conf.py

#
# PyTorch documentation build configuration file, created by
# sphinx-quickstart on Fri Dec 23 13:31:47 2016.
#
# This file is execfile()d with the current directory set to its
# containing dir.
#
# Note that not all possible configuration values are present in this
# autogenerated file.
#
# All configuration values have a default; values that are commented out
# serve to show the default.


import functools
import inspect
import os
import pkgutil
import re
import sys
from os import path

# source code directory, relative to this file, for sphinx-autobuild
# sys.path.insert(0, os.path.abspath('../..'))
import torch


# If extensions (or modules to document with autodoc) are in another directory,
# add these directories to sys.path here. If the directory is relative to the
# documentation root, use os.path.abspath to make it absolute, like shown here.


try:
    import torchvision  # noqa: F401
except ImportError:
    import warnings

    warnings.warn('unable to load "torchvision" package')

RELEASE = os.environ.get("RELEASE", False)

import pytorch_sphinx_theme2


html_theme = "pytorch_sphinx_theme2"
html_theme_path = [pytorch_sphinx_theme2.get_html_theme_path()]

# Add the source directory to sys.path so that redirects.py can be imported
sys.path.insert(0, os.path.abspath(os.path.dirname(__file__)))
from redirects import redirects  # noqa: F401


# -- General configuration ------------------------------------------------

# Add any Sphinx extension module names here, as strings. They can be
# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
# ones.
extensions = [
    "sphinx.ext.autodoc",
    "sphinx.ext.autosummary",
    "sphinx.ext.doctest",
    "sphinx.ext.intersphinx",
    "sphinx.ext.todo",
    "sphinx.ext.coverage",
    "sphinx.ext.napoleon",
    "sphinx.ext.autosectionlabel",
    "sphinxcontrib.katex",
    "sphinx_copybutton",
    "sphinx_design",
    "myst_nb",
    "sphinx.ext.linkcode",
    "sphinxcontrib.mermaid",
    "sphinx_sitemap",
    "sphinx_reredirects",
]

myst_enable_extensions = [
    "colon_fence",
    "deflist",
    "html_image",
]

# Don't execute notebooks during the docs build. Notebook correctness is
# verified by the separate docs_test CI job; re-executing them here just
# adds ~3 minutes to the build for no benefit.
nb_execution_mode = "off"
suppress_warnings = ["myst-nb.lexer"]

html_baseurl = "https://docs.pytorch.org/docs/stable/"  # needed for sphinx-sitemap
sitemap_locales = [None]
sitemap_excludes = [
    "search.html",
    "genindex.html",
]
sitemap_url_scheme = "{link}"

html_additional_pages = {
    "404": "404.html",
}

# build the templated autosummary files
autosummary_generate = True
numpydoc_show_class_members = False

# autosectionlabel throws warnings if section names are duplicated.
# The following tells autosectionlabel to not throw a warning for
# duplicated section names that are in different documents.
autosectionlabel_prefix_document = True

# katex options
#
#

katex_prerender = True

# General information about the project.
project = "PyTorch"
copyright = "PyTorch Contributors"
author = "PyTorch Contributors"
torch_version = str(torch.__version__)

# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
# built documents.
#
# The short X.Y version.
# TODO: change to [:2] at v1.0
version = "main (" + torch_version + " )"
# The full version, including alpha/beta/rc tags.
release = "main"

# Customized html_title here.
# Default is " ".join(project, release, "documentation") if not set
if RELEASE:
    # Turn 1.11.0aHASH into 1.11
    # Note: the release candidates should no longer have the aHASH suffix, but in any
    # case we wish to leave only major.minor, even for rc builds.
    version = ".".join(torch_version.split(".")[:2])
    html_title = " ".join((project, version, "documentation"))
    release = version

switcher_version = "main" if not RELEASE else version

html_static_path = ["_static"]
html_theme_options = {
    "analytics_id": "GTM-T8XT4PS",
    "canonical_url": "https://docs.pytorch.org/docs/stable/",
    "switcher": {
        "json_url": "https://docs.pytorch.org/docs/pytorch-versions.json",
        "version_match": switcher_version,
    },
    "show_toc_level": 2,
    "navigation_with_keys": False,
    "external_links": [
        {
            "name": "Tutorials",
            "url": "https://docs.pytorch.org/tutorials/",
        },
    ],
    "show_version_warning_banner": True,
    "llm_generate_full": "false",
    "icon_links": [
        {
            "name": "X",
            "url": "https://x.com/PyTorch",
            "icon": "fa-brands fa-x-twitter",
        },
        {
            "name": "GitHub",
            "url": "https://github.com/pytorch/pytorch",
            "icon": "fa-brands fa-github",
        },
        {
            "name": "PyTorch Forum",
            "url": "https://discuss.pytorch.org/",
            "icon": "fa-brands fa-discourse",
        },
        {
            "name": "PyPi",
            "url": "https://pypi.org/project/torch/",
            "icon": "fa-brands fa-python",
        },
    ],
    "navbar_align": "left",
    "navbar_start": ["navbar-logo", "version-switcher"],
    "navbar_center": ["navbar-nav"],
    "navbar_end": ["search-field-custom", "theme-switcher", "navbar-icon-links"],
    "header_links_before_dropdown": 6,
    "navbar_persistent": [],
    "use_edit_page_button": True,
    # RunLLM Widget Configuration (uncomment and set assistant_id to enable)
    # Each repository should have its own unique assistant_id from RunLLM
    "runllm_assistant_id": "834",
    "runllm_name": "PyTorch Assistant",
    "pytorch_project": "docs",
}

theme_variables = pytorch_sphinx_theme2.get_theme_variables()
html_context = {
    "github_url": "https://github.com",
    "github_user": "pytorch",
    "github_repo": "pytorch",
    "feedback_url": "https://github.com/pytorch/pytorch",
    "github_version": "main",
    "pytorch_project": "docs",
    "doc_path": "docs/source",
    "theme_variables": theme_variables,
    # library links are defined in
    # pytorch_sphinx_theme2/pytorch_sphinx_theme2/links.json
    "library_links": theme_variables.get("library_links", []),
    "version": version,
    "date_info": {
        "paths_to_skip": ["generated/", "index"],
    },
}

napoleon_use_ivar = True

# Add any paths that contain templates here, relative to this directory.
templates_path = [
    "_templates",
    os.path.join(os.path.dirname(pytorch_sphinx_theme2.__file__), "templates"),
]
# TODO: document these and remove them from here.
# Fixes the duplicated
autosummary_filename_map = {
    "torch.nn.utils.prune.identity": "torch.nn.utils.prune.identity_function",
    "torch.nn.utils.prune.Identity": "torch.nn.utils.prune.Identity_class",
    "torch.optim.adamw.adamw": "torch.optim.adamw.adamw_function",
    "torch.optim.adamw.AdamW": "torch.optim.adamw.AdamW_class",
    "torch.optim.asgd.asgd": "torch.optim.asgd.asgd_function",
    "torch.optim.asgd.ASGD": "torch.optim.asgd.ASGD_class",
    "torch.optim.nadam.nadam": "torch.optim.nadam.nadam_function",
    "torch.optim.nadam.NAdam": "torch.optim.nadam.NAdam_class",
    "torch.optim.radam.radam": "torch.optim.radam.radam_function",
    "torch.optim.radam.RAdam": "torch.optim.radam.RAdam_class",
    "torch.optim.rmsprop.rmsprop": "torch.optim.rmsprop.rmsprop_function",
    "torch.optim.rmsprop.RMSprop": "torch.optim.rmsprop.RMSprop_class",
    "torch.optim.rprop.rprop": "torch.optim.rprop.rprop_function",
    "torch.optim.rprop.Rprop": "torch.optim.rprop.Rprop_class",
    "torch.optim.sgd.sgd": "torch.optim.sgd.sgd_function",
    "torch.optim.sgd.SGD": "torch.optim.sgd.SGD_class",
    "torch.optim.adadelta.adadelta": "torch.optim.adadelta.adadelta_function",
    "torch.optim.adadelta.Adadelta": "torch.optim.adadelta.Adadelta_class",
    "torch.optim.adagrad.adagrad": "torch.optim.adagrad.adagrad_function",
    "torch.optim.adagrad.Adagrad": "torch.optim.adagrad.Adagrad_class",
    "torch.optim.adam.adam": "torch.optim.adam.adam_function",
    "torch.optim.adam.Adam": "torch.optim.adam.Adam_class",
    "torch.optim.adamax.adamax": "torch.optim.adamax.adamax_function",
    "torch.optim.adamax.Adamax": "torch.optim.adamax.Adamax_class",
    "torch.mtia.stream": "torch.mtia.stream_function",
    "torch.mtia.Stream": "torch.mtia.Stream_class",
    "torch.cpu.stream": "torch.cpu.stream_function",
    "torch.cpu.Stream": "torch.cpu.Stream_class",
    "torch.cuda.stream": "torch.cuda.stream_function",
    "torch.cuda.Stream": "torch.cuda.Stream_class",
    "torch.xpu.stream": "torch.xpu.stream_function",
    "torch.xpu.Stream": "torch.xpu.Stream_class",
}

