{ "__type": "IngestedDoc", "__tag": 4010, "_content": {}, "_ordered_sections": [], "item_file": null, "item_line": null, "item_type": null, "aliases": [], "example_section_data": { "__type": "Section", "__tag": 4015, "children": [], "title": [], "level": 0, "target": null }, "see_also": [], "signature": null, "references": null, "qa": "release:2.0.0-notes", "arbitrary": [ { "__type": "Section", "__tag": 4015, "children": [], "title": [], "level": 0, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy 2.0.0 is the first major release since 2006. It is the result of 11 months of development since the last feature release and is the work of 212 contributors spread over 1078 pull requests. It contains a large number of exciting new features as well as changes to both the Python and C APIs." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "This major release includes breaking changes that could not happen in a regular minor (feature) release - including an ABI break, changes to type promotion rules, and API changes which may not have been emitting deprecation warnings in 1.26.x. Key documents related to how to adapt to changes in NumPy 2.0, in addition to these release notes, include:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy-2-migration-guide", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "NumPy 2.0-specific advice", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "dev:depending_on_numpy" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " in " }, { "__type": "CrossRef", "__tag": 4002, "value": "for-downstream-package-authors", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "dev:depending_on_numpy" }, "kind": "exists" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "NumPy 2.0.0 Release Notes" } ], "level": 0, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Highlights of this release include:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "New features:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new variable-length string dtype, " }, { "__type": "CrossRef", "__tag": 4002, "value": "StringDType", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.dtypes:StringDType" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and a new " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.strings", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.strings" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " namespace with performant ufuncs for string operations," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Support for " }, { "__type": "InlineCode", "__tag": 4051, "value": "float32" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "longdouble" }, { "__type": "Text", "__tag": 4046, "value": " in all " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.fft", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " functions," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Support for the array API standard in the main " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy" }, { "__type": "Text", "__tag": 4046, "value": " namespace." } ] } ] } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Performance improvements:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Sorting functions (" }, { "__type": "InlineRole", "__tag": 4003, "value": "sort", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineRole", "__tag": 4003, "value": "argsort", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineRole", "__tag": 4003, "value": "partition", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineRole", "__tag": 4003, "value": "argpartition", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ") have been accelerated through the use of the Intel x86-simd-sort and Google Highway libraries, and may see large (hardware-specific) speedups," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "macOS Accelerate support and binary wheels for macOS >=14, with significant performance improvements for linear algebra operations on macOS, and wheels that are about 3 times smaller," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.char", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.char" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " fixed-length string operations have been accelerated by implementing ufuncs that also support " }, { "__type": "CrossRef", "__tag": 4002, "value": "StringDType", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.dtypes:StringDType" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " in addition to the fixed-length string dtypes," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new tracing and introspection API, " }, { "__type": "CrossRef", "__tag": 4002, "value": "opt_func_info", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.lib.introspect:opt_func_info" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", to determine which hardware-specific kernels are available and will be dispatched to." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.save", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:save" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " now uses pickle protocol version 4 for saving arrays with object dtype, which allows for pickle objects larger than 4GB and improves saving speed by about 5% for large arrays." } ] } ] } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Python API improvements:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A clear split between public and private API, with a new " }, { "__type": "CrossRef", "__tag": 4002, "value": "module structure", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:module_structure" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": ", and each public function now available in a single place," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Many removals of non-recommended functions and aliases. This should make it easier to learn and use NumPy. The number of objects in the main namespace decreased by ~10% and in " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.lib" }, { "__type": "Text", "__tag": 4046, "value": " by ~80%," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "Canonical dtype names", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "user:basics.types" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " and a new " }, { "__type": "CrossRef", "__tag": 4002, "value": "isdtype", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:isdtype" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " introspection function," } ] } ] } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "C API improvements:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "CrossRef", "__tag": 4002, "value": "public C API for creating custom dtypes", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:c-api:array" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Many outdated functions and macros removed, and private internals hidden to ease future extensibility," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "New, easier to use, initialization functions: " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ImportNumPyAPI" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_ImportUFuncAPI" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Improved behavior:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Improvements to type promotion behavior was changed by adopting " }, { "__type": "InlineRole", "__tag": 4003, "value": "NEP 50 ", "domain": null, "role": "ref", "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ". This fixes many user surprises about promotions which previously often depended on data values of input arrays rather than only their dtypes. Please see the NEP and the " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy-2-migration-guide", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for details as this change can lead to changes in output dtypes and lower precision results for mixed-dtype operations." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The default integer type on Windows is now " }, { "__type": "InlineCode", "__tag": 4051, "value": "int64" }, { "__type": "Text", "__tag": 4046, "value": " rather than " }, { "__type": "InlineCode", "__tag": 4051, "value": "int32" }, { "__type": "Text", "__tag": 4046, "value": ", matching the behavior on other platforms," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The maximum number of array dimensions is changed from 32 to 64" } ] } ] } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Documentation:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The reference guide navigation was significantly improved, and there is now documentation on NumPy's " }, { "__type": "CrossRef", "__tag": 4002, "value": "module structure", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:module_structure" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "," } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "building from source", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "building:index" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " documentation was completely rewritten," } ] } ] } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Furthermore there are many changes to NumPy internals, including continuing to migrate code from C to C++, that will make it easier to improve and maintain NumPy in the future." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The \"no free lunch\" theorem dictates that there is a price to pay for all these API and behavior improvements and better future extensibility. This price is:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": true, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Backwards compatibility. There are a significant number of breaking changes to both the Python and C APIs. In the majority of cases, there are clear error messages that will inform the user how to adapt their code. However, there are also changes in behavior for which it was not possible to give such an error message - these cases are all covered in the Deprecation and Compatibility sections below, and in the " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy-2-migration-guide", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Note that there is a " }, { "__type": "InlineCode", "__tag": 4051, "value": "ruff" }, { "__type": "Text", "__tag": 4046, "value": " mode to auto-fix many things in Python code." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Breaking changes to the NumPy ABI. As a result, binaries of packages that use the NumPy C API and were built against a NumPy 1.xx release will not work with NumPy 2.0. On import, such packages will see an " }, { "__type": "InlineCode", "__tag": 4051, "value": "ImportError" }, { "__type": "Text", "__tag": 4046, "value": " with a message about binary incompatibility." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "It is possible to build binaries against NumPy 2.0 that will work at runtime with both NumPy 2.0 and 1.x. See " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy-2-abi-handling", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "dev:depending_on_numpy" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for more details." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Strong", "__tag": 4048, "children": [ { "__type": "Text", "__tag": 4046, "value": "All downstream packages that depend on the NumPy ABI are advised to do a\n new release built against NumPy 2.0 and verify that that release works with\n both 2.0 and 1.26 - ideally in the period between 2.0.0rc1 (which will be\n ABI-stable) and the final 2.0.0 release to avoid problems for their users." } ] } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The Python versions supported by this release are 3.9-3.12." } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Highlights" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.geterrobj" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.seterrobj" }, { "__type": "Text", "__tag": 4046, "value": " and the related ufunc keyword argument " }, { "__type": "InlineCode", "__tag": 4051, "value": "extobj=" }, { "__type": "Text", "__tag": 4046, "value": " have been removed. The preferred replacement for all of these is using the context manager " }, { "__type": "InlineCode", "__tag": 4051, "value": "with np.errstate():" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23922" } ], "url": "https://github.com/numpy/numpy/pull/23922", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.cast" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. The literal replacement for " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.cast[dtype](arg)" }, { "__type": "Text", "__tag": 4046, "value": " is " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.asarray(arg, dtype=dtype)" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.source" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. The preferred replacement is " }, { "__type": "InlineCode", "__tag": 4051, "value": "inspect.getsource" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.lookfor" }, { "__type": "Text", "__tag": 4046, "value": " has been removed." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24144" } ], "url": "https://github.com/numpy/numpy/pull/24144", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.who" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. As an alternative for the removed functionality, one can use a variable explorer that is available in IDEs such as Spyder or Jupyter Notebook." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24321" } ], "url": "https://github.com/numpy/numpy/pull/24321", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Warnings and exceptions present in " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.exceptions", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.exceptions" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " (e.g, " }, { "__type": "CrossRef", "__tag": 4002, "value": "ComplexWarning", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.exceptions:ComplexWarning" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "VisibleDeprecationWarning", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.exceptions:VisibleDeprecationWarning" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ") are no longer exposed in the main namespace." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Multiple niche enums, expired members and functions have been removed from the main namespace, such as: " }, { "__type": "InlineCode", "__tag": 4051, "value": "ERR_*" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "SHIFT_*" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.fastCopyAndTranspose" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.kernel_version" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.numarray" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.oldnumeric" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.set_numeric_ops" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24316" } ], "url": "https://github.com/numpy/numpy/pull/24316", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Replaced " }, { "__type": "InlineCode", "__tag": 4051, "value": "from ... import *" }, { "__type": "Text", "__tag": 4046, "value": " in the " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy/__init__.py" }, { "__type": "Text", "__tag": 4046, "value": " with explicit imports. As a result, these main namespace members got removed: " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.FLOATING_POINT_SUPPORT" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.FPE_*" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.NINF" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.PINF" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.NZERO" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.PZERO" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.CLIP" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.WRAP" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.WRAP" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.RAISE" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.BUFSIZE" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.UFUNC_BUFSIZE_DEFAULT" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.UFUNC_PYVALS_NAME" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.ALLOW_THREADS" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.MAXDIMS" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.MAY_SHARE_EXACT" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.MAY_SHARE_BOUNDS" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "add_newdoc" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.add_docstring" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.add_newdoc_ufunc" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24357" } ], "url": "https://github.com/numpy/numpy/pull/24357", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.float_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.float64" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.complex_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.complex128" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.longfloat" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.longdouble" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.singlecomplex" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.complex64" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.cfloat" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.complex128" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.longcomplex" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.clongdouble" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.clongfloat" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.clongdouble" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.string_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.bytes_" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.unicode_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.str_" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.Inf" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.inf" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.Infinity" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.inf" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.NaN" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.nan" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.infty" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.inf" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.mat" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.asmatrix" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.issubclass_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use the " }, { "__type": "InlineCode", "__tag": 4051, "value": "issubclass" }, { "__type": "Text", "__tag": 4046, "value": " builtin instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.asfarray" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.asarray" }, { "__type": "Text", "__tag": 4046, "value": " with a proper dtype instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.set_string_function" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.set_printoptions" }, { "__type": "Text", "__tag": 4046, "value": " instead with a formatter for custom printing of NumPy objects." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.tracemalloc_domain" }, { "__type": "Text", "__tag": 4046, "value": " is now only available from " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.lib" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.recfromcsv" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.recfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " were removed from the main namespace. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " with comma delimiter instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.issctype" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.maximum_sctype" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.obj2sctype" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.sctype2char" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.sctypes" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.issubsctype" }, { "__type": "Text", "__tag": 4046, "value": " were all removed from the main namespace without replacement, as they where niche members." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Deprecated " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.deprecate" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.deprecate_with_doc" }, { "__type": "Text", "__tag": 4046, "value": " has been removed from the main namespace. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "DeprecationWarning" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Deprecated " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.safe_eval" }, { "__type": "Text", "__tag": 4046, "value": " has been removed from the main namespace. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "ast.literal_eval" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24376" } ], "url": "https://github.com/numpy/numpy/pull/24376", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.find_common_type" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.promote_types" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.result_type" }, { "__type": "Text", "__tag": 4046, "value": " instead. To achieve semantics for the " }, { "__type": "InlineCode", "__tag": 4051, "value": "scalar_types" }, { "__type": "Text", "__tag": 4046, "value": " argument, use " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.result_type" }, { "__type": "Text", "__tag": 4046, "value": " and pass " }, { "__type": "InlineCode", "__tag": 4051, "value": "0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "0.0" }, { "__type": "Text", "__tag": 4046, "value": ", or " }, { "__type": "InlineCode", "__tag": 4051, "value": "0j" }, { "__type": "Text", "__tag": 4046, "value": " as a Python scalar instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.round_" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.round" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.nbytes" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype().itemsize" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24477" } ], "url": "https://github.com/numpy/numpy/pull/24477", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.compare_chararrays" }, { "__type": "Text", "__tag": 4046, "value": " has been removed from the main namespace. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char.compare_chararrays" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "charrarray" }, { "__type": "Text", "__tag": 4046, "value": " in the main namespace has been deprecated. It can be imported without a deprecation warning from " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char.chararray" }, { "__type": "Text", "__tag": 4046, "value": " for now, but we are planning to fully deprecate and remove " }, { "__type": "InlineCode", "__tag": 4051, "value": "chararray" }, { "__type": "Text", "__tag": 4046, "value": " in the future." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.format_parser" }, { "__type": "Text", "__tag": 4046, "value": " has been removed from the main namespace. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.rec.format_parser" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24587" } ], "url": "https://github.com/numpy/numpy/pull/24587", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Support for seven data type string aliases has been removed from " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype" }, { "__type": "Text", "__tag": 4046, "value": ": " }, { "__type": "InlineCode", "__tag": 4051, "value": "int0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "uint0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "void0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "object0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "str0" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "bytes0" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "bool8" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24807" } ], "url": "https://github.com/numpy/numpy/pull/24807", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The experimental " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.array_api" }, { "__type": "Text", "__tag": 4046, "value": " submodule has been removed. Use the main " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy" }, { "__type": "Text", "__tag": 4046, "value": " namespace for regular usage instead, or the separate " }, { "__type": "InlineCode", "__tag": 4051, "value": "array-api-strict" }, { "__type": "Text", "__tag": 4046, "value": " package for the compliance testing use case for which " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.array_api" }, { "__type": "Text", "__tag": 4046, "value": " was mostly used." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25911" } ], "url": "https://github.com/numpy/numpy/pull/25911", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "NumPy 2.0 Python API removals" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "UFuncs called " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_prepare__" }, { "__type": "Text", "__tag": 4046, "value": " before running computations for normal ufunc calls (not generalized ufuncs, reductions, etc.). The function was also called instead of " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_wrap__" }, { "__type": "Text", "__tag": 4046, "value": " on the results of some linear algebra functions." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "It is now removed. If you use it, migrate to " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_ufunc__" }, { "__type": "Text", "__tag": 4046, "value": " or rely on " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_wrap__" }, { "__type": "Text", "__tag": 4046, "value": " which is called with a context in all cases, although only after the result array is filled. In those code paths, " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_wrap__" }, { "__type": "Text", "__tag": 4046, "value": " will now be passed a base class, rather than a subclass array." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25105" } ], "url": "https://github.com/numpy/numpy/pull/25105", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "__array_prepare__" }, { "__type": "Text", "__tag": 4046, "value": " is removed" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.compat" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated, as Python 2 is no longer supported." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.int8" }, { "__type": "Text", "__tag": 4046, "value": " and similar classes will no longer support conversion of out of bounds python integers to integer arrays. For example, conversion of 255 to int8 will not return -1. " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.iinfo(dtype)" }, { "__type": "Text", "__tag": 4046, "value": " can be used to check the machine limits for data types. For example, " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.iinfo(np.uint16)" }, { "__type": "Text", "__tag": 4046, "value": " returns min = 0 and max = 65535." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.array(value).astype(dtype)" }, { "__type": "Text", "__tag": 4046, "value": " will give the desired result." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.safe_eval" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated. " }, { "__type": "InlineCode", "__tag": 4051, "value": "ast.literal_eval" }, { "__type": "Text", "__tag": 4046, "value": " should be used instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23830" } ], "url": "https://github.com/numpy/numpy/pull/23830", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.recfromcsv" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.recfromtxt" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.disp" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.get_array_wrap" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.maximum_sctype" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.deprecate" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.deprecate_with_doc" }, { "__type": "Text", "__tag": 4046, "value": " have been deprecated." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24154" } ], "url": "https://github.com/numpy/numpy/pull/24154", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.trapz" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.trapezoid" }, { "__type": "Text", "__tag": 4046, "value": " or a " }, { "__type": "InlineCode", "__tag": 4051, "value": "scipy.integrate" }, { "__type": "Text", "__tag": 4046, "value": " function instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.in1d" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.isin" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Alias " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.row_stack" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.vstack" }, { "__type": "Text", "__tag": 4046, "value": " directly." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24445" } ], "url": "https://github.com/numpy/numpy/pull/24445", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "__array_wrap__" }, { "__type": "Text", "__tag": 4046, "value": " is now passed " }, { "__type": "InlineCode", "__tag": 4051, "value": "arr, context, return_scalar" }, { "__type": "Text", "__tag": 4046, "value": " and support for implementations not accepting all three are deprecated. Its signature should be " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array_wrap__(self, arr, context=None, return_scalar=False)" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25409" } ], "url": "https://github.com/numpy/numpy/pull/25409", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Arrays of 2-dimensional vectors for " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.cross" }, { "__type": "Text", "__tag": 4046, "value": " have been deprecated. Use arrays of 3-dimensional vectors instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24818" } ], "url": "https://github.com/numpy/numpy/pull/24818", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"a\")" }, { "__type": "Text", "__tag": 4046, "value": " alias for " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(np.bytes_)" }, { "__type": "Text", "__tag": 4046, "value": " was deprecated. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"S\")" }, { "__type": "Text", "__tag": 4046, "value": " alias instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24854" } ], "url": "https://github.com/numpy/numpy/pull/24854", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Use of keyword arguments " }, { "__type": "InlineCode", "__tag": 4051, "value": "x" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "y" }, { "__type": "Text", "__tag": 4046, "value": " with functions " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert_array_equal" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert_array_almost_equal" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated. Pass the first two arguments as positional arguments instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24978" } ], "url": "https://github.com/numpy/numpy/pull/24978", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Deprecations" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Using " }, { "__type": "InlineCode", "__tag": 4051, "value": "fftn" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "ifftn" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "rfftn" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "irfftn" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "fft2" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "ifft2" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "rfft2" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "irfft2" }, { "__type": "Text", "__tag": 4046, "value": " with the " }, { "__type": "InlineCode", "__tag": 4051, "value": "s" }, { "__type": "Text", "__tag": 4046, "value": " parameter set to a value that is not " }, { "__type": "InlineCode", "__tag": 4051, "value": "None" }, { "__type": "Text", "__tag": 4046, "value": " and the " }, { "__type": "InlineCode", "__tag": 4051, "value": "axes" }, { "__type": "Text", "__tag": 4046, "value": " parameter set to " }, { "__type": "InlineCode", "__tag": 4051, "value": "None" }, { "__type": "Text", "__tag": 4046, "value": " has been deprecated, in line with the array API standard. To retain current behaviour, pass a sequence [0, ..., k-1] to " }, { "__type": "InlineCode", "__tag": 4051, "value": "axes" }, { "__type": "Text", "__tag": 4046, "value": " for an array of dimension k." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Furthermore, passing an array to " }, { "__type": "InlineCode", "__tag": 4051, "value": "s" }, { "__type": "Text", "__tag": 4046, "value": " which contains " }, { "__type": "InlineCode", "__tag": 4051, "value": "None" }, { "__type": "Text", "__tag": 4046, "value": " values is deprecated as the parameter is documented to accept a sequence of integers in both the NumPy docs and the array API specification. To use the default behaviour of the corresponding 1-D transform, pass the value matching the default for its " }, { "__type": "InlineCode", "__tag": 4051, "value": "n" }, { "__type": "Text", "__tag": 4046, "value": " parameter. To use the default behaviour for every axis, the " }, { "__type": "InlineCode", "__tag": 4051, "value": "s" }, { "__type": "Text", "__tag": 4046, "value": " argument can be omitted." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25495" } ], "url": "https://github.com/numpy/numpy/pull/25495", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.fft" }, { "__type": "Text", "__tag": 4046, "value": " deprecations for n-D transforms with None values in arguments" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "lstsq", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:lstsq" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " now uses the new rcond value of the machine precision times " }, { "__type": "InlineCode", "__tag": 4051, "value": "max(M, N)" }, { "__type": "Text", "__tag": 4046, "value": ". Previously, the machine precision was used but a FutureWarning was given to notify that this change will happen eventually. That old behavior can still be achieved by passing " }, { "__type": "InlineCode", "__tag": 4051, "value": "rcond=-1" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25721" } ], "url": "https://github.com/numpy/numpy/pull/25721", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.linalg.lstsq" }, { "__type": "Text", "__tag": 4046, "value": " now defaults to a new " }, { "__type": "InlineCode", "__tag": 4051, "value": "rcond" }, { "__type": "Text", "__tag": 4046, "value": " value" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.core.umath_tests" }, { "__type": "Text", "__tag": 4046, "value": " submodule has been removed from the public API. (Deprecated in NumPy 1.15)" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23809" } ], "url": "https://github.com/numpy/numpy/pull/23809", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataMem_SetEventHook" }, { "__type": "Text", "__tag": 4046, "value": " deprecation has expired and it is removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "tracemalloc" }, { "__type": "Text", "__tag": 4046, "value": " and the " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.lib.tracemalloc_domain" }, { "__type": "Text", "__tag": 4046, "value": " domain. (Deprecated in NumPy 1.23)" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23921" } ], "url": "https://github.com/numpy/numpy/pull/23921", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The deprecation of " }, { "__type": "InlineCode", "__tag": 4051, "value": "set_numeric_ops" }, { "__type": "Text", "__tag": 4046, "value": " and the C functions " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_SetNumericOps" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_GetNumericOps" }, { "__type": "Text", "__tag": 4046, "value": " has been expired and the functions removed. (Deprecated in NumPy 1.16)" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23998" } ], "url": "https://github.com/numpy/numpy/pull/23998", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "fasttake" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "fastclip" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "fastputmask" }, { "__type": "Text", "__tag": 4046, "value": " " }, { "__type": "InlineCode", "__tag": 4051, "value": "ArrFuncs" }, { "__type": "Text", "__tag": 4046, "value": " deprecation is now finalized." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The deprecated function " }, { "__type": "InlineCode", "__tag": 4051, "value": "fastCopyAndTranspose" }, { "__type": "Text", "__tag": 4046, "value": " and its C counterpart are now removed." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The deprecation of " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ScalarFromObject" }, { "__type": "Text", "__tag": 4046, "value": " is now finalized." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24312" } ], "url": "https://github.com/numpy/numpy/pull/24312", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.msort" }, { "__type": "Text", "__tag": 4046, "value": " has been removed. For a replacement, " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.sort(a, axis=0)" }, { "__type": "Text", "__tag": 4046, "value": " should be used instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24494" } ], "url": "https://github.com/numpy/numpy/pull/24494", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype((\"f8\", 1)" }, { "__type": "Text", "__tag": 4046, "value": " will now return a shape 1 subarray dtype rather than a non-subarray one." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25761" } ], "url": "https://github.com/numpy/numpy/pull/25761", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Assigning to the " }, { "__type": "InlineCode", "__tag": 4051, "value": ".data" }, { "__type": "Text", "__tag": 4046, "value": " attribute of an ndarray is disallowed and will raise." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.binary_repr(a, width)" }, { "__type": "Text", "__tag": 4046, "value": " will raise if width is too small." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Using " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_CHAR" }, { "__type": "Text", "__tag": 4046, "value": " in " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_DescrFromType()" }, { "__type": "Text", "__tag": 4046, "value": " will raise, use " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_STRING" }, { "__type": "Text", "__tag": 4046, "value": " " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_UNICODE" }, { "__type": "Text", "__tag": 4046, "value": ", or " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_VSTRING" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25794" } ], "url": "https://github.com/numpy/numpy/pull/25794", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Expired deprecations" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [], "title": [ { "__type": "Text", "__tag": 4046, "value": "Compatibility notes" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "loadtxt" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " now both default to " }, { "__type": "InlineCode", "__tag": 4051, "value": "encoding=None" }, { "__type": "Text", "__tag": 4046, "value": " which may mainly modify how " }, { "__type": "InlineCode", "__tag": 4051, "value": "converters" }, { "__type": "Text", "__tag": 4046, "value": " work. These will now be passed " }, { "__type": "InlineCode", "__tag": 4051, "value": "str" }, { "__type": "Text", "__tag": 4046, "value": " rather than " }, { "__type": "InlineCode", "__tag": 4051, "value": "bytes" }, { "__type": "Text", "__tag": 4046, "value": ". Pass the encoding explicitly to always get the new or old behavior. For " }, { "__type": "InlineCode", "__tag": 4051, "value": "genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " the change also means that returned values will now be unicode strings rather than bytes." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25158" } ], "url": "https://github.com/numpy/numpy/pull/25158", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "loadtxt" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " default encoding changed" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " will no longer accept ambiguous " }, { "__type": "InlineCode", "__tag": 4051, "value": "-m" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": ".pyf" }, { "__type": "Text", "__tag": 4046, "value": " CLI combinations. When more than one " }, { "__type": "InlineCode", "__tag": 4051, "value": ".pyf" }, { "__type": "Text", "__tag": 4046, "value": " file is passed, an error is raised. When both " }, { "__type": "InlineCode", "__tag": 4051, "value": "-m" }, { "__type": "Text", "__tag": 4046, "value": " and a " }, { "__type": "InlineCode", "__tag": 4051, "value": ".pyf" }, { "__type": "Text", "__tag": 4046, "value": " is passed, a warning is emitted and the " }, { "__type": "InlineCode", "__tag": 4051, "value": "-m" }, { "__type": "Text", "__tag": 4046, "value": " provided name is ignored." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25181" } ], "url": "https://github.com/numpy/numpy/pull/25181", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py.compile()" }, { "__type": "Text", "__tag": 4046, "value": " helper has been removed because it leaked memory, has been marked as experimental for several years now, and was implemented as a thin " }, { "__type": "InlineCode", "__tag": 4051, "value": "subprocess.run" }, { "__type": "Text", "__tag": 4046, "value": " wrapper. It was also one of the test bottlenecks. See " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25122" } ], "url": "https://github.com/numpy/numpy/issues/25122", "title": "" }, { "__type": "Text", "__tag": 4046, "value": " for the full rationale. It also used several " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.distutils" }, { "__type": "Text", "__tag": 4046, "value": " features which are too fragile to be ported to work with " }, { "__type": "InlineCode", "__tag": 4051, "value": "meson" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Users are urged to replace calls to " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py.compile" }, { "__type": "Text", "__tag": 4046, "value": " with calls to " }, { "__type": "InlineCode", "__tag": 4051, "value": "subprocess.run(\"python\", \"-m\", \"numpy.f2py\",..." }, { "__type": "Text", "__tag": 4046, "value": " instead, and to use environment variables to interact with " }, { "__type": "InlineCode", "__tag": 4051, "value": "meson" }, { "__type": "Text", "__tag": 4046, "value": ". " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "Native files" } ], "url": "https://mesonbuild.com/Machine-files.html", "title": "" }, { "__type": "Text", "__tag": 4046, "value": " are also an option." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25193" } ], "url": "https://github.com/numpy/numpy/pull/25193", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " compatibility notes" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Due to algorithmic changes and use of SIMD code, sorting functions with methods that aren't stable may return slightly different results in 2.0.0 compared to 1.26.x. This includes the default method of " }, { "__type": "CrossRef", "__tag": 4002, "value": "argsort", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:argsort" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "argpartition", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:argpartition" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Minor changes in behavior of sorting functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The broadcasting rules for " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.solve(a, b)" }, { "__type": "Text", "__tag": 4046, "value": " were ambiguous when " }, { "__type": "InlineCode", "__tag": 4051, "value": "b" }, { "__type": "Text", "__tag": 4046, "value": " had 1 fewer dimensions than " }, { "__type": "InlineCode", "__tag": 4051, "value": "a" }, { "__type": "Text", "__tag": 4046, "value": ". This has been resolved in a backward-incompatible way and is now compliant with the Array API. The old behaviour can be reconstructed by using " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.solve(a, b[..., None])[..., 0]" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25914" } ], "url": "https://github.com/numpy/numpy/pull/25914", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Removed ambiguity when broadcasting in " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.solve" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The representation method for " }, { "__type": "CrossRef", "__tag": 4002, "value": "Polynomial", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.polynomial.polynomial:Polynomial" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " was updated to include the domain in the representation. The plain text and latex representations are now consistent. For example the output of " }, { "__type": "InlineCode", "__tag": 4051, "value": "str(np.polynomial.Polynomial([1, 1], domain=[.1, .2]))" }, { "__type": "Text", "__tag": 4046, "value": " used to be " }, { "__type": "InlineCode", "__tag": 4051, "value": "1.0 + 1.0 x" }, { "__type": "Text", "__tag": 4046, "value": ", but now is " }, { "__type": "InlineCode", "__tag": 4051, "value": "1.0 + 1.0 (-3.0000000000000004 + 20.0 x)" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-21760" } ], "url": "https://github.com/numpy/numpy/pull/21760", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Modified representation for " }, { "__type": "InlineCode", "__tag": 4051, "value": "Polynomial" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CGT" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CLT" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CGE" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CLE" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CEQ" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CNE" }, { "__type": "Text", "__tag": 4046, "value": " macros have been removed." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_MIN" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_MAX" }, { "__type": "Text", "__tag": 4046, "value": " have been moved from " }, { "__type": "InlineCode", "__tag": 4051, "value": "ndarraytypes.h" }, { "__type": "Text", "__tag": 4046, "value": " to " }, { "__type": "InlineCode", "__tag": 4051, "value": "npy_math.h" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24258" } ], "url": "https://github.com/numpy/numpy/pull/24258", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A C API for working with " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.dtypes.StringDType", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.dtypes:StringDType" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " arrays has been exposed. This includes functions for acquiring and releasing mutexes which lock access to the string data, as well as packing and unpacking UTF-8 bytestreams from array entries." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "NPY_NTYPES" }, { "__type": "Text", "__tag": 4046, "value": " has been renamed to " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_NTYPES_LEGACY" }, { "__type": "Text", "__tag": 4046, "value": " as it does not include new NumPy built-in DTypes. In particular the new string DType will likely not work correctly with code that handles legacy DTypes." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25347" } ], "url": "https://github.com/numpy/numpy/pull/25347", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The C-API now only exports the static inline function versions of the array accessors (previously this depended on using \"deprecated API\"). While we discourage it, the struct fields can still be used directly." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25789" } ], "url": "https://github.com/numpy/numpy/pull/25789", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy now defines " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_Pack" }, { "__type": "Text", "__tag": 4046, "value": " to set an individual memory address. Unlike " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_SETITEM" }, { "__type": "Text", "__tag": 4046, "value": " this function is equivalent to setting an individual array item and does not require a NumPy array input." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25954" } ], "url": "https://github.com/numpy/numpy/pull/25954", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "->f" }, { "__type": "Text", "__tag": 4046, "value": " slot has been removed from " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_Descr" }, { "__type": "Text", "__tag": 4046, "value": ". If you use this slot, replace accessing it with " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_GetArrFuncs" }, { "__type": "Text", "__tag": 4046, "value": " (see its documentation and the " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy-2-migration-guide", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "). In some cases using other functions like " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_GETITEM" }, { "__type": "Text", "__tag": 4046, "value": " may be an alternatives." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_GETITEM" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_SETITEM" }, { "__type": "Text", "__tag": 4046, "value": " now require the import of the NumPy API table to be used and are no longer defined in " }, { "__type": "InlineCode", "__tag": 4051, "value": "ndarraytypes.h" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25812" } ], "url": "https://github.com/numpy/numpy/pull/25812", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Due to runtime dependencies, the definition for functionality accessing the dtype flags was moved from " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy/ndarraytypes.h" }, { "__type": "Text", "__tag": 4046, "value": " and is only available after including " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy/ndarrayobject.h" }, { "__type": "Text", "__tag": 4046, "value": " as it requires " }, { "__type": "InlineCode", "__tag": 4051, "value": "import_array()" }, { "__type": "Text", "__tag": 4046, "value": ". This includes " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_FLAGCHK" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_REFCHK" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_BEGIN_THREADS_DESCR" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The dtype flags on " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_Descr" }, { "__type": "Text", "__tag": 4046, "value": " must now be accessed through the " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_FLAGS" }, { "__type": "Text", "__tag": 4046, "value": " inline function to be compatible with both 1.x and 2.x. This function is defined in " }, { "__type": "InlineCode", "__tag": 4051, "value": "npy_2_compat.h" }, { "__type": "Text", "__tag": 4046, "value": " to allow backporting. Most or all users should use " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_FLAGCHK" }, { "__type": "Text", "__tag": 4046, "value": " which is available on 1.x and does not require backporting. Cython users should use Cython 3. Otherwise access will go through Python unless they use " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_FLAGCHK" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25816" } ], "url": "https://github.com/numpy/numpy/pull/25816", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "C API changes" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The functions " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_ConvertDatetime64ToDatetimeStruct" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_ConvertDatetimeStructToDatetime64" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_ConvertPyDateTimeToDatetimeStruct" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_GetDatetimeISO8601StrLen" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_MakeISO8601Datetime" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "NpyDatetime_ParseISO8601Datetime" }, { "__type": "Text", "__tag": 4046, "value": " have been added to the C API to facilitate converting between strings, Python datetimes, and NumPy datetimes in external libraries." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-21199" } ], "url": "https://github.com/numpy/numpy/pull/21199", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Datetime functionality exposed in the C API and Cython bindings" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The NumPy C API's functions for constructing generalized ufuncs (" }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_FromFuncAndData" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_FromFuncAndDataAndSignature" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_FromFuncAndDataAndSignatureAndIdentity" }, { "__type": "Text", "__tag": 4046, "value": ") take " }, { "__type": "InlineCode", "__tag": 4051, "value": "types" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "data" }, { "__type": "Text", "__tag": 4046, "value": " arguments that are not modified by NumPy's internals. Like the " }, { "__type": "InlineCode", "__tag": 4051, "value": "name" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "doc" }, { "__type": "Text", "__tag": 4046, "value": " arguments, third-party Python extension modules are likely to supply these arguments from static constants. The " }, { "__type": "InlineCode", "__tag": 4051, "value": "types" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "data" }, { "__type": "Text", "__tag": 4046, "value": " arguments are now const-correct: they are declared as " }, { "__type": "InlineCode", "__tag": 4051, "value": "const char *types" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "void *const *data" }, { "__type": "Text", "__tag": 4046, "value": ", respectively. C code should not be affected, but C++ code may be." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23847" } ], "url": "https://github.com/numpy/numpy/pull/23847", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Const correctness for the generalized ufunc C API" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXDIMS" }, { "__type": "Text", "__tag": 4046, "value": " is now 64, you may want to review its use. This is usually used in a stack allocation, where the increase should be safe. However, we do encourage generally to remove any use of " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXDIMS" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXARGS" }, { "__type": "Text", "__tag": 4046, "value": " to eventually allow removing the constraint completely. For the conversion helper and C-API functions mirroring Python ones such as " }, { "__type": "InlineCode", "__tag": 4051, "value": "take" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXDIMS" }, { "__type": "Text", "__tag": 4046, "value": " was used to mean " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" }, { "__type": "Text", "__tag": 4046, "value": ". Such usage must be replaced with " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_RAVEL_AXIS" }, { "__type": "Text", "__tag": 4046, "value": ". See also " }, { "__type": "CrossRef", "__tag": 4002, "value": "migration_maxdims", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25149" } ], "url": "https://github.com/numpy/numpy/pull/25149", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Larger " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXDIMS" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXARGS" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_RAVEL_AXIS" }, { "__type": "Text", "__tag": 4046, "value": " introduced" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Since " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXARGS" }, { "__type": "Text", "__tag": 4046, "value": " was increased, it is now a runtime constant and not compile-time constant anymore. We expect almost no users to notice this. But if used for stack allocations it now must be replaced with a custom constant using " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXARGS" }, { "__type": "Text", "__tag": 4046, "value": " as an additional runtime check." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "sizeof(PyArrayMultiIterObject)" }, { "__type": "Text", "__tag": 4046, "value": " no longer includes the full size of the object. We expect nobody to notice this change. It was necessary to avoid issues with Cython." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25271" } ], "url": "https://github.com/numpy/numpy/pull/25271", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAXARGS" }, { "__type": "Text", "__tag": 4046, "value": " not constant and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArrayMultiIterObject" }, { "__type": "Text", "__tag": 4046, "value": " size change" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "In order to improve our DTypes it is unfortunately necessary to break the ABI, which requires some changes for dtypes registered with " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_RegisterDataType" }, { "__type": "Text", "__tag": 4046, "value": ". Please see the documentation of " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_RegisterDataType" }, { "__type": "Text", "__tag": 4046, "value": " for how to adapt your code and achieve compatibility with both 1.x and 2.x." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25792" } ], "url": "https://github.com/numpy/numpy/pull/25792", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Required changes for custom legacy user dtypes" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The C implementation of the NEP 42 DType API is now public. While the DType API has shipped in NumPy for a few versions, it was only usable in sessions with a special environment variable set. It is now possible to write custom DTypes outside of NumPy using the new DType API and the normal " }, { "__type": "InlineCode", "__tag": 4051, "value": "import_array()" }, { "__type": "Text", "__tag": 4046, "value": " mechanism for importing the numpy C API." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "See " }, { "__type": "CrossRef", "__tag": 4002, "value": "dtype-api", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:c-api:array" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for more details about the API. As always with a new feature, please report any bugs you run into implementing or using a new DType. It is likely that downstream C code that works with dtypes will need to be updated to work correctly with new DTypes." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25754" } ], "url": "https://github.com/numpy/numpy/pull/25754", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New Public DType API" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "We have now added " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ImportNumPyAPI" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_ImportUFuncAPI" }, { "__type": "Text", "__tag": 4046, "value": " as static inline functions to import the NumPy C-API tables. The new functions have two advantages over " }, { "__type": "InlineCode", "__tag": 4051, "value": "import_array" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "import_ufunc" }, { "__type": "Text", "__tag": 4046, "value": ":" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "They check whether the import was already performed and are light-weight if not, allowing to add them judiciously (although this is not preferable in most cases)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The old mechanisms were macros rather than functions which included a " }, { "__type": "InlineCode", "__tag": 4051, "value": "return" }, { "__type": "Text", "__tag": 4046, "value": " statement." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ImportNumPyAPI()" }, { "__type": "Text", "__tag": 4046, "value": " function is included in " }, { "__type": "InlineCode", "__tag": 4051, "value": "npy_2_compat.h" }, { "__type": "Text", "__tag": 4046, "value": " for simpler backporting." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25866" } ], "url": "https://github.com/numpy/numpy/pull/25866", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New C-API import functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The dtype structures fields " }, { "__type": "InlineCode", "__tag": 4051, "value": "c_metadata" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "names" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "fields" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "subarray" }, { "__type": "Text", "__tag": 4046, "value": " must now be accessed through new functions following the same names, such as " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_NAMES" }, { "__type": "Text", "__tag": 4046, "value": ". Direct access of the fields is not valid as they do not exist for all " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_Descr" }, { "__type": "Text", "__tag": 4046, "value": " instances. The " }, { "__type": "InlineCode", "__tag": 4051, "value": "metadata" }, { "__type": "Text", "__tag": 4046, "value": " field is kept, but the macro version should also be preferred." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25802" } ], "url": "https://github.com/numpy/numpy/pull/25802", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Structured dtype information access through functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Unless compiling only with NumPy 2 support, the " }, { "__type": "InlineCode", "__tag": 4051, "value": "elsize" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "alignment" }, { "__type": "Text", "__tag": 4046, "value": " fields must now be accessed via " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_ELSIZE" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_SET_ELSIZE" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_ALIGNMENT" }, { "__type": "Text", "__tag": 4046, "value": ". In cases where the descriptor is attached to an array, we advise using " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ITEMSIZE" }, { "__type": "Text", "__tag": 4046, "value": " as it exists on all NumPy versions. Please see " }, { "__type": "CrossRef", "__tag": 4002, "value": "migration_c_descr", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for more information." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25943" } ], "url": "https://github.com/numpy/numpy/pull/25943", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Descriptor " }, { "__type": "InlineCode", "__tag": 4051, "value": "elsize" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "alignment" }, { "__type": "Text", "__tag": 4046, "value": " access" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "npy_interrupt.h" }, { "__type": "Text", "__tag": 4046, "value": " and the corresponding macros like " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_SIGINT_ON" }, { "__type": "Text", "__tag": 4046, "value": " have been removed. We recommend querying " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyErr_CheckSignals()" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyOS_InterruptOccurred()" }, { "__type": "Text", "__tag": 4046, "value": " periodically (these do currently require holding the GIL though)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "noprefix.h" }, { "__type": "Text", "__tag": 4046, "value": " header has been removed. Replace missing symbols with their prefixed counterparts (usually an added " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "npy_" }, { "__type": "Text", "__tag": 4046, "value": ")." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23919" } ], "url": "https://github.com/numpy/numpy/pull/23919", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_GetPyVals" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_handlefperr" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUFunc_checkfperr" }, { "__type": "Text", "__tag": 4046, "value": " have been removed. If needed, a new backwards compatible function to raise floating point errors could be restored. Reason for removal: there are no known users and the functions would have made " }, { "__type": "InlineCode", "__tag": 4051, "value": "with np.errstate()" }, { "__type": "Text", "__tag": 4046, "value": " fixes much more difficult)." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23922" } ], "url": "https://github.com/numpy/numpy/pull/23922", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy/old_defines.h" }, { "__type": "Text", "__tag": 4046, "value": " which was part of the API deprecated since NumPy 1.7 has been removed. This removes macros of the form " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CONSTANT" }, { "__type": "Text", "__tag": 4046, "value": ". The " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "replace_old_macros.sed" } ], "url": "https://github.com/numpy/numpy/blob/main/tools/replace_old_macros.sed", "title": "" }, { "__type": "Text", "__tag": 4046, "value": " script may be useful to convert them to the " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_CONSTANT" }, { "__type": "Text", "__tag": 4046, "value": " version." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24011" } ], "url": "https://github.com/numpy/numpy/pull/24011", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "legacy_inner_loop_selector" }, { "__type": "Text", "__tag": 4046, "value": " member of the ufunc struct is removed to simplify improvements to the dispatching system. There are no known users overriding or directly accessing this member." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24271" } ], "url": "https://github.com/numpy/numpy/pull/24271", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "NPY_INTPLTR" }, { "__type": "Text", "__tag": 4046, "value": " has been removed to avoid confusion (see " }, { "__type": "InlineCode", "__tag": 4051, "value": "intp" }, { "__type": "Text", "__tag": 4046, "value": " redefinition)." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24888" } ], "url": "https://github.com/numpy/numpy/pull/24888", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The advanced indexing " }, { "__type": "InlineCode", "__tag": 4051, "value": "MapIter" }, { "__type": "Text", "__tag": 4046, "value": " and related API has been removed. The (truly) public part of it was not well tested and had only one known user (Theano). Making it private will simplify improvements to speed up " }, { "__type": "InlineCode", "__tag": 4051, "value": "ufunc.at" }, { "__type": "Text", "__tag": 4046, "value": ", make advanced indexing more maintainable, and was important for increasing the maximum number of dimensions of arrays to 64. Please let us know if this API is important to you so we can find a solution together." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25138" } ], "url": "https://github.com/numpy/numpy/pull/25138", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_MAX_ELSIZE" }, { "__type": "Text", "__tag": 4046, "value": " macro has been removed, as it only ever reflected builtin numeric types and served no internal purpose." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25149" } ], "url": "https://github.com/numpy/numpy/pull/25149", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_REFCNT" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_REFCOUNT" }, { "__type": "Text", "__tag": 4046, "value": " are removed. Use " }, { "__type": "InlineCode", "__tag": 4051, "value": "Py_REFCNT" }, { "__type": "Text", "__tag": 4046, "value": " instead." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25156" } ], "url": "https://github.com/numpy/numpy/pull/25156", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArrayFlags_Type" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_NewFlagsObject" }, { "__type": "Text", "__tag": 4046, "value": " as well as " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArrayFlagsObject" }, { "__type": "Text", "__tag": 4046, "value": " are private now. There is no known use-case; use the Python API if needed." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_MoveInto" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CastTo" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CastAnyTo" }, { "__type": "Text", "__tag": 4046, "value": " are removed use " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CopyInto" }, { "__type": "Text", "__tag": 4046, "value": " and if absolutely needed " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CopyAnyInto" }, { "__type": "Text", "__tag": 4046, "value": " (the latter does a flat copy)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_FillObjectArray" }, { "__type": "Text", "__tag": 4046, "value": " is removed, its only true use was for implementing " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.empty" }, { "__type": "Text", "__tag": 4046, "value": ". Create a new empty array or use " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_FillWithScalar()" }, { "__type": "Text", "__tag": 4046, "value": " (decrefs existing objects)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CompareUCS4" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CompareString" }, { "__type": "Text", "__tag": 4046, "value": " are removed. Use the standard C string comparison functions." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_ISPYTHON" }, { "__type": "Text", "__tag": 4046, "value": " is removed as it is misleading, has no known use-cases, and is easy to replace." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_FieldNames" }, { "__type": "Text", "__tag": 4046, "value": " is removed, as it is unclear what it would be useful for. It also has incorrect semantics in some possible use-cases." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_TypestrConvert" }, { "__type": "Text", "__tag": 4046, "value": " is removed, since it seems a misnomer and unlikely to be used by anyone. If you know the size or are limited to few types, just use it explicitly, otherwise go via Python strings." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25292" } ], "url": "https://github.com/numpy/numpy/pull/25292", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyDataType_GetDatetimeMetaData" }, { "__type": "Text", "__tag": 4046, "value": " is removed, it did not actually do anything since at least NumPy 1.7." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25802" } ], "url": "https://github.com/numpy/numpy/pull/25802", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_GetCastFunc" }, { "__type": "Text", "__tag": 4046, "value": " is removed. Note that custom legacy user dtypes can still provide a castfunc as their implementation, but any access to them is now removed. The reason for this is that NumPy never used these internally for many years. If you use simple numeric types, please just use C casts directly. In case you require an alternative, please let us know so we can create new API such as " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_CastBuffer()" }, { "__type": "Text", "__tag": 4046, "value": " which could use old or new cast functions depending on the NumPy version." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25161" } ], "url": "https://github.com/numpy/numpy/pull/25161", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "NumPy 2.0 C API removals" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [], "title": [ { "__type": "Text", "__tag": 4046, "value": "New Features" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Blockquote", "__tag": 4059, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24858" } ], "url": "https://github.com/numpy/numpy/pull/24858", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.add" }, { "__type": "Text", "__tag": 4046, "value": " was extended to work with " }, { "__type": "InlineCode", "__tag": 4051, "value": "unicode" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "bytes" }, { "__type": "Text", "__tag": 4046, "value": " dtypes." } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "This new function counts the number of 1-bits in a number. " }, { "__type": "CrossRef", "__tag": 4002, "value": "bitwise_count", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:bitwise_count" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " works on all the numpy integer types and integer-like objects." } ] }, { "__type": "Code", "__tag": 4050, "value": ">>> a = np.array([2**i - 1 for i in range(16)])\n>>> np.bitwise_count(a)\narray([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15],\n dtype=uint8)", "execution_status": null }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-19355" } ], "url": "https://github.com/numpy/numpy/pull/19355", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "InlineCode", "__tag": 4051, "value": "bitwise_count" }, { "__type": "Text", "__tag": 4046, "value": " function" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Support for the updated Accelerate BLAS/LAPACK library, including ILP64 (64-bit integer) support, in macOS 13.3 has been added. This brings arm64 support, and significant performance improvements of up to 10x for commonly used linear algebra operations. When Accelerate is selected at build time, or if no explicit BLAS library selection is done, the 13.3+ version will automatically be used if available." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24053" } ], "url": "https://github.com/numpy/numpy/pull/24053", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Binary wheels are also available. On macOS >=14.0, users who install NumPy from PyPI will get wheels built against Accelerate rather than OpenBLAS." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25255" } ], "url": "https://github.com/numpy/numpy/pull/25255", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "macOS Accelerate support, including the ILP64" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A " }, { "__type": "InlineCode", "__tag": 4051, "value": "weights" }, { "__type": "Text", "__tag": 4046, "value": " keyword is now available for " }, { "__type": "CrossRef", "__tag": 4002, "value": "quantile", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:quantile" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "percentile", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:percentile" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "nanquantile", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:nanquantile" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "nanpercentile", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:nanpercentile" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ". Only " }, { "__type": "InlineCode", "__tag": 4051, "value": "method=\"inverted_cdf\"" }, { "__type": "Text", "__tag": 4046, "value": " supports weights." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24254" } ], "url": "https://github.com/numpy/numpy/pull/24254", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Option to use weights for quantile and percentile functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new tracer mechanism is available which enables tracking of the enabled targets for each optimized function (i.e., that uses hardware-specific SIMD instructions) in the NumPy library. With this enhancement, it becomes possible to precisely monitor the enabled CPU dispatch targets for the dispatched functions." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new function named " }, { "__type": "InlineCode", "__tag": 4051, "value": "opt_func_info" }, { "__type": "Text", "__tag": 4046, "value": " has been added to the new namespace " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.lib.introspect", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.lib.introspect" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", offering this tracing capability. This function allows you to retrieve information about the enabled targets based on function names and data type signatures." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24420" } ], "url": "https://github.com/numpy/numpy/pull/24420", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Improved CPU optimization tracking" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " in compile mode (i.e. " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py -c" }, { "__type": "Text", "__tag": 4046, "value": ") now accepts the " }, { "__type": "InlineCode", "__tag": 4051, "value": "--backend meson" }, { "__type": "Text", "__tag": 4046, "value": " option. This is the default option for Python >=3.12. For older Python versions, " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " will still default to " }, { "__type": "InlineCode", "__tag": 4051, "value": "--backend distutils" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "To support this in realistic use-cases, in compile mode " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " takes a " }, { "__type": "InlineCode", "__tag": 4051, "value": "--dep" }, { "__type": "Text", "__tag": 4046, "value": " flag one or many times which maps to " }, { "__type": "InlineCode", "__tag": 4051, "value": "dependency()" }, { "__type": "Text", "__tag": 4046, "value": " calls in the " }, { "__type": "InlineCode", "__tag": 4051, "value": "meson" }, { "__type": "Text", "__tag": 4046, "value": " backend, and does nothing in the " }, { "__type": "InlineCode", "__tag": 4051, "value": "distutils" }, { "__type": "Text", "__tag": 4046, "value": " backend." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "There are no changes for users of " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " only as a code generator, i.e. without " }, { "__type": "InlineCode", "__tag": 4051, "value": "-c" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24532" } ], "url": "https://github.com/numpy/numpy/pull/24532", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A new Meson backend for " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Both functions and subroutines can be annotated with " }, { "__type": "InlineCode", "__tag": 4051, "value": "bind(c)" }, { "__type": "Text", "__tag": 4046, "value": ". " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " will handle both the correct type mapping, and preserve the unique label for other C interfaces." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Strong", "__tag": 4048, "children": [ { "__type": "Text", "__tag": 4046, "value": "Note:" } ] }, { "__type": "Text", "__tag": 4046, "value": " " }, { "__type": "InlineCode", "__tag": 4051, "value": "bind(c, name = 'routine_name_other_than_fortran_routine')" }, { "__type": "Text", "__tag": 4046, "value": " is not honored by the " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " bindings by design, since " }, { "__type": "InlineCode", "__tag": 4051, "value": "bind(c)" }, { "__type": "Text", "__tag": 4046, "value": " with the " }, { "__type": "InlineCode", "__tag": 4051, "value": "name" }, { "__type": "Text", "__tag": 4046, "value": " is meant to guarantee only the same name in C and Fortran, not in Python and Fortran." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24555" } ], "url": "https://github.com/numpy/numpy/pull/24555", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "bind(c)" }, { "__type": "Text", "__tag": 4046, "value": " support for " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "strict" }, { "__type": "Text", "__tag": 4046, "value": " keyword is now available for " }, { "__type": "CrossRef", "__tag": 4002, "value": "assert_allclose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.testing._private.utils:assert_allclose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "assert_equal", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.testing._private.utils:assert_equal" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "CrossRef", "__tag": 4002, "value": "assert_array_less", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.testing._private.utils:assert_array_less" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ". Setting " }, { "__type": "InlineCode", "__tag": 4051, "value": "strict=True" }, { "__type": "Text", "__tag": 4046, "value": " will disable the broadcasting behaviour for scalars and ensure that input arrays have the same data type." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24680" } ], "url": "https://github.com/numpy/numpy/pull/24680", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24770" } ], "url": "https://github.com/numpy/numpy/pull/24770", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24775" } ], "url": "https://github.com/numpy/numpy/pull/24775", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "InlineCode", "__tag": 4051, "value": "strict" }, { "__type": "Text", "__tag": 4046, "value": " option for several testing functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Add two " }, { "__type": "InlineCode", "__tag": 4051, "value": "find" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "rfind" }, { "__type": "Text", "__tag": 4046, "value": " UFuncs that operate on unicode or byte strings and are used in " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char" }, { "__type": "Text", "__tag": 4046, "value": ". They operate similar to " }, { "__type": "InlineCode", "__tag": 4051, "value": "str.find" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "str.rfind" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24868" } ], "url": "https://github.com/numpy/numpy/pull/24868", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Add " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.core.umath.find" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.core.umath.rfind" }, { "__type": "Text", "__tag": 4046, "value": " UFuncs" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.diagonal", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:diagonal" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.trace", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:trace" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " have been added, which are array API standard-compatible variants of " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.diagonal", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:diagonal" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.trace", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:trace" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ". They differ in the default axis selection which define 2-D sub-arrays." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24887" } ], "url": "https://github.com/numpy/numpy/pull/24887", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "diagonal" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "trace" }, { "__type": "Text", "__tag": 4046, "value": " for " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.long", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:int64" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.ulong", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:uint64" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " have been added as NumPy integers mapping to C's " }, { "__type": "InlineCode", "__tag": 4051, "value": "long" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "unsigned long" }, { "__type": "Text", "__tag": 4046, "value": ". Prior to NumPy 1.24, " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.long" }, { "__type": "Text", "__tag": 4046, "value": " was an alias to Python's " }, { "__type": "InlineCode", "__tag": 4051, "value": "int" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24922" } ], "url": "https://github.com/numpy/numpy/pull/24922", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New " }, { "__type": "InlineCode", "__tag": 4051, "value": "long" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "ulong" }, { "__type": "Text", "__tag": 4046, "value": " dtypes" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.svdvals", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:svdvals" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " has been added. It computes singular values for (a stack of) matrices. Executing " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.svdvals(x)" }, { "__type": "Text", "__tag": 4046, "value": " is the same as calling " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.svd(x, compute_uv=False, hermitian=False)" }, { "__type": "Text", "__tag": 4046, "value": ". This function is compatible with the array API standard." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24940" } ], "url": "https://github.com/numpy/numpy/pull/24940", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "svdvals" }, { "__type": "Text", "__tag": 4046, "value": " for " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.isdtype", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:isdtype" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " was added to provide a canonical way to classify NumPy's dtypes in compliance with the array API standard." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25054" } ], "url": "https://github.com/numpy/numpy/pull/25054", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "InlineCode", "__tag": 4051, "value": "isdtype" }, { "__type": "Text", "__tag": 4046, "value": " function" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.astype", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:astype" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " was added to provide an array API standard-compatible alternative to the " }, { "__type": "InlineRole", "__tag": 4003, "value": "numpy.ndarray.astype", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": " method." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25079" } ], "url": "https://github.com/numpy/numpy/pull/25079", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "InlineCode", "__tag": 4051, "value": "astype" }, { "__type": "Text", "__tag": 4046, "value": " function" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "13 aliases for existing functions were added to improve compatibility with the array API standard:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Trigonometry: " }, { "__type": "InlineCode", "__tag": 4051, "value": "acos" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "acosh" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "asin" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "asinh" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "atan" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "atanh" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "atan2" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Bitwise: " }, { "__type": "InlineCode", "__tag": 4051, "value": "bitwise_left_shift" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "bitwise_invert" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "bitwise_right_shift" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Misc: " }, { "__type": "InlineCode", "__tag": 4051, "value": "concat" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "permute_dims" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "pow" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "In " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg" }, { "__type": "Text", "__tag": 4046, "value": ": " }, { "__type": "InlineCode", "__tag": 4051, "value": "tensordot" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "matmul" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25086" } ], "url": "https://github.com/numpy/numpy/pull/25086", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Array API compatible functions' aliases" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "unique_all", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:unique_all" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "unique_counts", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:unique_counts" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "unique_inverse", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:unique_inverse" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "CrossRef", "__tag": 4002, "value": "unique_values", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:unique_values" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " functions have been added. They provide functionality of " }, { "__type": "CrossRef", "__tag": 4002, "value": "unique", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:unique" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " with different sets of flags. They are array API standard-compatible, and because the number of arrays they return does not depend on the values of input arguments, they are easier to target for JIT compilation." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25088" } ], "url": "https://github.com/numpy/numpy/pull/25088", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New " }, { "__type": "InlineCode", "__tag": 4051, "value": "unique_*" }, { "__type": "Text", "__tag": 4046, "value": " functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy now offers support for calculating the matrix transpose of an array (or stack of arrays). The matrix transpose is equivalent to swapping the last two axes of an array. Both " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.ndarray" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.ma.MaskedArray" }, { "__type": "Text", "__tag": 4046, "value": " now expose a " }, { "__type": "InlineCode", "__tag": 4051, "value": ".mT" }, { "__type": "Text", "__tag": 4046, "value": " attribute, and there is a matching new " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.matrix_transpose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:matrix_transpose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " function." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23762" } ], "url": "https://github.com/numpy/numpy/pull/23762", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Matrix transpose support for ndarrays" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Six new functions and two aliases were added to improve compatibility with the Array API standard for " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ":" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.matrix_norm", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:matrix_norm" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " - Computes the matrix norm of a matrix (or a stack of matrices)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.vector_norm", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:vector_norm" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " - Computes the vector norm of a vector (or batch of vectors)." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.vecdot", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:vecdot" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " - Computes the (vector) dot product of two arrays." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.vecdot", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:vecdot" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " - An alias for " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.vecdot", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:vecdot" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.matrix_transpose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:matrix_transpose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " - An alias for " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.matrix_transpose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:matrix_transpose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25155" } ], "url": "https://github.com/numpy/numpy/pull/25155", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.outer", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:outer" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " has been added. It computes the outer product of two vectors. It differs from " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.outer", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:outer" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " by accepting one-dimensional arrays only. This function is compatible with the array API standard." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25101" } ], "url": "https://github.com/numpy/numpy/pull/25101", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.cross", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:cross" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " has been added. It computes the cross product of two (arrays of) 3-dimensional vectors. It differs from " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.cross", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:cross" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " by accepting three-dimensional vectors only. This function is compatible with the array API standard." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25145" } ], "url": "https://github.com/numpy/numpy/pull/25145", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Array API compatible functions for " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A " }, { "__type": "InlineCode", "__tag": 4051, "value": "correction" }, { "__type": "Text", "__tag": 4046, "value": " argument was added to " }, { "__type": "CrossRef", "__tag": 4002, "value": "var", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:var" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "std", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:std" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", which is an array API standard compatible alternative to " }, { "__type": "InlineCode", "__tag": 4051, "value": "ddof" }, { "__type": "Text", "__tag": 4046, "value": ". As both arguments serve a similar purpose, only one of them can be provided at the same time." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25169" } ], "url": "https://github.com/numpy/numpy/pull/25169", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A " }, { "__type": "InlineCode", "__tag": 4051, "value": "correction" }, { "__type": "Text", "__tag": 4046, "value": " argument for " }, { "__type": "InlineCode", "__tag": 4051, "value": "var" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "std" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "An " }, { "__type": "InlineCode", "__tag": 4051, "value": "ndarray.device" }, { "__type": "Text", "__tag": 4046, "value": " attribute and " }, { "__type": "InlineCode", "__tag": 4051, "value": "ndarray.to_device" }, { "__type": "Text", "__tag": 4046, "value": " method were added to " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.ndarray" }, { "__type": "Text", "__tag": 4046, "value": " for array API standard compatibility." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Additionally, " }, { "__type": "InlineCode", "__tag": 4051, "value": "device" }, { "__type": "Text", "__tag": 4046, "value": " keyword-only arguments were added to: " }, { "__type": "CrossRef", "__tag": 4002, "value": "asarray", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:asarray" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "arange", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:arange" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "empty", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:empty" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "empty_like", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:empty_like" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "eye", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:eye" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "full", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:full" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "full_like", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:full_like" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "linspace", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:linspace" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "ones", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:ones" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "ones_like", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:ones_like" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "zeros", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:zeros" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "CrossRef", "__tag": 4002, "value": "zeros_like", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:zeros_like" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "For all these new arguments, only " }, { "__type": "InlineCode", "__tag": 4051, "value": "device=\"cpu\"" }, { "__type": "Text", "__tag": 4046, "value": " is supported." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25233" } ], "url": "https://github.com/numpy/numpy/pull/25233", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "ndarray.device" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "ndarray.to_device" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "We have added a new variable-width UTF-8 encoded string data type, implementing a \"NumPy array of Python strings\", including support for a user-provided missing data sentinel. It is intended as a drop-in replacement for arrays of Python strings and missing data sentinels using the object dtype. See " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "NEP 55" } ], "url": "https://numpy.org/neps/nep-0055-string_dtype.html", "title": "" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "the documentation", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "user:basics.strings" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for more details." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25347" } ], "url": "https://github.com/numpy/numpy/pull/25347", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "StringDType has been added to NumPy" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "upper" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "rtol" }, { "__type": "Text", "__tag": 4046, "value": " keywords were added to " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.cholesky", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:cholesky" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg.pinv", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:pinv" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", respectively, to improve array API standard compatibility." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "For " }, { "__type": "CrossRef", "__tag": 4002, "value": "pinv", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:pinv" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", if neither " }, { "__type": "InlineCode", "__tag": 4051, "value": "rcond" }, { "__type": "Text", "__tag": 4046, "value": " nor " }, { "__type": "InlineCode", "__tag": 4051, "value": "rtol" }, { "__type": "Text", "__tag": 4046, "value": " is specified, the " }, { "__type": "InlineCode", "__tag": 4051, "value": "rcond" }, { "__type": "Text", "__tag": 4046, "value": "'s default is used. We plan to deprecate and remove " }, { "__type": "InlineCode", "__tag": 4051, "value": "rcond" }, { "__type": "Text", "__tag": 4046, "value": " in the future." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25388" } ], "url": "https://github.com/numpy/numpy/pull/25388", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New keywords for " }, { "__type": "InlineCode", "__tag": 4051, "value": "cholesky" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "pinv" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "New keyword parameters were added to improve array API standard compatibility:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "rtol" }, { "__type": "Text", "__tag": 4046, "value": " was added to " }, { "__type": "CrossRef", "__tag": 4002, "value": "matrix_rank", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg:matrix_rank" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "stable" }, { "__type": "Text", "__tag": 4046, "value": " was added to " }, { "__type": "CrossRef", "__tag": 4002, "value": "sort", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:sort" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "argsort", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:argsort" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25437" } ], "url": "https://github.com/numpy/numpy/pull/25437", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New keywords for " }, { "__type": "InlineCode", "__tag": 4051, "value": "sort" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "argsort" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "linalg.matrix_rank" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy now implements some string operations as ufuncs. The old " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char" }, { "__type": "Text", "__tag": 4046, "value": " namespace is still available, and where possible the string manipulation functions in that namespace have been updated to use the new ufuncs, substantially improving their performance." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Where possible, we suggest updating code to use functions in " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.strings" }, { "__type": "Text", "__tag": 4046, "value": " instead of " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char" }, { "__type": "Text", "__tag": 4046, "value": ". In the future we may deprecate " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.char" }, { "__type": "Text", "__tag": 4046, "value": " in favor of " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.strings" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25463" } ], "url": "https://github.com/numpy/numpy/pull/25463", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.strings" }, { "__type": "Text", "__tag": 4046, "value": " namespace for string ufuncs" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The various FFT routines in " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.fft", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " now do their calculations natively in float, double, or long double precision, depending on the input precision, instead of always calculating in double precision. Hence, the calculation will now be less precise for single and more precise for long double precision. The data type of the output array will now be adjusted accordingly." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Furthermore, all FFT routines have gained an " }, { "__type": "InlineCode", "__tag": 4051, "value": "out" }, { "__type": "Text", "__tag": 4046, "value": " argument that can be used for in-place calculations." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25536" } ], "url": "https://github.com/numpy/numpy/pull/25536", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.fft" }, { "__type": "Text", "__tag": 4046, "value": " support for different precisions and in-place calculations" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy-config" }, { "__type": "Text", "__tag": 4046, "value": " CLI script is available that can be queried for the NumPy version and for compile flags needed to use the NumPy C API. This will allow build systems to better support the use of NumPy as a dependency. Also, a " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.pc" }, { "__type": "Text", "__tag": 4046, "value": " pkg-config file is now included with Numpy. In order to find its location for use with " }, { "__type": "InlineCode", "__tag": 4051, "value": "PKG_CONFIG_PATH" }, { "__type": "Text", "__tag": 4046, "value": ", use " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy-config --pkgconfigdir" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25730" } ], "url": "https://github.com/numpy/numpy/pull/25730", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "configtool and pkg-config support" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The main " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy" }, { "__type": "Text", "__tag": 4046, "value": " namespace now supports the array API standard. See " }, { "__type": "CrossRef", "__tag": 4002, "value": "array-api-standard-compatibility", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:array_api" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for details." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25911" } ], "url": "https://github.com/numpy/numpy/pull/25911", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Array API standard support in the main namespace" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [], "title": [ { "__type": "Text", "__tag": 4046, "value": "Improvements" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Blockquote", "__tag": 4059, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25651" } ], "url": "https://github.com/numpy/numpy/pull/25651", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Strings are now supported by " }, { "__type": "InlineCode", "__tag": 4051, "value": "any" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "all" }, { "__type": "Text", "__tag": 4046, "value": ", and the logical ufuncs." } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "CrossRef", "__tag": 4002, "value": "numpy.memmap", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:memmap" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " can now be created with any integer sequence as the " }, { "__type": "InlineCode", "__tag": 4051, "value": "shape" }, { "__type": "Text", "__tag": 4046, "value": " argument, such as a list or numpy array of integers. Previously, only the types of tuple and int could be used without raising an error." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23729" } ], "url": "https://github.com/numpy/numpy/pull/23729", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Integer sequences as the shape argument for " }, { "__type": "InlineCode", "__tag": 4051, "value": "memmap" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.errstate", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:errstate" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " context manager/decorator is now faster and safer. Previously, it was not context safe and had (rare) issues with thread-safety." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23936" } ], "url": "https://github.com/numpy/numpy/pull/23936", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "errstate" }, { "__type": "Text", "__tag": 4046, "value": " is now faster and context safe" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The first introduction of the Google Highway library, using VQSort on AArch64. Execution time is improved by up to 16x in some cases, see the PR for benchmark results. Extensions to other platforms will be done in the future." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24018" } ], "url": "https://github.com/numpy/numpy/pull/24018", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "AArch64 quicksort speed improved by using Highway's VQSort" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The underlying C types for all of NumPy's complex types have been changed to use C99 complex types." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "While this change does not affect the memory layout of complex types, it changes the API to be used to directly retrieve or write the real or complex part of the complex number, since direct field access (as in " }, { "__type": "InlineCode", "__tag": 4051, "value": "c.real" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "c.imag" }, { "__type": "Text", "__tag": 4046, "value": ") is no longer an option. You can now use utilities provided in " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy/npy_math.h" }, { "__type": "Text", "__tag": 4046, "value": " to do these operations, like this:" } ] }, { "__type": "Code", "__tag": 4050, "value": "npy_cdouble c;\n npy_csetreal(&c, 1.0);\n npy_csetimag(&c, 0.0);\n printf(\"%d + %di\\n\", npy_creal(c), npy_cimag(c));", "execution_status": null } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "To ease cross-version compatibility, equivalent macros and a compatibility layer have been added which can be used by downstream packages to continue to support both NumPy 1.x and 2.x. See " }, { "__type": "CrossRef", "__tag": 4002, "value": "complex-numbers", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "reference:c-api:coremath" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": " for more info." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy/npy_common.h" }, { "__type": "Text", "__tag": 4046, "value": " now includes " }, { "__type": "InlineCode", "__tag": 4051, "value": "complex.h" }, { "__type": "Text", "__tag": 4046, "value": ", which means that " }, { "__type": "InlineCode", "__tag": 4051, "value": "complex" }, { "__type": "Text", "__tag": 4046, "value": " is now a reserved keyword." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24085" } ], "url": "https://github.com/numpy/numpy/pull/24085", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Complex types - underlying C type changes" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Previously, users would have to define their own custom " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2cmap" }, { "__type": "Text", "__tag": 4046, "value": " file to use type mappings defined by the Fortran2003 " }, { "__type": "InlineCode", "__tag": 4051, "value": "iso_c_binding" }, { "__type": "Text", "__tag": 4046, "value": " intrinsic module. These type maps are now natively supported by " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24555" } ], "url": "https://github.com/numpy/numpy/pull/24555", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "f2py" }, { "__type": "Text", "__tag": 4046, "value": " now handles " }, { "__type": "InlineCode", "__tag": 4051, "value": "common" }, { "__type": "Text", "__tag": 4046, "value": " blocks which have " }, { "__type": "InlineCode", "__tag": 4051, "value": "kind" }, { "__type": "Text", "__tag": 4046, "value": " specifications from modules. This further expands the usability of intrinsics like " }, { "__type": "InlineCode", "__tag": 4051, "value": "iso_fortran_env" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "iso_c_binding" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25186" } ], "url": "https://github.com/numpy/numpy/pull/25186", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "iso_c_binding" }, { "__type": "Text", "__tag": 4046, "value": " support and improved common blocks for " }, { "__type": "InlineCode", "__tag": 4051, "value": "f2py" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The third argument to functions like " }, { "__type": "CrossRef", "__tag": 4002, "value": "assert_equal", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.testing._private.utils:assert_equal" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " now has " }, { "__type": "InlineCode", "__tag": 4051, "value": "str" }, { "__type": "Text", "__tag": 4046, "value": " called on it automatically. This way it mimics the built-in " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert" }, { "__type": "Text", "__tag": 4046, "value": " statement, where " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert_equal(a, b, obj)" }, { "__type": "Text", "__tag": 4046, "value": " works like " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert a == b, obj" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24877" } ], "url": "https://github.com/numpy/numpy/pull/24877", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Call " }, { "__type": "InlineCode", "__tag": 4051, "value": "str" }, { "__type": "Text", "__tag": 4046, "value": " automatically on third argument to functions like " }, { "__type": "InlineCode", "__tag": 4051, "value": "assert_equal" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The keywords " }, { "__type": "InlineCode", "__tag": 4051, "value": "atol" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "rtol" }, { "__type": "Text", "__tag": 4046, "value": " in " }, { "__type": "CrossRef", "__tag": 4002, "value": "isclose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:isclose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "allclose", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:allclose" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " now accept both scalars and arrays. An array, if given, must broadcast to the shapes of the first two array arguments." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24878" } ], "url": "https://github.com/numpy/numpy/pull/24878", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Support for array-like " }, { "__type": "InlineCode", "__tag": 4051, "value": "atol" }, { "__type": "Text", "__tag": 4046, "value": "/" }, { "__type": "InlineCode", "__tag": 4051, "value": "rtol" }, { "__type": "Text", "__tag": 4046, "value": " in " }, { "__type": "InlineCode", "__tag": 4051, "value": "isclose" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "allclose" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Previously, some " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.testing", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.testing" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " assertions printed messages that referred to the actual and desired results as " }, { "__type": "InlineCode", "__tag": 4051, "value": "x" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "y" }, { "__type": "Text", "__tag": 4046, "value": ". Now, these values are consistently referred to as " }, { "__type": "InlineCode", "__tag": 4051, "value": "ACTUAL" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "DESIRED" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24931" } ], "url": "https://github.com/numpy/numpy/pull/24931", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Consistent failure messages in test functions" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "fftn", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:fftn" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "ifftn", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:ifftn" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "rfftn", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:rfftn" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "irfftn", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:irfftn" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "fft2", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:fft2" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "ifft2", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:ifft2" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "rfft2", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:rfft2" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "irfft2", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft:irfft2" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " functions now use the whole input array along the axis " }, { "__type": "InlineCode", "__tag": 4051, "value": "i" }, { "__type": "Text", "__tag": 4046, "value": " if " }, { "__type": "InlineCode", "__tag": 4051, "value": "s[i] == -1" }, { "__type": "Text", "__tag": 4046, "value": ", in line with the array API standard." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25495" } ], "url": "https://github.com/numpy/numpy/pull/25495", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "n-D FFT transforms allow " }, { "__type": "InlineCode", "__tag": 4051, "value": "s[i] == -1" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "PyUnicodeScalarObject" }, { "__type": "Text", "__tag": 4046, "value": " holds a " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyUnicodeObject" }, { "__type": "Text", "__tag": 4046, "value": ", which is not available when using " }, { "__type": "InlineCode", "__tag": 4051, "value": "Py_LIMITED_API" }, { "__type": "Text", "__tag": 4046, "value": ". Add guards to hide it and consequently also make the " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArrayScalar_VAL" }, { "__type": "Text", "__tag": 4046, "value": " macro hidden." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25531" } ], "url": "https://github.com/numpy/numpy/pull/25531", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Guard PyArrayScalar_VAL and PyUnicodeScalarObject for the limited API" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.gradient()" }, { "__type": "Text", "__tag": 4046, "value": " now returns a tuple rather than a list making the return value immutable." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23861" } ], "url": "https://github.com/numpy/numpy/pull/23861", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Being fully context and thread-safe, " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.errstate" }, { "__type": "Text", "__tag": 4046, "value": " can only be entered once now." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.setbufsize" }, { "__type": "Text", "__tag": 4046, "value": " is now tied to " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.errstate()" }, { "__type": "Text", "__tag": 4046, "value": ": leaving an " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.errstate" }, { "__type": "Text", "__tag": 4046, "value": " context will also reset the " }, { "__type": "InlineCode", "__tag": 4051, "value": "bufsize" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23936" } ], "url": "https://github.com/numpy/numpy/pull/23936", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "A new public " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.lib.array_utils" }, { "__type": "Text", "__tag": 4046, "value": " submodule has been introduced and it currently contains three functions: " }, { "__type": "InlineCode", "__tag": 4051, "value": "byte_bounds" }, { "__type": "Text", "__tag": 4046, "value": " (moved from " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.lib.utils" }, { "__type": "Text", "__tag": 4046, "value": "), " }, { "__type": "InlineCode", "__tag": 4051, "value": "normalize_axis_tuple" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "normalize_axis_index" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24540" } ], "url": "https://github.com/numpy/numpy/pull/24540", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Introduce " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.bool", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:bool" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " as the new canonical name for NumPy's boolean dtype, and make " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.bool_", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:bool" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " an alias to it. Note that until NumPy 1.24, " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.bool" }, { "__type": "Text", "__tag": 4046, "value": " was an alias to Python's builtin " }, { "__type": "InlineCode", "__tag": 4051, "value": "bool" }, { "__type": "Text", "__tag": 4046, "value": ". The new name helps with array API standard compatibility and is a more intuitive name." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25080" } ], "url": "https://github.com/numpy/numpy/pull/25080", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "dtype.flags" }, { "__type": "Text", "__tag": 4046, "value": " value was previously stored as a signed integer. This means that the aligned dtype struct flag lead to negative flags being set (-128 rather than 128). This flag is now stored unsigned (positive). Code which checks flags manually may need to adapt. This may include code compiled with Cython 0.29.x." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25816" } ], "url": "https://github.com/numpy/numpy/pull/25816", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ] } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Changes" } ], "level": 1, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "As per " }, { "__type": "InlineRole", "__tag": 4003, "value": "NEP 51 ", "domain": null, "role": "ref", "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ", the scalar representation has been updated to include the type information to avoid confusion with Python scalars." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Scalars are now printed as " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.float64(3.0)" }, { "__type": "Text", "__tag": 4046, "value": " rather than just " }, { "__type": "InlineCode", "__tag": 4051, "value": "3.0" }, { "__type": "Text", "__tag": 4046, "value": ". This may disrupt workflows that store representations of numbers (e.g., to files) making it harder to read them. They should be stored as explicit strings, for example by using " }, { "__type": "InlineCode", "__tag": 4051, "value": "str()" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "f\"{scalar!s}\"" }, { "__type": "Text", "__tag": 4046, "value": ". For the time being, affected users can use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.set_printoptions(legacy=\"1.25\")" }, { "__type": "Text", "__tag": 4046, "value": " to get the old behavior (with possibly a few exceptions). Documentation of downstream projects may require larger updates, if code snippets are tested. We are working on tooling for " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "doctest-plus" } ], "url": "https://github.com/scientific-python/pytest-doctestplus/issues/107", "title": "" }, { "__type": "Text", "__tag": 4046, "value": " to facilitate updates." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-22449" } ], "url": "https://github.com/numpy/numpy/pull/22449", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Representation of NumPy scalars changed" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy strings previously were inconsistent about how they defined if the string is " }, { "__type": "InlineCode", "__tag": 4051, "value": "True" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "False" }, { "__type": "Text", "__tag": 4046, "value": " and the definition did not match the one used by Python. Strings are now considered " }, { "__type": "InlineCode", "__tag": 4051, "value": "True" }, { "__type": "Text", "__tag": 4046, "value": " when they are non-empty and " }, { "__type": "InlineCode", "__tag": 4051, "value": "False" }, { "__type": "Text", "__tag": 4046, "value": " when they are empty. This changes the following distinct cases:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Casts from string to boolean were previously roughly equivalent to " }, { "__type": "InlineCode", "__tag": 4051, "value": "string_array.astype(np.int64).astype(bool)" }, { "__type": "Text", "__tag": 4046, "value": ", meaning that only valid integers could be cast. Now a string of " }, { "__type": "InlineCode", "__tag": 4051, "value": "\"0\"" }, { "__type": "Text", "__tag": 4046, "value": " will be considered " }, { "__type": "InlineCode", "__tag": 4051, "value": "True" }, { "__type": "Text", "__tag": 4046, "value": " since it is not empty. If you need the old behavior, you may use the above step (casting to integer first) or " }, { "__type": "InlineCode", "__tag": 4051, "value": "string_array == \"0\"" }, { "__type": "Text", "__tag": 4046, "value": " (if the input is only ever " }, { "__type": "InlineCode", "__tag": 4051, "value": "0" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "1" }, { "__type": "Text", "__tag": 4046, "value": "). To get the new result on old NumPy versions use " }, { "__type": "InlineCode", "__tag": 4051, "value": "string_array != \"\"" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.nonzero(string_array)" }, { "__type": "Text", "__tag": 4046, "value": " previously ignored whitespace so that a string only containing whitespace was considered " }, { "__type": "InlineCode", "__tag": 4051, "value": "False" }, { "__type": "Text", "__tag": 4046, "value": ". Whitespace is now considered " }, { "__type": "InlineCode", "__tag": 4051, "value": "True" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "This change does not affect " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.loadtxt" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.fromstring" }, { "__type": "Text", "__tag": 4046, "value": ", or " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": ". The first two still use the integer definition, while " }, { "__type": "InlineCode", "__tag": 4051, "value": "genfromtxt" }, { "__type": "Text", "__tag": 4046, "value": " continues to match for " }, { "__type": "InlineCode", "__tag": 4051, "value": "\"true\"" }, { "__type": "Text", "__tag": 4046, "value": " (ignoring case). However, if " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.bool_" }, { "__type": "Text", "__tag": 4046, "value": " is used as a converter the result will change." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The change does affect " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.fromregex" }, { "__type": "Text", "__tag": 4046, "value": " as it uses direct assignments." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-23871" } ], "url": "https://github.com/numpy/numpy/pull/23871", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Truthiness of NumPy strings changed" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Often when the standard deviation is needed the mean is also needed. The same holds for the variance and the mean. Until now the mean is then calculated twice, the change introduced here for the " }, { "__type": "CrossRef", "__tag": 4002, "value": "var", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:var" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "std", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:std" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " functions allows for passing in a precalculated mean as an keyword argument. See the docstrings for details and an example illustrating the speed-up." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24126" } ], "url": "https://github.com/numpy/numpy/pull/24126", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "A " }, { "__type": "InlineCode", "__tag": 4051, "value": "mean" }, { "__type": "Text", "__tag": 4046, "value": " keyword was added to var and std function" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.datetime64", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:datetime64" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " method now issues a UserWarning rather than a DeprecationWarning whenever a timezone is included in the datetime string that is provided." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24193" } ], "url": "https://github.com/numpy/numpy/pull/24193", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Remove datetime64 deprecation warning when constructing with timezone" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The default NumPy integer is now 64-bit on all 64-bit systems as the historic 32-bit default on Windows was a common source of issues. Most users should not notice this. The main issues may occur with code interfacing with libraries written in a compiled language like C. For more information see " }, { "__type": "CrossRef", "__tag": 4002, "value": "migration_windows_int64", "reference": { "__type": "LocalRef", "__tag": 4022, "kind": "docs", "path": "numpy_2_0_migration_guide" }, "kind": "exists" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24224" } ], "url": "https://github.com/numpy/numpy/pull/24224", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Default integer dtype is now 64-bit on 64-bit Windows" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Accessing " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.core" }, { "__type": "Text", "__tag": 4046, "value": " now emits a DeprecationWarning. In practice we have found that most downstream usage of " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.core" }, { "__type": "Text", "__tag": 4046, "value": " was to access functionality that is available in the main " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy" }, { "__type": "Text", "__tag": 4046, "value": " namespace. If for some reason you are using functionality in " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.core" }, { "__type": "Text", "__tag": 4046, "value": " that is not available in the main " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy" }, { "__type": "Text", "__tag": 4046, "value": " namespace, this means you are likely using private NumPy internals. You can still access these internals via " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy._core" }, { "__type": "Text", "__tag": 4046, "value": " without a deprecation warning but we do not provide any backward compatibility guarantees for NumPy internals. Please open an issue if you think a mistake was made and something needs to be made public." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24634" } ], "url": "https://github.com/numpy/numpy/pull/24634", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The \"relaxed strides\" debug build option, which was previously enabled through the " }, { "__type": "InlineCode", "__tag": 4051, "value": "NPY_RELAXED_STRIDES_DEBUG" }, { "__type": "Text", "__tag": 4046, "value": " environment variable or the " }, { "__type": "InlineCode", "__tag": 4051, "value": "-Drelaxed-strides-debug" }, { "__type": "Text", "__tag": 4046, "value": " config-settings flag has been removed." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24717" } ], "url": "https://github.com/numpy/numpy/pull/24717", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Renamed " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.core" }, { "__type": "Text", "__tag": 4046, "value": " to " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy._core" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Due to the actual use of these types almost always matching the use of " }, { "__type": "InlineCode", "__tag": 4051, "value": "size_t" }, { "__type": "Text", "__tag": 4046, "value": "/" }, { "__type": "InlineCode", "__tag": 4051, "value": "Py_ssize_t" }, { "__type": "Text", "__tag": 4046, "value": " this is now the definition in C. Previously, it matched " }, { "__type": "InlineCode", "__tag": 4051, "value": "intptr_t" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "uintptr_t" }, { "__type": "Text", "__tag": 4046, "value": " which would often have been subtly incorrect. This has no effect on the vast majority of machines since the size of these types only differ on extremely niche platforms." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "However, it means that:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Pointers may not necessarily fit into an " }, { "__type": "InlineCode", "__tag": 4051, "value": "intp" }, { "__type": "Text", "__tag": 4046, "value": " typed array anymore. The " }, { "__type": "InlineCode", "__tag": 4051, "value": "p" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "P" }, { "__type": "Text", "__tag": 4046, "value": " character codes can still be used, however." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Creating " }, { "__type": "InlineCode", "__tag": 4051, "value": "intptr_t" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "uintptr_t" }, { "__type": "Text", "__tag": 4046, "value": " typed arrays in C remains possible in a cross-platform way via " }, { "__type": "InlineCode", "__tag": 4051, "value": "PyArray_DescrFromType('p')" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The new character codes " }, { "__type": "InlineCode", "__tag": 4051, "value": "nN" }, { "__type": "Text", "__tag": 4046, "value": " were introduced." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "It is now correct to use the Python C-API functions when parsing to " }, { "__type": "InlineCode", "__tag": 4051, "value": "npy_intp" }, { "__type": "Text", "__tag": 4046, "value": " typed arguments." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24888" } ], "url": "https://github.com/numpy/numpy/pull/24888", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Redefinition of " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.intp" }, { "__type": "Text", "__tag": 4046, "value": "/" }, { "__type": "InlineCode", "__tag": 4051, "value": "np.uintp" }, { "__type": "Text", "__tag": 4046, "value": " (almost never a change)" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.fft.helper" }, { "__type": "Text", "__tag": 4046, "value": " was renamed to " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.fft._helper" }, { "__type": "Text", "__tag": 4046, "value": " to indicate that it is a private submodule. All public functions exported by it should be accessed from " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.fft", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.fft" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24945" } ], "url": "https://github.com/numpy/numpy/pull/24945", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.fft.helper" }, { "__type": "Text", "__tag": 4046, "value": " made private" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg.linalg" }, { "__type": "Text", "__tag": 4046, "value": " was renamed to " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg._linalg" }, { "__type": "Text", "__tag": 4046, "value": " to indicate that it is a private submodule. All public functions exported by it should be accessed from " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.linalg", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy.linalg" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-24946" } ], "url": "https://github.com/numpy/numpy/pull/24946", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "numpy.linalg.linalg" }, { "__type": "Text", "__tag": 4046, "value": " made private" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "In some cases " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=32" }, { "__type": "Text", "__tag": 4046, "value": " or for concatenate any large value was the same as " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" }, { "__type": "Text", "__tag": 4046, "value": ". Except for " }, { "__type": "InlineCode", "__tag": 4051, "value": "concatenate" }, { "__type": "Text", "__tag": 4046, "value": " this was deprecate. Any out of bound axis value will now error, make sure to use " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25149" } ], "url": "https://github.com/numpy/numpy/pull/25149", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Out-of-bound axis not the same as " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Now " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.array", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:array" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.asarray", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:asarray" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " support three values for " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " parameter:" } ] }, { "__type": "BulletList", "__tag": 4053, "ordered": false, "start": 1, "children": [ { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "None" }, { "__type": "Text", "__tag": 4046, "value": " - A copy will only be made if it is necessary." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "True" }, { "__type": "Text", "__tag": 4046, "value": " - Always make a copy." } ] } ] }, { "__type": "ListItem", "__tag": 4054, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "False" }, { "__type": "Text", "__tag": 4046, "value": " - Never make a copy. If a copy is required a " }, { "__type": "InlineCode", "__tag": 4051, "value": "ValueError" }, { "__type": "Text", "__tag": 4046, "value": " is raised." } ] } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The meaning of " }, { "__type": "InlineCode", "__tag": 4051, "value": "False" }, { "__type": "Text", "__tag": 4046, "value": " changed as it now raises an exception if a copy is needed." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25168" } ], "url": "https://github.com/numpy/numpy/pull/25168", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "New " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " keyword meaning for " }, { "__type": "InlineCode", "__tag": 4051, "value": "array" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "asarray" }, { "__type": "Text", "__tag": 4046, "value": " constructors" } ], "level": 2, "target": "copy-keyword-changes-2.0" }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "NumPy will pass " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " to the " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array__" }, { "__type": "Text", "__tag": 4046, "value": " special method in situations where it would be set to a non-default value (e.g. in a call to " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.asarray(some_object, copy=False)" }, { "__type": "Text", "__tag": 4046, "value": "). Currently, if an unexpected keyword argument error is raised after this, NumPy will print a warning and re-try without the " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " keyword argument. Implementations of objects implementing the " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array__" }, { "__type": "Text", "__tag": 4046, "value": " protocol should accept a " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " keyword argument with the same meaning as when passed to " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.array", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:array" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "CrossRef", "__tag": 4002, "value": "numpy.asarray", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:asarray" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25168" } ], "url": "https://github.com/numpy/numpy/pull/25168", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "__array__" }, { "__type": "Text", "__tag": 4046, "value": " special method now takes a " }, { "__type": "InlineCode", "__tag": 4051, "value": "copy" }, { "__type": "Text", "__tag": 4046, "value": " keyword argument." } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The interpretation of strings with commas is changed slightly, in that a trailing comma will now always create a structured dtype. E.g., where previously " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"i\")" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"i,\")" }, { "__type": "Text", "__tag": 4046, "value": " were treated as identical, now " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"i,\")" }, { "__type": "Text", "__tag": 4046, "value": " will create a structured dtype, with a single field. This is analogous to " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"i,i\")" }, { "__type": "Text", "__tag": 4046, "value": " creating a structured dtype with two fields, and makes the behaviour consistent with that expected of tuples." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "At the same time, the use of single number surrounded by parenthesis to indicate a sub-array shape, like in " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"(2)i,\")" }, { "__type": "Text", "__tag": 4046, "value": ", is deprecated. Instead; one should use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"(2,)i\")" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"2i\")" }, { "__type": "Text", "__tag": 4046, "value": ". Eventually, using a number in parentheses will raise an exception, like is the case for initializations without a comma, like " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.dtype(\"(2)i\")" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25434" } ], "url": "https://github.com/numpy/numpy/pull/25434", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Cleanup of initialization of " }, { "__type": "InlineCode", "__tag": 4051, "value": "numpy.dtype" }, { "__type": "Text", "__tag": 4046, "value": " with strings with commas" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Following the array API standard, the complex sign is now calculated as " }, { "__type": "InlineCode", "__tag": 4051, "value": "z / |z|" }, { "__type": "Text", "__tag": 4046, "value": " (instead of the rather less logical case where the sign of the real part was taken, unless the real part was zero, in which case the sign of the imaginary part was returned). Like for real numbers, zero is returned if " }, { "__type": "InlineCode", "__tag": 4051, "value": "z==0" }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25441" } ], "url": "https://github.com/numpy/numpy/pull/25441", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Change in how complex sign is calculated" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "Functions that returned a list of ndarrays have been changed to return a tuple of ndarrays instead. Returning tuples consistently whenever a sequence of arrays is returned makes it easier for JIT compilers like Numba, as well as for static type checkers in some cases, to support these functions. Changed functions are: " }, { "__type": "CrossRef", "__tag": 4002, "value": "atleast_1d", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:atleast_1d" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "atleast_2d", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:atleast_2d" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "atleast_3d", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:atleast_3d" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "broadcast_arrays", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:broadcast_arrays" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "meshgrid", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:meshgrid" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "InlineRole", "__tag": 4003, "value": "~numpy.ogrid", "domain": null, "role": null, "inventory": null }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "CrossRef", "__tag": 4002, "value": "histogramdd", "reference": { "__type": "RefInfo", "__tag": 4000, "module": "numpy", "version": "*", "kind": "api", "path": "numpy:histogramdd" }, "kind": "module" }, { "__type": "Text", "__tag": 4046, "value": "." } ] } ], "title": [ { "__type": "Text", "__tag": 4046, "value": "Return types of functions that returned a list of arrays" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "When multi-dimensional inputs are passed to " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.unique" }, { "__type": "Text", "__tag": 4046, "value": " with " }, { "__type": "InlineCode", "__tag": 4051, "value": "return_inverse=True" }, { "__type": "Text", "__tag": 4046, "value": ", the " }, { "__type": "InlineCode", "__tag": 4051, "value": "unique_inverse" }, { "__type": "Text", "__tag": 4046, "value": " output is now shaped such that the input can be reconstructed directly using " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.take(unique, unique_inverse)" }, { "__type": "Text", "__tag": 4046, "value": " when " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" }, { "__type": "Text", "__tag": 4046, "value": ", and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.take_along_axis(unique, unique_inverse, axis=axis)" }, { "__type": "Text", "__tag": 4046, "value": " otherwise." } ] }, { "__type": "Admonition", "__tag": 4056, "kind": "note", "base_type": "note", "children": [ { "__type": "AdmonitionTitle", "__tag": 4055, "children": [ { "__type": "Text", "__tag": 4046, "value": "note " } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "This change was reverted in 2.0.1 except for " }, { "__type": "InlineCode", "__tag": 4051, "value": "axis=None" }, { "__type": "Text", "__tag": 4046, "value": ". The correct reconstruction is always " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.take(unique, unique_inverse, axis=axis)" }, { "__type": "Text", "__tag": 4046, "value": ". When 2.0.0 needs to be supported, add " }, { "__type": "InlineCode", "__tag": 4051, "value": "unique_inverse.reshape(-1)" }, { "__type": "Text", "__tag": 4046, "value": " to code." } ] } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25553" } ], "url": "https://github.com/numpy/numpy/pull/25553", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ", " }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25570" } ], "url": "https://github.com/numpy/numpy/pull/25570", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.unique" }, { "__type": "Text", "__tag": 4046, "value": " " }, { "__type": "InlineCode", "__tag": 4051, "value": "return_inverse" }, { "__type": "Text", "__tag": 4046, "value": " shape for multi-dimensional inputs" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "The " }, { "__type": "InlineCode", "__tag": 4051, "value": "any" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "all" }, { "__type": "Text", "__tag": 4046, "value": " functions and methods now return booleans also for object arrays. Previously, they did a reduction which behaved like the Python " }, { "__type": "InlineCode", "__tag": 4051, "value": "or" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "and" }, { "__type": "Text", "__tag": 4046, "value": " operators which evaluates to one of the arguments. You can use " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.logical_or.reduce" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "np.logical_and.reduce" }, { "__type": "Text", "__tag": 4046, "value": " to achieve the previous behavior." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-25712" } ], "url": "https://github.com/numpy/numpy/pull/25712", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "any" }, { "__type": "Text", "__tag": 4046, "value": " and " }, { "__type": "InlineCode", "__tag": 4051, "value": "all" }, { "__type": "Text", "__tag": 4046, "value": " return booleans for object arrays" } ], "level": 2, "target": null }, { "__type": "Section", "__tag": 4015, "children": [ { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.can_cast" }, { "__type": "Text", "__tag": 4046, "value": " cannot be called with Python int, float, or complex instances anymore. This is because NEP 50 means that the result of " }, { "__type": "InlineCode", "__tag": 4051, "value": "can_cast" }, { "__type": "Text", "__tag": 4046, "value": " must not depend on the value passed in. Unfortunately, for Python scalars whether a cast should be considered " }, { "__type": "InlineCode", "__tag": 4051, "value": "\"same_kind\"" }, { "__type": "Text", "__tag": 4046, "value": " or " }, { "__type": "InlineCode", "__tag": 4051, "value": "\"safe\"" }, { "__type": "Text", "__tag": 4046, "value": " may depend on the context and value so that this is currently not implemented. In some cases, this means you may have to add a specific path for: " }, { "__type": "InlineCode", "__tag": 4051, "value": "if type(obj) in (int, float, complex): ..." }, { "__type": "Text", "__tag": 4046, "value": "." } ] }, { "__type": "Paragraph", "__tag": 4045, "children": [ { "__type": "Text", "__tag": 4046, "value": "(" }, { "__type": "Link", "__tag": 4049, "children": [ { "__type": "Text", "__tag": 4046, "value": "gh-26393" } ], "url": "https://github.com/numpy/numpy/pull/26393", "title": "" }, { "__type": "Text", "__tag": 4046, "value": ")" } ] } ], "title": [ { "__type": "InlineCode", "__tag": 4051, "value": "np.can_cast" }, { "__type": "Text", "__tag": 4046, "value": " cannot be called on Python int, float, or complex" } ], "level": 2, "target": null } ], "local_refs": [] }