coverage_ignore_functions = [
    "main",  # utility script
    "run",  # utility script
    # TorchVitals (C++ bindings still in torch._C, Python wrappers removed)
    "read_vitals",
    "set_vital",
    "vitals_enabled",
    # torch.hub
    "import_module",
    # torch.jit.unsupported_tensor_ops
    "execWrapper",
    # torch.backends
    "disable_global_flags",
    "flags_frozen",
    # torch.nn.parallel
    "DistributedDataParallelCPU",
    # torch.utils
    "burn_in_info",
    "get_info_and_burn_skeleton",
    "get_storage_info",
    "hierarchical_pickle",
    # torch.amp.autocast_mode
    "autocast_decorator",
    # torch.ao.nn.quantized.dynamic.modules.rnn
    "apply_permutation",
    "pack_weight_bias",
    # torch.ao.nn.quantized.reference.modules.rnn
    "get_quantized_weight",
    # torch.ao.ns.fx.graph_matcher
    "get_matching_subgraph_pairs",
    # torch.ao.ns.fx.graph_passes
    "add_loggers_to_model",
    "create_a_shadows_b",
    # torch.ao.ns.fx.mappings
    "add_op_to_sets_of_related_ops",
    "get_base_name_for_op",
    "get_base_name_to_sets_of_related_ops",
    "get_node_type_to_io_type_map",
    "get_unmatchable_types_map",
    # torch.ao.ns.fx.n_shadows_utils
    "create_add_loggers_graph",
    "create_n_transformed_and_logged_copies_of_subgraph",
    "create_one_transformed_and_logged_copy_of_subgraph",
    "create_results_comparison",
    "create_submodule_from_subgraph",
    "extract_weight_comparison",
    "group_results_by_subgraph",
    "print_n_shadows_summary",
    # torch.ao.ns.fx.pattern_utils
    "end_node_matches_reversed_fusion",
    "get_reversed_fusions",
    "get_type_a_related_to_b",
    # torch.ao.ns.fx.utils
    "get_arg_indices_of_inputs_to_log",
    "get_node_first_input_and_output_type",
    "get_node_input_qparams",
    "get_normalized_nth_input",
    "get_number_of_non_param_args",
    "get_target_type_str",
    "maybe_add_missing_fqns",
    "maybe_dequantize_first_two_tensor_args_and_handle_tuples",
    "op_type_supports_shadowing",
    "rekey_logger_info_on_node_name_of_model",
    "return_first_non_observer_node",
    # torch.ao.ns.fx.weight_utils
    "extract_weight_from_node",
    "get_conv_fun_weight",
    "get_conv_mod_weight",
    "get_linear_fun_weight",
    "get_linear_mod_weight",
    "get_lstm_mod_weights",
    "get_lstm_weight",
    "get_op_to_type_to_weight_extraction_fn",
    "get_qconv_fun_weight",
    "get_qlinear_fun_weight",
    "get_qlstm_weight",
    "mod_0_weight_detach",
    "mod_weight_bias_0",
    "mod_weight_detach",
    # torch.ao.pruning.sparsifier.utils
    "fqn_to_module",
    "get_arg_info_from_tensor_fqn",
    "module_contains_param",
    "module_to_fqn",
    "swap_module",
    # torch.ao.quantization.backend_config.native
    "get_native_backend_config",
    "get_native_backend_config_dict",
    "get_test_only_legacy_native_backend_config",
    "get_test_only_legacy_native_backend_config_dict",
    # torch.ao.quantization.backend_config.qnnpack
    "get_qnnpack_backend_config",
    # torch.ao.quantization.backend_config.tensorrt
    "get_tensorrt_backend_config",
    "get_tensorrt_backend_config_dict",
    "get_x86_backend_config",
    # torch.ao.quantization.fuse_modules
    "fuse_known_modules",
    "fuse_modules_qat",
    # torch.ao.quantization.fuser_method_mappings
    "fuse_conv_bn",
    "fuse_conv_bn_relu",
    "fuse_convtranspose_bn",
    "fuse_linear_bn",
    "get_fuser_method",
    "get_fuser_method_new",
    # torch.ao.quantization.fx.convert
    "convert",
    "convert_custom_module",
    "convert_standalone_module",
    "convert_weighted_module",
    # torch.ao.quantization.fx.fuse
    "fuse",
    # torch.ao.quantization.fx.lower_to_fbgemm
    "lower_to_fbgemm",
    # torch.ao.quantization.fx.lower_to_qnnpack
    "lower_to_qnnpack",
    # torch.ao.quantization.fx.pattern_utils
    "get_default_fusion_patterns",
    "get_default_output_activation_post_process_map",
    "get_default_quant_patterns",
    # torch.ao.quantization.fx.prepare
    "insert_observers_for_model",
    "prepare",
    "propagate_dtypes_for_known_nodes",
    "bfs_trace_with_node_process",
    "find_sequential_partitions",
    "get_equivalent_types",
    "update_equivalent_types_dict",
    # torch.ao.quantization.pt2e.prepare
    "prepare",
    # torch.ao.quantization.pt2e.representation.rewrite
    "reference_representation_rewrite",
    # torch.ao.quantization.pt2e.utils
    "fold_bn_weights_into_conv_node",
    "remove_tensor_overload_for_qdq_ops",
    # torch.ao.quantization.qconfig
    "get_default_qat_qconfig_dict",
    "get_default_qconfig_dict",
    # torch.ao.quantization.quantize
    "get_default_custom_config_dict",
    # torch.ao.quantization.quantize_jit
    "convert_dynamic_jit",
    "convert_jit",
    "fuse_conv_bn_jit",
    "prepare_dynamic_jit",
    "prepare_jit",
    "quantize_dynamic_jit",
    "quantize_jit",
    "script_qconfig",
    "script_qconfig_dict",
    # torch.ao.quantization.quantize_pt2e
    "convert_pt2e",
    "prepare_pt2e",
    "prepare_qat_pt2e",
    # torch.ao.quantization.quantizer.embedding_quantizer
    "get_embedding_operators_config",
    # torch.ao.quantization.quantizer.xnnpack_quantizer_utils
    "get_bias_qspec",
    "get_input_act_qspec",
    "get_output_act_qspec",
    "get_weight_qspec",
    "propagate_annotation",
    "register_annotator",
    "activation_dtype",
    "check_node",
    "has_no_children_ignoring_parametrizations",
    "is_per_channel",
    "is_per_tensor",
    "op_is_int8_dynamically_quantized",
    "to_underlying_dtype",
    "weight_dtype",
    "weight_is_quantized",
    "weight_is_statically_quantized",
    # torch.backends.cudnn.rnn
    "get_cudnn_mode",
    "init_dropout_state",
    # torch.cuda.amp.autocast_mode
    "custom_bwd",
    "custom_fwd",
    # torch.cuda.amp.common
    "amp_definitely_not_available",
    # torch.cuda.nccl
    "all_gather",
    "all_reduce",
    "broadcast",
    "init_rank",
    "reduce",
    "reduce_scatter",
    "unique_id",
    # torch.cuda.profiler
    "init",
    # torch.distributed.checkpoint.default_planner
    "create_default_local_load_plan",
    # torch.distributed.checkpoint.utils
    "find_state_dict_object",
    "find_tensor_shard",
    "isend",
    "monitored_barrier",
    "new_group",
    "new_subgroups",
    "new_subgroups_by_enumeration",
    "recv",
    "reduce",
    "reduce_scatter",
    "reduce_scatter_tensor",
    "scatter",
    "scatter_object_list",
    "send",
    "supports_complex",
    # torch.distributed.elastic.events.handlers
    "get_logging_handler",
    # torch.distributed.elastic.metrics.api
    "configure",
    "getStream",
    "get_elapsed_time_ms",
    "prof",
    "profile",
    "publish_metric",
    "put_metric",
    # torch.distributed.elastic.multiprocessing.api
    "get_std_cm",
    "to_map",
    # torch.distributed.elastic.multiprocessing.errors.handlers
    "get_error_handler",
    # torch.distributed.elastic.multiprocessing.redirects
    "get_libc",
    "redirect",
    # torch.distributed.elastic.multiprocessing.tail_log
    "tail_logfile",
    # torch.distributed.elastic.rendezvous.dynamic_rendezvous
    "get_method_name",
    # torch.distributed.elastic.rendezvous.etcd_rendezvous
    "create_rdzv_handler",
    # torch.distributed.elastic.rendezvous.etcd_server
    "find_free_port",
    "stop_etcd",
    # torch.distributed.elastic.rendezvous.etcd_store
    "cas_delay",
    # torch.distributed.elastic.rendezvous.static_tcp_rendezvous
    "create_rdzv_handler",
    # torch.distributed.elastic.rendezvous.utils
    "parse_rendezvous_endpoint",
    # torch.distributed.elastic.timer.api
    "configure",
    "expires",
    # torch.distributed.elastic.utils.api
    "get_socket_with_port",
    # torch.distributed.elastic.utils.distributed
    "create_c10d_store",
    "get_socket_with_port",
    # torch.distributed.elastic.utils.store
    "barrier",
    "get_all",
    "synchronize",
    "store_timeout",
    # torch.distributed.fsdp.wrap
    # torch.distributed.nn.functional
    "all_to_all",
    "all_to_all_single",
    # torch.distributed.nn.jit.instantiator
    "get_arg_return_types_from_interface",
    "instantiate_non_scriptable_remote_module_template",
    "instantiate_scriptable_remote_module_template",
    # torch.distributed.nn.jit.templates.remote_module_template
    "get_remote_module_template",
    # torch.distributed.optim.utils
    "as_functional_optim",
    # torch.distributed.rendezvous
    "rendezvous",
    # torch.distributed.rpc.api
    "method_factory",
    "new_method",
    "remote",
    "rpc_async",
    "rpc_sync",
    "shutdown",
    # torch.distributed.rpc.backend_registry
    "construct_rpc_backend_options",
    "init_backend",
    # torch.distributed.rpc.internal
    "deserialize",
    "serialize",
    # torch.distributed.tensor.parallel.style
    "make_sharded_output_tensor",
    # torch.fx.passes.dialect.common.cse_pass
    "get_CSE_banned_ops",
    # torch.fx.passes.graph_manipulation
    "get_tensor_meta",
    # torch.fx.passes.split_module
    "split_module",
    "split_module_simple",
    # torch.fx.passes.split_utils
    "getattr_recursive",
    # torch.fx.passes.tools_common
    "get_acc_ops_name",
    "get_node_target",
    "legalize_graph",
    # torch.fx.passes.utils.fuser_utils
    "fuse_by_partitions",
    "insert_subgm",
    # torch.fx.proxy
    "assert_fn",
    # torch.fx.traceback
    "format_stack",
    "get_current_meta",
    "has_preserved_node_meta",
    "preserve_node_meta",
    "reset_grad_fn_seq_nr",
    "set_current_meta",
    "set_current_replay_node",
    "set_grad_fn_seq_nr",
    "set_stack_trace",
    "get_current_replay_node",
    # torch.fx.experimental.proxy_tensor
    "wrap_key",
    "wrapper_and_args_for_make_fx",
    # torch.fx.experimental.sym_node
    "wrap_node",
    # torch.jit.annotations
    "ann_to_type",
    "check_fn",
    "get_enum_value_type",
    "get_param_names",
    "get_signature",
    "get_type_line",
    "is_function_or_method",
    "is_tensor",
    "is_vararg",
    "parse_type_line",
    "split_type_line",
    "try_ann_to_type",
    "try_real_annotations",
    # torch.jit.frontend
    "build_class_def",
    "build_def",
    "build_ignore_context_manager",
    "build_param",
    "build_param_list",
    "build_stmts",
    "build_withitems",
    "find_before",
    "get_class_assigns",
    "get_class_properties",
    "get_default_args",
    "get_default_args_for_class",
    "get_jit_class_def",
    "get_jit_def",
    "is_reserved_name",
    "is_torch_jit_ignore_context_manager",
    # torch.jit.generate_bytecode
    "format_bytecode",
    "generate_upgraders_bytecode",
    # torch.jit.quantized
    "apply_permutation",
    "quantize_linear_modules",
    "quantize_rnn_cell_modules",
    "quantize_rnn_modules",
    # torch.masked.maskedtensor.creation
    "as_masked_tensor",
    "masked_tensor",
    # torch.multiprocessing.pool
    "clean_worker",
    # torch.multiprocessing.reductions
    "fd_id",
    "init_reductions",
    "rebuild_cuda_tensor",
    "rebuild_meta_tensor",
    "rebuild_event",
    "rebuild_nested_tensor",
    "rebuild_sparse_coo_tensor",
    "rebuild_sparse_compressed_tensor",
    "rebuild_storage_empty",
    "rebuild_storage_fd",
    "rebuild_storage_filename",
    "rebuild_tensor",
    "rebuild_typed_storage",
    "rebuild_typed_storage_child",
    "reduce_event",
    "reduce_storage",
    "reduce_tensor",
    "reduce_typed_storage",
    "reduce_typed_storage_child",
    "storage_from_cache",
    # torch.multiprocessing.spawn
    "start_processes",
    # torch.nn.functional
    "adaptive_max_pool1d_with_indices",  # documented as adaptive_max_pool1d
    "adaptive_max_pool2d_with_indices",  # documented as adaptive_max_pool2d
    "adaptive_max_pool3d_with_indices",  # documented as adaptive_max_pool3d
    "assert_int_or_pair",  # looks unintentionally public
    "fractional_max_pool2d_with_indices",  # documented as fractional_max_pool2d
    "fractional_max_pool3d_with_indices",  # documented as fractional_max_pool3d
    "max_pool1d_with_indices",  # documented as max_pool1d
    "max_pool2d_with_indices",  # documented as max_pool2d
    "max_pool3d_with_indices",  # documented as max_pool3d
    "multi_head_attention_forward",
    # torch.nn.grad
    "conv1d_input",  # legacy helper for gradient computation
    "conv1d_weight",  # legacy helper for gradient computation
    "conv2d_input",  # legacy helper for gradient computation
    "conv2d_weight",  # legacy helper for gradient computation
    "conv3d_input",  # legacy helper for gradient computation
    "conv3d_weight",  # legacy helper for gradient computation
    # torch.nn.init
    "constant",  # deprecated
    "dirac",  # deprecated
    "eye",  # deprecated
    "kaiming_normal",  # deprecated
    "kaiming_uniform",  # deprecated
    "normal",  # deprecated
    "orthogonal",  # deprecated
    "sparse",  # deprecated
    "uniform",  # deprecated
    "xavier_normal",  # deprecated
    "xavier_uniform",  # deprecated
    # torch.nn.modules.rnn
    "apply_permutation",  # deprecated
    # torch.nn.parallel.comm
    "broadcast",
    "broadcast_coalesced",
    "gather",
    "reduce_add",
    "reduce_add_coalesced",
    "scatter",
    # torch.nn.parallel.data_parallel
    "data_parallel",
    # torch.nn.parallel.parallel_apply
    "get_a_var",
    "parallel_apply",
    # torch.nn.parallel.replicate
    "replicate",
    # torch.nn.parallel.scatter_gather
    "gather",
    "is_namedtuple",
    "scatter",
    "scatter_kwargs",
    # torch.nn.utils.rnn
    "bind",  # looks unintentionally public
    # torch.onnx.operators
    "reshape_from_tensor_shape",
    "shape_as_tensor",
    # torch.onnx.symbolic_caffe2
    "add",
    "avg_pool2d",
    "cat",
    "conv2d",
    "conv2d_relu",
    "conv_prepack",
    "dequantize",
    "linear",
    "linear_prepack",
    "max_pool2d",
    "nchw2nhwc",
    "nhwc2nchw",
    "quantize_per_tensor",
    "register_quantized_ops",
    "relu",
    "reshape",
    "sigmoid",
    "slice",
    "upsample_nearest2d",
    # torch.onnx.symbolic_helper
    "args_have_same_dtype",
    "check_training_mode",
    "dequantize_helper",
    "is_complex_value",
    "quantize_helper",
    "quantized_args",
    "requantize_bias_helper",
    # torch.onnx.symbolic_opset10
    "dequantize",
    "div",
    "embedding_bag",
    "fake_quantize_per_tensor_affine",
    "flip",
    "fmod",
    "isfinite",
    "isinf",
    "nan_to_num",
    "quantize_per_tensor",
    "quantized_add",
    "quantized_add_relu",
    "quantized_cat",
    "quantized_conv1d",
    "quantized_conv1d_relu",
    "quantized_conv2d",
    "quantized_conv2d_relu",
    "quantized_conv3d",
    "quantized_conv3d_relu",
    "quantized_conv_transpose1d",
    "quantized_conv_transpose2d",
    "quantized_conv_transpose3d",
    "quantized_group_norm",
    "quantized_hardswish",
    "quantized_instance_norm",
    "quantized_layer_norm",
    "quantized_leaky_relu",
    "quantized_linear",
    "quantized_linear_relu",
    "quantized_mul",
    "quantized_sigmoid",
    "slice",
    "sort",
    "topk",
    # torch.onnx.symbolic_opset11
    "Delete",
    "add",
    "append",
    "arange",
    "argsort",
    "atleast_1d",
    "atleast_2d",
    "atleast_3d",
    "cat",
    "chunk",
    "clamp",
    "clamp_max",
    "clamp_min",
    "constant_pad_nd",
    "cumsum",
    "embedding_bag",
    "embedding_renorm",
    "flatten",
    "gather",
    "hardtanh",
    "hstack",
    "im2col",
    "index",
    "index_copy",
    "index_fill",
    "index_put",
    "insert",
    "linalg_det",
    "linalg_vector_norm",
    "logdet",
    "masked_scatter",
    "masked_select",
    "mm",
    "narrow",
    "normal",
    "pad",
    "pixel_shuffle",
    "pop",
    "prim_constant_chunk",
    "reflection_pad",
    "relu6",
    "remainder",
    "replication_pad",
    "round",
    "scatter",
    "select",
    "size",
    "sort",
    "split",
    "split_with_sizes",
    "squeeze",
    "stack",
    "topk",
    "unbind",
    "unique_dim",
    "unsqueeze",
    "vstack",
    # torch.onnx.symbolic_opset12
    "argmax",
    "argmin",
    "binary_cross_entropy_with_logits",
    "celu",
    "cross_entropy_loss",
    "dropout",
    "einsum",
    "ge",
    "le",
    "native_dropout",
    "nll_loss",
    "nll_loss2d",
    "nll_loss_nd",
    "outer",
    "pow",
    "tensordot",
    "unfold",
    # torch.onnx.symbolic_opset13
    "diagonal",
    "fake_quantize_per_channel_affine",
    "fake_quantize_per_tensor_affine",
    "frobenius_norm",
    "log_softmax",
    "nonzero_numpy",
    "quantized_conv1d",
    "quantized_conv1d_relu",
    "quantized_conv2d",
    "quantized_conv2d_relu",
    "quantized_conv3d",
    "quantized_conv3d_relu",
    "quantized_conv_transpose1d",
    "quantized_conv_transpose2d",
    "quantized_conv_transpose3d",
    "quantized_linear",
    "quantized_linear_relu",
    "repeat_interleave",
    "softmax",
    "split",
    "split_with_sizes",
    "tensor_split",
    "tile",
    "unbind",
    "unflatten",
    "unsafe_chunk",
    "unsafe_split",
    "unsafe_split_with_sizes",
    "where",
    # torch.onnx.symbolic_opset14
    "batch_norm",
    "hardswish",
    "quantized_hardswish",
    "reshape",
    "scaled_dot_product_attention",
    "tril",
    "triu",
    # torch.onnx.symbolic_opset15
    "aten__is_",
    "aten__isnot_",
    "bernoulli",
    "prim_unchecked_cast",
    # torch.onnx.symbolic_opset16
    "grid_sampler",
    "scatter_add",
    "scatter_reduce",
    # torch.onnx.symbolic_opset17
    "layer_norm",
    "stft",
    # torch.onnx.symbolic_opset18
    "col2im",
    # torch.onnx.symbolic_opset7
    "max",
    "min",
    "dim",
    "export",
    "check_export_model_diff",
    "verify",
    "verify_aten_graph",
    # torch.overrides
    "enable_reentrant_dispatch",
    # torch.return_types
    "pytree_register_structseq",
    # torch.serialization
    "load",
    "location_tag",
    "mkdtemp",
    "normalize_storage_type",
    "save",
    "storage_to_tensor_type",
    "validate_cuda_device",
    "validate_hpu_device",
    # torch.signal.windows.windows
    "bartlett",
    "blackman",
    "cosine",
    "exponential",
    "gaussian",
    "general_cosine",
    "general_hamming",
    "hamming",
    "hann",
    "kaiser",
    "nuttall",
    # torch.utils.benchmark.examples.op_benchmark
    # torch.utils.benchmark.op_fuzzers.spectral
    "power_range",
    # torch.utils.benchmark.utils.common
    "ordered_unique",
    "set_torch_threads",
    "unit_to_english",
    # torch.utils.benchmark.utils.compare
    "optional_min",
    # torch.utils.benchmark.utils.compile
    "bench_loop",
    # torch.utils.benchmark.utils.cpp_jit
    "compile_callgrind_template",
    "compile_timeit_template",
    "get_compat_bindings",
    # torch.utils.benchmark.utils.fuzzer
    "dtype_size",
    # torch.utils.benchmark.utils.timer
    "timer",
    # torch.utils.benchmark.utils.valgrind_wrapper.timer_interface
    "wrapper_singleton",
    # torch.utils.bundled_inputs
    "augment_many_model_functions_with_bundled_inputs",
    "augment_model_with_bundled_inputs",
    "bundle_inputs",
    "bundle_large_tensor",
    "bundle_randn",
    # torch.utils.checkpoint
    "check_backward_validity",
    "detach_variable",
    "get_device_states",
    "noop_context_fn",
    # torch.utils.cpp_extension
    "check_compiler_is_gcc",
    "check_compiler_ok_for_platform",
    "get_cxx_compiler",
    "get_default_build_root",
    "library_paths",
    "remove_extension_h_precompiler_headers",
    # torch.utils.data.backward_compatibility
    "worker_init_fn",
    # torch.utils.data.datapipes.dataframe.dataframe_wrapper
    "concat",
    "create_dataframe",
    "get_columns",
    "get_df_wrapper",
    "get_item",
    "get_len",
    "is_column",
    "is_dataframe",
    "iterate",
    "set_df_wrapper",
    # torch.utils.data.datapipes.dataframe.dataframes
    "disable_capture",
    "get_val",
    # torch.utils.data.datapipes.gen_pyi
    "extract_class_name",
    "extract_method_name",
    "find_file_paths",
    "gen_from_template",
    "get_method_definitions",
    "materialize_lines",
    "parse_datapipe_file",
    "parse_datapipe_files",
    "process_signature",
    "split_outside_bracket",
    # torch.utils.data.datapipes.map.callable
    "default_fn",
    # torch.utils.data.datapipes.utils.common
    "get_file_binaries_from_pathnames",
    "get_file_pathnames_from_root",
    "match_masks",
    "validate_pathname_binary_tuple",
    # torch.utils.data.datapipes.utils.decoder
    "audiohandler",
    "extension_extract_fn",
    "imagehandler",
    "mathandler",
    "videohandler",
    # torch.utils.data.graph_settings
    "apply_random_seed",
    "apply_sharding",
    "apply_shuffle_seed",
    "apply_shuffle_settings",
    "get_all_graph_pipes",
    # torch.utils.hooks
    "warn_if_has_hooks",
    # torch.utils.jit.log_extract
    "extract_ir",
    "load_graph_and_inputs",
    "make_tensor_from_type",
    "no_fuser",
    "time_cpu",
    "time_cuda",
    # torch.utils.mkldnn
    "to_mkldnn",
    # torch.utils.mobile_optimizer
    # torch.utils.tensorboard.summary
    "audio",
    "compute_curve",
    "custom_scalars",
    "draw_boxes",
    "half_to_int",
    "histogram",
    "histogram_raw",
    "hparams",
    "image",
    "image_boxes",
    "int_to_half",
    "make_histogram",
    "make_image",
    "make_video",
    "mesh",
    "pr_curve",
    "pr_curve_raw",
    "scalar",
    "tensor_proto",
    "text",
    "video",
]

coverage_ignore_classes = [
    # torch
    "FatalError",
    "QUInt2x4Storage",
    "Size",
    "Storage",
    "Stream",
    "Tensor",
    "finfo",
    "iinfo",
    "qscheme",
    "AggregationType",
    "AliasDb",
    "AnyType",
    "Argument",
    "ArgumentSpec",
    "AwaitType",
    "BenchmarkConfig",
    "BenchmarkExecutionStats",
    "Block",
    "BoolType",
    "BufferDict",
    "CallStack",
    "Capsule",
    "ClassType",
    "Code",
    "CompleteArgumentSpec",
    "ComplexType",
    "ConcreteModuleType",
    "ConcreteModuleTypeBuilder",
    "DeepCopyMemoTable",
    "DeserializationStorageContext",
    "DeviceObjType",
    "DictType",
    "DispatchKey",
    "DispatchKeySet",
    "EnumType",
    "ExcludeDispatchKeyGuard",
    "ExecutionPlan",
    "FileCheck",
    "FloatType",
    "FunctionSchema",
    "Gradient",
    "Graph",
    "GraphExecutorState",
    "IODescriptor",
    "InferredType",
    "IntType",
    "InterfaceType",
    "ListType",
    "LockingLogger",
    "MobileOptimizerType",
    "ModuleDict",
    "Node",
    "NoneType",
    "NoopLogger",
    "NumberType",
    "OperatorInfo",
    "OptionalType",
    "ParameterDict",
    "PyObjectType",
    "PyTorchFileReader",
    "PyTorchFileWriter",
    "RRefType",
    "ScriptClass",
    "ScriptClassFunction",
    "ScriptDict",
    "ScriptDictIterator",
    "ScriptDictKeyIterator",
    "ScriptList",
    "ScriptListIterator",
    "ScriptMethod",
    "ScriptModule",
    "ScriptModuleSerializer",
    "ScriptObject",
    "ScriptObjectProperty",
    "SerializationStorageContext",
    "StaticModule",
    "StringType",
    "SymIntType",
    "SymBoolType",
    "ThroughputBenchmark",
    "TracingState",
    "TupleType",
    "Type",
    "UnionType",
    "Use",
    "Value",
    # torch.cuda
    "BFloat16Storage",
    "BFloat16Tensor",
    "BoolStorage",
    "BoolTensor",
    "ByteStorage",
    "ByteTensor",
    "CharStorage",
    "CharTensor",
    "ComplexDoubleStorage",
    "ComplexFloatStorage",
    "CudaError",
    "DeferredCudaCallError",
    "DoubleStorage",
    "DoubleTensor",
    "FloatStorage",
    "FloatTensor",
    "HalfStorage",
    "HalfTensor",
    "IntStorage",
    "IntTensor",
    "LongStorage",
    "LongTensor",
    "ShortStorage",
    "ShortTensor",
    "cudaStatus",
    # torch.cuda._sanitizer
    "Access",
    "AccessType",
    "Await",
    "CUDASanitizer",
    "CUDASanitizerDispatchMode",
    "CUDASanitizerErrors",
    "EventHandler",
    "SynchronizationError",
    "UnsynchronizedAccessError",
    # torch.distributed.elastic.multiprocessing.errors
    "ChildFailedError",
    "ProcessFailure",
    # torch.distributions.constraints
    "cat",
    "greater_than",
    "greater_than_eq",
    "half_open_interval",
    "independent",
    "integer_interval",
    "interval",
    "less_than",
    "multinomial",
    "stack",
    # torch.distributions.transforms
    "AffineTransform",
    "CatTransform",
    "ComposeTransform",
    "CorrCholeskyTransform",
    "CumulativeDistributionTransform",
    "ExpTransform",
    "IndependentTransform",
    "PowerTransform",
    "ReshapeTransform",
    "SigmoidTransform",
    "SoftmaxTransform",
    "SoftplusTransform",
    "StackTransform",
    "StickBreakingTransform",
    "TanhTransform",
    "Transform",
    # torch.jit
    "CompilationUnit",
    "Error",
    "Future",
    "ScriptFunction",
    # torch.onnx
    "CheckerError",
    "ExportTypes",
    # torch.backends
    "ContextProp",
    "PropModule",
    # torch.backends.cuda
    "cuBLASModule",
    "cuFFTPlanCache",
    "cuFFTPlanCacheAttrContextProp",
    "cuFFTPlanCacheManager",
    # torch.distributed.algorithms.ddp_comm_hooks
    "DDPCommHookType",
    # torch.jit.mobile
    "LiteScriptModule",
    # torch.ao.nn.quantized.modules
    "DeQuantize",
    "Quantize",
    # torch.utils.backcompat
    "Warning",
    # torch.ao.nn.quantized.modules.activation
    "ELU",
    "Hardswish",
    "LeakyReLU",
    "MultiheadAttention",
    "PReLU",
    "ReLU6",
    "Sigmoid",
    "Softmax",
    # torch.ao.nn.quantized.modules.batchnorm
    "BatchNorm2d",
    "BatchNorm3d",
    # torch.ao.nn.quantized.modules.conv
    "Conv1d",
    "Conv2d",
    "Conv3d",
    "ConvTranspose1d",
    "ConvTranspose2d",
    "ConvTranspose3d",
    # torch.ao.nn.quantized.modules.dropout
    "Dropout",
    # torch.ao.nn.quantized.modules.embedding_ops
    "Embedding",
    "EmbeddingBag",
    "EmbeddingPackedParams",
    # torch.ao.nn.quantized.modules.functional_modules
    "FXFloatFunctional",
    "FloatFunctional",
    "QFunctional",
    # torch.ao.nn.quantized.modules.linear
    "Linear",
    "LinearPackedParams",
    # torch.ao.nn.quantized.modules.normalization
    "GroupNorm",
    "InstanceNorm1d",
    "InstanceNorm2d",
    "InstanceNorm3d",
    "LayerNorm",
    # torch.ao.nn.quantized.modules.rnn
    "LSTM",
    # torch.ao.nn.quantized.modules.utils
    "WeightedQuantizedModule",
    # torch.ao.nn.quantized.reference.modules.conv
    "Conv1d",
    "Conv2d",
    "Conv3d",
    "ConvTranspose1d",
    "ConvTranspose2d",
    "ConvTranspose3d",
    # torch.ao.nn.quantized.reference.modules.linear
    "Linear",
    # torch.ao.nn.quantized.reference.modules.rnn
    "GRU",
    "GRUCell",
    "LSTM",
    "LSTMCell",
    "RNNBase",
    "RNNCell",
    "RNNCellBase",
    # torch.ao.nn.quantized.reference.modules.sparse
    "Embedding",
    "EmbeddingBag",
    # torch.ao.nn.quantized.reference.modules.utils
    "ReferenceQuantizedModule",
    # torch.ao.nn.sparse.quantized.dynamic.linear
    "Linear",
    # torch.ao.nn.sparse.quantized.linear
    "Linear",
    "LinearPackedParams",
    # torch.ao.nn.sparse.quantized.utils
    "LinearBlockSparsePattern",
    # torch.ao.ns.fx.graph_matcher
    "SubgraphTypeRelationship",
    # torch.ao.ns.fx.n_shadows_utils
    "OutputProp",
    # torch.ao.ns.fx.ns_types
    "NSSingleResultValuesType",
    "NSSubgraph",
    # torch.ao.ns.fx.qconfig_multi_mapping
    "QConfigMultiMapping",
    # torch.ao.pruning.scheduler.base_scheduler
    "BaseScheduler",
    # torch.ao.pruning.scheduler.cubic_scheduler
    "CubicSL",
    # torch.ao.pruning.scheduler.lambda_scheduler
    "LambdaSL",
    # torch.ao.pruning.sparsifier.base_sparsifier
    "BaseSparsifier",
    # torch.ao.pruning.sparsifier.nearly_diagonal_sparsifier
    "NearlyDiagonalSparsifier",
    # torch.ao.pruning.sparsifier.utils
    "FakeSparsity",
    # torch.ao.pruning.sparsifier.weight_norm_sparsifier
    "WeightNormSparsifier",
    # torch.ao.quantization.backend_config.backend_config
    "BackendConfig",
    "BackendPatternConfig",
    "DTypeConfig",
    # torch.ao.quantization.fake_quantize
    "FakeQuantize",
    "FakeQuantizeBase",
    "FixedQParamsFakeQuantize",
    "FusedMovingAvgObsFakeQuantize",
    # torch.ao.quantization.fx.fuse_handler
    "DefaultFuseHandler",
    "FuseHandler",
    # torch.ao.quantization.fx.graph_module
    "FusedGraphModule",
    "ObservedGraphModule",
    "ObservedStandaloneGraphModule",
    # torch.ao.quantization.fx.quantize_handler
    "BatchNormQuantizeHandler",
    "BinaryOpQuantizeHandler",
    "CatQuantizeHandler",
    "ConvReluQuantizeHandler",
    "CopyNodeQuantizeHandler",
    "CustomModuleQuantizeHandler",
    "DefaultNodeQuantizeHandler",
    "EmbeddingQuantizeHandler",
    "FixedQParamsOpQuantizeHandler",
    "GeneralTensorShapeOpQuantizeHandler",
    "LinearReLUQuantizeHandler",
    "RNNDynamicQuantizeHandler",
    "StandaloneModuleQuantizeHandler",
    # torch.ao.quantization.fx.tracer
    "QuantizationTracer",
    "ScopeContextManager",
    # torch.ao.quantization.fx.utils
    "ObservedGraphModuleAttrs",
    # torch.ao.quantization.observer
    "FixedQParamsObserver",
    "HistogramObserver",
    "MinMaxObserver",
    "MovingAverageMinMaxObserver",
    "MovingAveragePerChannelMinMaxObserver",
    "NoopObserver",
    "ObserverBase",
    "PerChannelMinMaxObserver",
    "PlaceholderObserver",
    "RecordingObserver",
    "ReuseInputObserver",
    "UniformQuantizationObserverBase",
    "default_debug_observer",
    "default_placeholder_observer",
    "default_reuse_input_observer",
    # torch.ao.quantization.qconfig
    "QConfigDynamic",
    # torch.ao.quantization.quant_type
    "QuantType",
    # torch.ao.quantization.stubs
    "DeQuantStub",
    "QuantStub",
    "QuantWrapper",
    # torch.ao.quantization.utils
    "MatchAllNode",
    # torch.backends.cudnn.rnn
    "Unserializable",
    # torch.amp.grad_scaler
    "GradScaler",
    "OptState",
    # torch.distributed.algorithms.ddp_comm_hooks.post_localSGD_hook
    "PostLocalSGDState",
    # torch.distributed.algorithms.ddp_comm_hooks.powerSGD_hook
    "PowerSGDState",
    # torch.distributed.algorithms.join
    "Join",
    "JoinHook",
    "Joinable",
    # torch.distributed.algorithms.model_averaging.averagers
    "ModelAverager",
    "PeriodicModelAverager",
    # torch.distributed.algorithms.model_averaging.hierarchical_model_averager
    "HierarchicalModelAverager",
    # torch.distributed.argparse_util
    "check_env",
    "env",
    # torch.distributed.checkpoint.api
    "CheckpointException",
    # torch.distributed.checkpoint.default_planner
    "DefaultLoadPlanner",
    "DefaultSavePlanner",
    # torch.distributed.checkpoint.filesystem
    "FileSystemReader",
    "FileSystemWriter",
    # torch.distributed.checkpoint.hf_storage
    "HuggingFaceStorageReader",
    "HuggingFaceStorageWriter",
    # torch.distributed.checkpoint.quantized_hf_storage
    "QuantizedHuggingFaceStorageReader",
    # torch.distributed.checkpoint.metadata
    "BytesStorageMetadata",
    "ChunkStorageMetadata",
    "Metadata",
    "MetadataIndex",
    # torch.distributed.checkpoint.planner
    "LoadItemType",
    "LoadPlanner",
    "SavePlanner",
    "WriteItemType",
    # torch.distributed.checkpoint.state_dict
    "DistributedStateDictOptions",
    # torch.distributed.checkpoint.storage
    "WriteResult",
    # torch.distributed.collective_utils
    "SyncPayload",
    # torch.distributed.distributed_c10d
    "AllToAllOptions",
    "AllreduceCoalescedOptions",
    "AllreduceOptions",
    "Backend",
    "BackendConfig",
    "BarrierOptions",
    "BroadcastOptions",
    "DebugLevel",
    "GatherOptions",
    "GroupMember",
    "ProcessGroup",
    "ProcessGroupGloo",
    "ProcessGroupNCCL",
    "ReduceOptions",
    "ReduceScatterOptions",
    "ScatterOptions",
    "Work",
    "group",
    # torch.distributed.elastic.agent.server.api
    "ElasticAgent",
    "RunResult",
    "SimpleElasticAgent",
    "WorkerSpec",
    # torch.distributed.elastic.events.api
    "Event",
    "RdzvEvent",
    # torch.distributed.elastic.metrics.api
    "ConsoleMetricHandler",
    "MetricData",
    "MetricHandler",
    "MetricStream",
    "MetricsConfig",
    "NullMetricHandler",
    # torch.distributed.elastic.multiprocessing.api
    "MultiprocessContext",
    "PContext",
    "RunProcsResult",
    "SignalException",
    "Std",
    "SubprocessContext",
    "SubprocessHandler",
    # torch.distributed.elastic.multiprocessing.tail_log
    "TailLog",
    # torch.distributed.elastic.rendezvous.api
    "RendezvousHandler",
    "RendezvousHandlerRegistry",
    "RendezvousParameters",
    # torch.distributed.elastic.rendezvous.dynamic_rendezvous
    "DynamicRendezvousHandler",
    "RendezvousSettings",
    # torch.distributed.elastic.rendezvous.etcd_rendezvous
    "EtcdRendezvous",
    "EtcdRendezvousHandler",
    "EtcdRendezvousRetryImmediately",
    "EtcdRendezvousRetryableFailure",
    # torch.distributed.elastic.rendezvous.etcd_server
    "EtcdServer",
    # torch.distributed.elastic.rendezvous.static_tcp_rendezvous
    "StaticTCPRendezvous",
    # torch.distributed.elastic.timer.api
    "RequestQueue",
    "TimerClient",
    "TimerServer",
    # torch.distributed.elastic.timer.file_based_local_timer
    "FileTimerClient",
    "FileTimerRequest",
    "FileTimerServer",
    # torch.distributed.elastic.timer.local_timer
    "LocalTimerClient",
    "LocalTimerServer",
    "MultiprocessingRequestQueue",
    # torch.distributed.elastic.utils.api
    "macros",
    # torch.distributed.elastic.utils.data.cycling_iterator
    "CyclingIterator",
    # torch.distributed.elastic.utils.data.elastic_distributed_sampler
    "ElasticDistributedSampler",
    # torch.distributed.fsdp.api
    "StateDictType",
    # torch.distributed.fsdp.fully_sharded_data_parallel
    "FullyShardedDataParallel",
    "OptimStateKeyType",
    # torch.distributed.fsdp.sharded_grad_scaler
    "ShardedGradScaler",
    # torch.distributed.fsdp.wrap
    "CustomPolicy",
    "ModuleWrapPolicy",
    # torch.distributed.launcher.api
    "LaunchConfig",
    "elastic_launch",
    # torch.distributed.optim.optimizer
    "DistributedOptimizer",
    # torch.distributed.optim.post_localSGD_optimizer
    "PostLocalSGDOptimizer",
    # torch.distributed.optim.zero_redundancy_optimizer
    "ZeroRedundancyOptimizer",
    # torch.distributed.rpc.api
    "AllGatherStates",
    "RRef",
    # torch.distributed.rpc.backend_registry
    "BackendValue",
    # torch.distributed.rpc.internal
    "PythonUDF",
    "RPCExecMode",
    "RemoteException",
    # torch.distributed.rpc.rref_proxy
    "RRefProxy",
    # torch.distributed.tensor.parallel.fsdp
    "DTensorExtensions",
    # torch.distributed.tensor.parallel.style
    "ParallelStyle",
    # torch.distributions.logistic_normal
    "LogisticNormal",
    # torch.distributions.one_hot_categorical
    "OneHotCategoricalStraightThrough",
    # torch.distributions.relaxed_categorical
    "ExpRelaxedCategorical",
    # torch.distributions.utils
    "lazy_property",
    # torch.export.unflatten
    "UnflattenedModule",
    # torch.export.exported_program
    "ConstantArgument",
    "ExportedProgram",
    # torch.fx.experimental.accelerator_partitioner
    "DAG",
    "DAGNode",
    "PartitionResult",
    "Partitioner",
    # torch.fx.experimental.const_fold
    "FoldedGraphModule",
    # torch.fx.experimental.graph_gradual_typechecker
    "Refine",
    # torch.fx.experimental.meta_tracer
    "MetaAttribute",
    "MetaDeviceAttribute",
    "MetaProxy",
    "MetaTracer",
    # torch.fx.experimental.migrate_gradual_types.constraint
    "ApplyBroadcasting",
    "BVar",
    "BinConstraintD",
    "BinConstraintT",
    "BinaryConstraint",
    "CalcConv",
    "CalcMaxPool",
    "CalcProduct",
    "CanReshape",
    "Conj",
    "Constraint",
    "DGreatestUpperBound",
    "DVar",
    "Disj",
    "F",
    "GetItem",
    "GetItemTensor",
    "IndexSelect",
    "Prod",
    "T",
    "TGreatestUpperBound",
    "TVar",
    "Transpose",
    # torch.fx.experimental.migrate_gradual_types.constraint_generator
    "ConstraintGenerator",
    # torch.fx.experimental.normalize
    "NormalizeArgs",
    "NormalizeOperators",
    # torch.fx.experimental.optimization
    "MklSubgraph",
    "UnionFind",
    # torch.fx.experimental.partitioner_utils
    "Device",
    "Partition",
    "PartitionLatency",
    "PartitionMode",
    "PartitionerConfig",
    # torch.fx.experimental.proxy_tensor
    "DecompositionInterpreter",
    "PreDispatchTorchFunctionMode",
    "ProxySymDispatchMode",
    "ProxyTorchDispatchMode",
    "PythonKeyTracer",
    # torch.fx.experimental.recording
    "FakeTensorMeta",
    "NotEqualError",
    "ShapeEnvEvent",
    # torch.fx.experimental.refinement_types
    "Equality",
    # torch.fx.experimental.rewriter
    "AST_Rewriter",
    "RewritingTracer",
    # torch.fx.experimental.schema_type_annotation
    "AnnotateTypesWithSchema",
    # torch.fx.experimental.sym_node
    "SymNode",
    # torch.fx.experimental.symbolic_shapes
    "Constraint",
    "ConstraintViolationError",
    "DynamicDimConstraintPrinter",
    "GuardOnDataDependentSymNode",
    "PendingUnbackedSymbolNotFound",
    "LoggingShapeGuardPrinter",
    "SymExprPrinter",
    "RelaxedUnspecConstraint",
    "RuntimeAssert",
    "ShapeGuardPrinter",
    "ShapeGuardPythonPrinter",
    "SymDispatchMode",
    "SymbolicContext",
    # torch.fx.experimental.unification.match
    "Dispatcher",
    "VarDispatcher",
    # torch.fx.experimental.unification.multipledispatch.conflict
    "AmbiguityWarning",
    # torch.fx.experimental.unification.multipledispatch.dispatcher
    "Dispatcher",
    "MDNotImplementedError",
    "MethodDispatcher",
    # torch.fx.experimental.unification.multipledispatch.variadic
    "Variadic",
    "VariadicSignatureMeta",
    "VariadicSignatureType",
    # torch.fx.experimental.unification.variable
    "Var",
    # torch.fx.experimental.validator
    "BisectValidationException",
    "PopulateValidator",
    "SympyToZ3",
    "ValidationException",
    # torch.fx.graph
    "PythonCode",
    # torch.fx.immutable_collections
    "immutable_dict",
    "immutable_list",
    # torch.fx.interpreter
    "Interpreter",
    # torch.fx.operator_schemas
    "ArgsKwargsPair",
    # torch.fx.passes.backends.cudagraphs
    "CudaGraphsSupport",
    # torch.fx.passes.dialect.common.cse_pass
    "CSEPass",
    # torch.fx.passes.fake_tensor_prop
    "FakeTensorProp",
    # torch.fx.passes.graph_drawer
    "FxGraphDrawer",
    # torch.fx.passes.graph_manipulation
    "size_bytes",
    # torch.fx.passes.infra.partitioner
    "CapabilityBasedPartitioner",
    "Partition",
    # torch.fx.passes.infra.pass_base
    "PassBase",
    "PassResult",
    # torch.fx.passes.infra.pass_manager
    "PassManager",
    # torch.fx.passes.net_min_base
    "FxNetMinimizerBadModuleError",
    "FxNetMinimizerResultMismatchError",
    "FxNetMinimizerRunFuncError",
    # torch.fx.passes.operator_support
    "OpSupports",
    "OperatorSupport",
    "OperatorSupportBase",
    # torch.fx.passes.pass_manager
    "PassManager",
    # torch.fx.passes.shape_prop
    "ShapeProp",
    # torch.fx.passes.split_module
    "Partition",
    # torch.fx.passes.split_utils
    "Component",
    # torch.fx.passes.splitter_base
    "FxNetAccNodesFinder",
    "FxNetSplitterInternalError",
    "SplitResult",
    "Subgraph",
    # torch.fx.passes.tests.test_pass_manager
    "TestPassManager",
    # torch.fx.passes.tools_common
    "FxNetAccFusionsFinder",
    # torch.fx.passes.utils.common
    "HolderModule",
    # torch.fx.passes.utils.matcher_utils
    "InternalMatch",
    "SubgraphMatcher",
    # torch.fx.passes.utils.source_matcher_utils
    "SourcePartition",
    # torch.fx.proxy
    "Attribute",
    "ParameterProxy",
    "Proxy",
    "Scope",
    "ScopeContextManager",
    "TraceError",
    "TracerBase",
    # torch.fx.subgraph_rewriter
    "Match",
    "ReplacedPatterns",
    # torch.jit.annotations
    "EvalEnv",
    "Module",
    # torch.jit.frontend
    "Builder",
    "ExprBuilder",
    "FrontendError",
    "FrontendTypeError",
    "NotSupportedError",
    "StmtBuilder",
    "UnsupportedNodeError",
    "WithItemBuilder",
    # torch.masked.maskedtensor.core
    "MaskedTensor",
    # torch.multiprocessing.pool
    "Pool",
    # torch.multiprocessing.queue
    "ConnectionWrapper",
    "Queue",
    "SimpleQueue",
    # torch.multiprocessing.reductions
    "SharedCache",
    # torch.multiprocessing.spawn
    "ProcessContext",
    "ProcessException",
    "ProcessExitedException",
    "ProcessRaisedException",
    "SpawnContext",
    # torch.nn.cpp
    "ModuleWrapper",
    "OrderedDictWrapper",
    # torch.nn.modules.container
    "Container",  # deprecated
    # torch.nn.modules.linear
    "NonDynamicallyQuantizableLinear",
    # torch.nn.modules.module
    # TODO: causes multiple sphinx warnings
    # WARNING: more than one target found for cross-reference 'Module'
    "Module",
    # torch.nn.modules.loss
    "NLLLoss2d",  # deprecated
    # torch.nn.modules.normalization
    "CrossMapLRN2d",
    # torch.nn.parallel.data_parallel
    "DataParallel",
    # torch.nn.parallel.distributed
    "DistributedDataParallel",
    # torch.nn.parameter
    "UninitializedTensorMixin",
    # torch.nn.utils.parametrize
    "ParametrizationList",
    # torch.nn.utils.prune
    "CustomFromMask",
    "Identity",
    "L1Unstructured",
    "RandomUnstructured",
    # torch.nn.utils.rnn
    "PackedSequence",
    "PackedSequence_",
    # torch.nn.utils.spectral_norm
    "SpectralNorm",
    "SpectralNormLoadStateDictPreHook",
    "SpectralNormStateDictHook",
    # torch.nn.utils.weight_norm
    "WeightNorm",
    # torch.onnx.errors
    "OnnxExporterError",
    "OnnxExporterWarning",
    "SymbolicValueError",
    "UnsupportedOperatorError",
    # torch.onnx.verification
    "OnnxBackend",
    "OnnxTestCaseRepro",
    # torch.optim.optimizer
    "Optimizer",
    # torch.overrides
    "BaseTorchFunctionMode",
    "TorchFunctionMode",
    # torch.package.file_structure_representation
    "Directory",
    # torch.package.glob_group
    "GlobGroup",
    # torch.package.importer
    "Importer",
    "ObjMismatchError",
    "ObjNotFoundError",
    "OrderedImporter",
    # torch.package.package_exporter
    "PackageExporter",
    "PackagingErrorReason",
    # torch.package.package_importer
    "PackageImporter",
    # torch.autograd.profiler_util
    "EventMetadata",
    # torch.profiler.profiler
    "ExecutionTraceObserver",
    "profile",
    # torch.return_types
    "aminmax",
    "aminmax_out",
    "cummax",
    "cummax_out",
    "cummin",
    "cummin_out",
    "frexp",
    "frexp_out",
    "geqrf",
    "geqrf_out",
    "histogram",
    "histogram_out",
    "histogramdd",
    "kthvalue",
    "kthvalue_out",
    "linalg_cholesky_ex",
    "linalg_cholesky_ex_out",
    "linalg_eig",
    "linalg_eig_out",
    "linalg_eigh",
    "linalg_eigh_out",
    "linalg_inv_ex",
    "linalg_inv_ex_out",
    "linalg_ldl_factor",
    "linalg_ldl_factor_ex",
    "linalg_ldl_factor_ex_out",
    "linalg_ldl_factor_out",
    "linalg_lstsq",
    "linalg_lstsq_out",
    "linalg_lu",
    "linalg_lu_factor",
    "linalg_lu_factor_ex",
    "linalg_lu_factor_ex_out",
    "linalg_lu_factor_out",
    "linalg_lu_out",
    "linalg_qr",
    "linalg_qr_out",
    "linalg_slogdet",
    "linalg_slogdet_out",
    "linalg_solve_ex",
    "linalg_solve_ex_out",
    "linalg_svd",
    "linalg_svd_out",
    "lu_unpack",
    "lu_unpack_out",
    "max",
    "max_out",
    "median",
    "median_out",
    "min",
    "min_out",
    "mode",
    "mode_out",
    "nanmedian",
    "nanmedian_out",
    "qr",
    "qr_out",
    "slogdet",
    "slogdet_out",
    "sort",
    "sort_out",
    "svd",
    "svd_out",
    "topk",
    "topk_out",
    "triangular_solve",
    "triangular_solve_out",
    # torch.serialization
    "LoadEndianness",
    "SourceChangeWarning",
    # torch.sparse.semi_structured
    "SparseSemiStructuredTensor",
    # torch.storage
    "UntypedStorage",
    # torch.torch_version
    "TorchVersion",
    # torch.types
    "SymInt",
    # torch.utils.benchmark.examples.compare
    "FauxTorch",
    # torch.utils.benchmark.examples.spectral_ops_fuzz_test
    "Benchmark",
    # torch.utils.benchmark.op_fuzzers.binary
    "BinaryOpFuzzer",
    # torch.utils.benchmark.op_fuzzers.sparse_binary
    "BinaryOpSparseFuzzer",
    # torch.utils.benchmark.op_fuzzers.sparse_unary
    "UnaryOpSparseFuzzer",
    # torch.utils.benchmark.op_fuzzers.spectral
    "SpectralOpFuzzer",
    # torch.utils.benchmark.op_fuzzers.unary
    "UnaryOpFuzzer",
    # torch.utils.benchmark.utils.common
    "Measurement",
    "TaskSpec",
    # torch.utils.benchmark.utils.compare
    "Colorize",
    "Compare",
    "Table",
    # torch.utils.benchmark.utils.fuzzer
    "FuzzedParameter",
    "FuzzedTensor",
    "Fuzzer",
    "ParameterAlias",
    # torch.utils.benchmark.utils.sparse_fuzzer
    "FuzzedSparseTensor",
    # torch.utils.benchmark.utils.timer
    "CPPTimer",
    "Language",
    "Timer",
    # torch.utils.benchmark.utils.valgrind_wrapper.timer_interface
    "CallgrindStats",
    "CopyIfCallgrind",
    "FunctionCount",
    "FunctionCounts",
    "GlobalsBridge",
    "Serialization",
    # torch.utils.bundled_inputs
    "InflatableArg",
    # torch.utils.checkpoint
    "CheckpointError",
    "CheckpointFunction",
    "DefaultDeviceType",
    # torch.utils.collect_env
    "SystemEnv",
    # torch.utils.cpp_extension
    "BuildExtension",
    # torch.utils.data.dataloader
    "DataLoader",
    # torch.utils.data.datapipes.dataframe.dataframe_wrapper
    "PandasWrapper",
    "default_wrapper",
    # torch.utils.data.datapipes.dataframe.dataframes
    "Capture",
    "CaptureA",
    "CaptureAdd",
    "CaptureCall",
    "CaptureControl",
    "CaptureDataFrame",
    "CaptureDataFrameWithDataPipeOps",
    "CaptureF",
    "CaptureGetAttr",
    "CaptureGetItem",
    "CaptureInitial",
    "CaptureLikeMock",
    "CaptureMul",
    "CaptureSetItem",
    "CaptureSub",
    "CaptureVariable",
    "CaptureVariableAssign",
    "DataFrameTracedOps",
    "DataFrameTracer",
    # torch.utils.data.datapipes.dataframe.datapipes
    "ConcatDataFramesPipe",
    "DataFramesAsTuplesPipe",
    "ExampleAggregateAsDataFrames",
    "FilterDataFramesPipe",
    "PerRowDataFramesPipe",
    "ShuffleDataFramesPipe",
    # torch.utils.data.datapipes.dataframe.structures
    "DataChunkDF",
    # torch.utils.data.datapipes.datapipe
    "DFIterDataPipe",
    "DataChunk",
    "IterDataPipe",
    "MapDataPipe",
    # torch.utils.data.datapipes.iter.callable
    "CollatorIterDataPipe",
    "MapperIterDataPipe",
    # torch.utils.data.datapipes.iter.combinatorics
    "SamplerIterDataPipe",
    "ShufflerIterDataPipe",
    # torch.utils.data.datapipes.iter.combining
    "ConcaterIterDataPipe",
    "DemultiplexerIterDataPipe",
    "ForkerIterDataPipe",
    "MultiplexerIterDataPipe",
    "ZipperIterDataPipe",
    # torch.utils.data.datapipes.iter.filelister
    "FileListerIterDataPipe",
    # torch.utils.data.datapipes.iter.fileopener
    "FileOpenerIterDataPipe",
    # torch.utils.data.datapipes.iter.grouping
    "BatcherIterDataPipe",
    "GrouperIterDataPipe",
    "UnBatcherIterDataPipe",
    # torch.utils.data.datapipes.iter.routeddecoder
    "RoutedDecoderIterDataPipe",
    # torch.utils.data.datapipes.iter.selecting
    "FilterIterDataPipe",
    # torch.utils.data.datapipes.iter.sharding
    "SHARDING_PRIORITIES",
    "ShardingFilterIterDataPipe",
    # torch.utils.data.datapipes.iter.utils
    "IterableWrapperIterDataPipe",
    # torch.utils.data.datapipes.map.callable
    "MapperMapDataPipe",
    # torch.utils.data.datapipes.map.combinatorics
    "ShufflerIterDataPipe",
    # torch.utils.data.datapipes.map.combining
    "ConcaterMapDataPipe",
    "ZipperMapDataPipe",
    # torch.utils.data.datapipes.map.grouping
    "BatcherMapDataPipe",
    # torch.utils.data.datapipes.map.utils
    "SequenceWrapperMapDataPipe",
    # torch.utils.data.datapipes.utils.decoder
    "Decoder",
    "ImageHandler",
    "MatHandler",
    # torch.utils.data.dataset
    "ConcatDataset",
    # torch.utils.data.distributed
    "DistributedSampler",
    # torch.utils.dlpack
    "DLDeviceType",
    # torch.utils.file_baton
    "FileBaton",
    # torch.utils.flop_counter
    "FlopCounterMode",
    # torch.utils.hipify.hipify_python
    "CurrentState",
    "GeneratedFileCleaner",
    "HipifyResult",
    "InputError",
    "Trie",
    "bcolors",
    # torch.utils.hooks
    "BackwardHook",
    "RemovableHandle",
    # torch.utils.mkldnn
    "MkldnnBatchNorm",
    "MkldnnConv1d",
    "MkldnnConv2d",
    "MkldnnConv3d",
    "MkldnnLinear",
    "MkldnnPrelu",
    # torch.utils.mobile_optimizer
    "LintCode",
    # torch.utils.show_pickle
    "DumpUnpickler",
    "FakeClass",
    "FakeObject",
    # torch.utils.tensorboard.writer
    "FileWriter",
    "SummaryWriter",
    # torch.utils.throughput_benchmark
    "ExecutionStats",
    # torch.utils.weak
    "WeakIdKeyDictionary",
    "WeakIdRef",
    "WeakTensorKeyDictionary",
]

# The suffix(es) of source filenames.
# You can specify multiple suffix as a list of string:
#
# source_suffix = ['.rst', '.md']
source_suffix = ".rst"

# The master toctree document.
master_doc = "index"


# Use the linkcode extension to override [SOURCE] links to point
# to the repo. Use the torch_version variable defined above to
# determine link
@functools.cache
def _add_docstr_source_lines(module_file):
    pattern = re.compile(r"^([A-Za-z_]\w*)\s*=\s*_add_docstr\(")
    source_lines = {}
    try:
        with open(module_file, encoding="utf-8") as f:
            for lineno, line in enumerate(f, 1):
                match = pattern.match(line)
                if match is not None:
                    source_lines[match.group(1)] = lineno
    except OSError:
        return {}

    return source_lines


def _find_add_docstr_source(module, fullname):
    if "." in fullname or not re.match(r"^[A-Za-z_]\w*$", fullname):
        return None

    module_file = getattr(module, "__file__", None)
    if module_file is None or not module_file.endswith(".py"):
        return None

    lineno = _add_docstr_source_lines(module_file).get(fullname)
    if lineno is None:
        return None
    return module_file, lineno


def linkcode_resolve(domain, info):
    if domain != "py":
        return None
    if not info["module"]:
        return None

    try:
        module = __import__(info["module"], fromlist=[""])
    except Exception:
        return None

    try:
        obj = module
        for part in info["fullname"].split("."):
            obj = getattr(obj, part)
        obj = inspect.unwrap(obj)
        fn = inspect.getsourcefile(obj)
        source, lineno = inspect.getsourcelines(obj)
    except Exception:
        resolved = _find_add_docstr_source(module, info["fullname"])
        if resolved is None:
            return None
        fn, lineno = resolved

    # Determine the tag based on the torch_version
    if RELEASE:
        version_parts = torch_version.split(
            "."
        )  # For release versions, format as "vX.Y.Z" for correct path in repo
        patch_version = (
            version_parts[2].split("+")[0].split("a")[0]
        )  # assuming a0 always comes after release version in versions.txt
        version_path = f"v{version_parts[0]}.{version_parts[1]}.{patch_version}"
    else:
        version_path = torch.version.git_version
    fn = os.path.relpath(fn, start=os.path.dirname(torch.__file__))
    return (
        f"https://github.com/pytorch/pytorch/blob/{version_path}/torch/{fn}#L{lineno}"
    )


# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
#
# This is also used if you do content translation via gettext catalogs.
# Usually you set "language" from the command line for these cases.
language = "en"

# List of patterns, relative to source directory, that match files and
# directories to ignore when looking for source files.
# This patterns also effect to html_static_path and html_extra_path
exclude_patterns = []

# The name of the Pygments (syntax highlighting) style to use.
pygments_style = "sphinx"

# If true, `todo` and `todoList` produce output, else they produce nothing.
# Disable docstring inheritance
autodoc_inherit_docstrings = False

# Show type hints in the description
autodoc_typehints = "description"

# Add parameter types if the parameter is documented in the docstring
autodoc_typehints_description_target = "documented_params"

# Type aliases for common types
# Sphinx type aliases only works with Postponed Evaluation of Annotations
# (PEP 563) enabled (via `from __future__ import annotations`), which keeps the
# type annotations in string form instead of resolving them to actual types.
# However, PEP 563 does not work well with JIT, which uses the type information
# to generate the code. Therefore, the following dict does not have any effect
# until PEP 563 is supported by JIT and enabled in files.
autodoc_type_aliases = {
    "_size_1_t": "int or tuple[int]",
    "_size_2_t": "int or tuple[int, int]",
    "_size_3_t": "int or tuple[int, int, int]",
    "_size_4_t": "int or tuple[int, int, int, int]",
    "_size_5_t": "int or tuple[int, int, int, int, int]",
    "_size_6_t": "int or tuple[int, int, int, int, int, int]",
    "_size_any_opt_t": "int or None or tuple",
    "_size_2_opt_t": "int or None or 2-tuple",
    "_size_3_opt_t": "int or None or 3-tuple",
    "_ratio_2_t": "float or tuple[float, float]",
    "_ratio_3_t": "float or tuple[float, float, float]",
    "_ratio_any_t": "float or tuple",
    "_tensor_list_t": "Tensor or tuple[Tensor]",
}

# Enable overriding of function signatures in the first line of the docstring.
autodoc_docstring_signature = True

# Exclude inherited IntEnum methods that have RST formatting issues in their docstrings
autodoc_default_options = {
    "exclude-members": "from_bytes, to_bytes",
}

# -- katex javascript in header
#
#    def setup(app):
#    app.add_javascript("https://cdn.jsdelivr.net/npm/katex@0.10.0-beta/dist/katex.min.js")


# -- Options for HTML output ----------------------------------------------
#
# The theme to use for HTML and HTML Help pages.  See the documentation for
# a list of builtin themes.
#
#
#


# Theme options are theme-specific and customize the look and feel of a theme
# further.  For a list of options available for each theme, see the
# documentation.

# Add any paths that contain custom static files (such as style sheets) here,
# relative to this directory. They are copied after the builtin static files,
# so a file named "default.css" will overwrite the builtin "default.css".

html_css_files = [
    "css/jit.css",
    "https://cdn.jsdelivr.net/npm/katex@0.10.0-beta/dist/katex.min.css",
]


from sphinx.ext.coverage import CoverageBuilder


# NB: Due to some duplications of the following modules/functions, we keep
# them as expected failures for the time being instead of return 1
ignore_duplicated_modules = {
    "torch.nn.utils.weight_norm",
    "torch.nn.utils.spectral_norm",
    "torch.nn.parallel.data_parallel",
    "torch.ao.quantization.quantize",
}


def coverage_post_process(app, exception):
    if exception is not None:
        return

    # Only run this test for the coverage build
    if not isinstance(app.builder, CoverageBuilder):
        return

    if not torch.distributed.is_available():
        raise RuntimeError(
            "The coverage tool cannot run with a version "
            "of PyTorch that was built with USE_DISTRIBUTED=0 "
            "as this module's API changes."
        )

    # These are all the modules that have "automodule" in an rst file
    # These modules are the ones for which coverage is checked
    modules = app.env.domaindata["py"]["modules"]

    def is_not_internal(modname):
        split_name = modname.split(".")
        for name in split_name:
            if name[0] == "_":
                return False
        return True

    # Sphinx's built-in coverage only catches items that pass
    # inspect.isfunction() or inspect.isclass(), missing decorated/wrapped
    # callables (e.g. @cache, @lru_cache, @deprecated). Cross-reference
    # __all__ exports directly against the documented objects dict.
    objects = app.env.domaindata["py"]["objects"]
    ignore_names = set(app.config.coverage_ignore_functions) | set(
        app.config.coverage_ignore_classes
    )
    # pybind11 objects from torch._C that are in __all__ but cannot be
    # documented via autosummary: the first three have .str members that
    # create ambiguous cross-references with Python's builtin str;
    # unify_type_list has a C++ arglist Sphinx cannot parse.
    ignore_names |= {"ErrorReport", "FutureType", "StreamObjType", "unify_type_list"}
    undocumented = []
    for mod_name in modules:
        if not is_not_internal(mod_name):
            continue
        try:
            mod = __import__(mod_name, fromlist=["__all__"])
        except ImportError:
            continue
        for name in getattr(mod, "__all__", []):
            if name.startswith("_") or name in ignore_names:
                continue
            full_name = f"{mod_name}.{name}"
            obj = getattr(mod, name, None)
            if obj is None or not callable(obj):
                continue
            if getattr(obj, "__module__", mod_name) != mod_name:
                continue
            if full_name in objects:
                continue
            # Check if documented under a parent module (e.g.
            # torch.distributed.distributed_c10d.get_rank is documented
            # as torch.distributed.get_rank via currentmodule).
            parts = mod_name.split(".")
            if any(
                f"{'.'.join(parts[:i])}.{name}" in objects for i in range(1, len(parts))
            ):
                continue
            undocumented.append(full_name)

    if undocumented:
        items = "\n".join(f"  - {u}" for u in sorted(undocumented))
        print(
            f"\nThe following public APIs are in __all__ but not documented:\n{items}\n"
            "Either add them to the appropriate .rst/.md doc file or "
            "remove from __all__."
        )
        sys.exit(1)

    # We go through all the torch submodules and make sure they are
    # properly documented
    missing = set()

    # The walk function does not return the top module
    if "torch" not in modules:
        missing.add("torch")

    for _, modname, ispkg in pkgutil.walk_packages(
        path=torch.__path__, prefix=torch.__name__ + "."
    ):
        if is_not_internal(modname):
            if modname not in modules and modname not in ignore_duplicated_modules:
                missing.add(modname)

    output = []

    if missing:
        mods = ", ".join(missing)
        output.append(
            f"\nYou added the following module(s) to the PyTorch namespace '{mods}' "
            "but they have no corresponding entry in a doc .rst file. You should "
            "either make sure that the .rst file that contains the module's documentation "
            "properly contains either '.. automodule:: mod_name' (if you do not want "
            "the paragraph added by the automodule, you can simply use '.. py:module:: mod_name') "
            " or make the module private (by appending an '_' at the beginning of its name)."
        )

    # The output file is hard-coded by the coverage tool
    # Our CI is setup to fail if any line is added to this file
    output_file = path.join(app.outdir, "python.txt")

    if output:
        with open(output_file, "a") as f:
            for o in output:
                f.write(o)


def process_docstring(app, what_, name, obj, options, lines):
    """
    Custom process to transform docstring lines Remove "Ignore" blocks

    Args:
        app (sphinx.application.Sphinx): the Sphinx application object

        what (str):
            the type of the object which the docstring belongs to (one of
            "module", "class", "exception", "function", "method", "attribute")

        name (str): the fully qualified name of the object

        obj: the object itself

        options: the options given to the directive: an object with
            attributes inherited_members, undoc_members, show_inheritance
            and noindex that are true if the flag option of same name was
            given to the auto directive

        lines (List[str]): the lines of the docstring, see above

    References:
        https://www.sphinx-doc.org/en/master/usage/extensions/autodoc.html
    """
    import re

    remove_directives = [
        # Remove all xdoctest directives
        re.compile(r"\s*>>>\s*#\s*x?doctest:\s*.*"),
        re.compile(r"\s*>>>\s*#\s*x?doc:\s*.*"),
    ]
    filtered_lines = [
        line for line in lines if not any(pat.match(line) for pat in remove_directives)
    ]
    # Modify the lines inplace
    lines[:] = filtered_lines

    # make sure there is a blank line at the end
    if lines and lines[-1].strip():
        lines.append("")


def setup(app):
    app.connect("build-finished", coverage_post_process)
    app.connect("autodoc-process-docstring", process_docstring)
    app.connect("html-page-context", hide_edit_button_for_pages)
    app.config.add_last_updated = True

    # Force serial reads to avoid pipe congestion from large env pickles.
    # Sphinx's parallel read sends the entire environment (100s of MB for
    # PyTorch) through a 64KB OS pipe per worker, which causes extreme
    # slowdowns with many workers. Serial reads avoid this overhead while
    # parallel writes (which send trivial payloads) remain enabled.
    from sphinx.builders import Builder

    _orig_read_serial = Builder._read_serial

    def _serial_read_ignoring_nproc(self, docnames, nproc=1):
        return _orig_read_serial(self, docnames)

    Builder._read_parallel = _serial_read_ignoring_nproc

    _fix_katex_server_race(app)

    return {"version": "0.1", "parallel_read_safe": True}


def _fix_katex_server_race(app):
    """Retry on ConnectionRefusedError when connecting to the KaTeX server.

    sphinxcontrib-katex 0.9.x starts a Node.js server and connects via Unix
    socket. There's a race: Python sees the socket file (created by bind())
    and calls connect() before Node.js has called listen(). On slow CI
    machines this causes ConnectionRefusedError. Fix by retrying connect().
    """
    try:
        from sphinxcontrib.katex import KaTeXServer
    except ImportError:
        return

    import socket as _socket
    import time as _time

    @classmethod
    def _start_with_retry(cls, rundir, timeout):
        from subprocess import PIPE, Popen

        from sphinxcontrib.katex import ONE_MILLISECOND

        socket_path = rundir / "katex.sock"
        cmd = cls.build_command(socket=socket_path)
        process = Popen(cmd, stdin=PIPE, stdout=PIPE, cwd=rundir)

        startup_start = _time.monotonic()
        while not socket_path.is_socket():
            _time.sleep(ONE_MILLISECOND)
            if _time.monotonic() - startup_start > timeout:
                raise cls.timeout_error(timeout)

        # Retry connect() — bind() creates the socket file but listen()
        # is async in Node.js and may not be ready yet.
        while True:
            remaining = startup_start + timeout - _time.monotonic()
            if remaining <= 0:
                raise cls.timeout_error(timeout)
            try:
                sock = _socket.socket(_socket.AF_UNIX, _socket.SOCK_STREAM)
                sock.settimeout(remaining)
                sock.connect(str(socket_path))
                break
            except ConnectionRefusedError:
                sock.close()
                _time.sleep(ONE_MILLISECOND)
            except TimeoutError:
                sock.close()
                raise cls.timeout_error(timeout)  # noqa: B904

        return process, sock

    KaTeXServer.start_server_process = _start_with_retry


def hide_edit_button_for_pages(app, pagename, templatename, context, doctree):
    if pagename.startswith("generated/"):
        context["theme_use_edit_page_button"] = False
        # Limit sidebar depth and collapse inactive sections for generated
        # API pages to avoid embedding ~300 KB of navigation markup per page.
        # Only the active section's children are shown.
        context["theme_navigation_depth"] = 2
        context["theme_collapse_navigation"] = "true"


# From PyTorch 1.5, we now use autogenerated files to document classes and
# functions. This breaks older references since
# https://pytorch.org/docs/stable/torch.html#torch.flip
# moved to
# https://pytorch.org/docs/stable/generated/torch.flip.html
# which breaks older links from blog posts, stack overflow answers and more.
# To mitigate that, we add an id="torch.flip" in an appropriated place
# in torch.html by overriding the visit_reference method of html writers.
# Someday this can be removed, once the old links fade away

from sphinx.writers import html, html5


def replace(Klass):
    old_call = Klass.visit_reference

    def visit_reference(self, node):
        if "refuri" in node and "generated" in node.get("refuri"):
            ref = node.get("refuri")
            ref_anchor = ref.split("#")
            if len(ref_anchor) > 1:
                # Only add the id if the node href and the text match,
                # i.e. the href is "torch.flip#torch.flip" and the content is
                # "torch.flip" or "flip" since that is a signal the node refers
                # to autogenerated content
                anchor = ref_anchor[1]
                txt = node.parent.astext()
                if txt == anchor or txt == anchor.split(".")[-1]:
                    self.body.append(f'<p id="{ref_anchor[1]}"/>')
        return old_call(self, node)

    Klass.visit_reference = visit_reference


replace(html.HTMLTranslator)
replace(html5.HTML5Translator)

# -- Options for HTMLHelp output ------------------------------------------

# Output file base name for HTML help builder.
htmlhelp_basename = "PyTorchdoc"


# -- Options for LaTeX output ---------------------------------------------

latex_engine = "lualatex"
latex_show_urls = "footnote"

latex_elements = {
    "papersize": "letterpaper",
    "pointsize": "10pt",
    "tableofcontents": r"\pdfbookmark[0]{Contents}{toc}\tableofcontents",
    "preamble": r"""
       \usepackage{tocloft}
       \setcounter{tocdepth}{3}
       \setcounter{secnumdepth}{3}
       % Fix table column widths
       \renewenvironment{tabulary}{\begin{longtable}{p{0.3\linewidth}p{0.7\linewidth}}}{\end{longtable}}

       % Ensure tables don't overflow
       \AtBeginEnvironment{tabular}{\sloppy}
    """,
    "fncychap": r"\usepackage[Bjornstrup]{fncychap}",
    "extraclassoptions": "oneside",
}

# Grouping the document tree into LaTeX files. List of tuples
# (source start file, target name, title,
#  author, documentclass [howto, manual, or own class]).


latex_documents = [
    (
        master_doc,
        "pytorch.tex",
        "PyTorch Documentation",
        "Torch Contributors",
        "manual",
    ),
]
latex_use_xindy = False


# -- Options for manual page output ---------------------------------------

# One entry per manual page. List of tuples
# (source start file, name, description, authors, manual section).
man_pages = [(master_doc, "PyTorch", "PyTorch Documentation", [author], 1)]


# -- Options for Texinfo output -------------------------------------------

# Grouping the document tree into Texinfo files. List of tuples
# (source start file, target name, title, author,
#  dir menu entry, description, category)
texinfo_documents = [
    (
        master_doc,
        "PyTorch",
        "PyTorch Documentation",
        author,
        "PyTorch",
        "One line description of project.",
        "Miscellaneous",
    ),
]


# Example configuration for intersphinx: refer to the Python standard library.
intersphinx_mapping = {
    "python": ("https://docs.python.org/3", None),
    "numpy": ("https://numpy.org/doc/stable", None),
}

import sphinx.ext.doctest

# -- A patch that prevents Sphinx from cross-referencing ivar tags -------
# See http://stackoverflow.com/a/41184353/3343043
from docutils import nodes
from sphinx import addnodes
from sphinx.util.docfields import TypedField


# Without this, doctest adds any example with a `>>>` as a test
doctest_test_doctest_blocks = ""
doctest_default_flags = sphinx.ext.doctest.doctest.ELLIPSIS
doctest_global_setup = """
import torch
try:
    import torchvision
except ImportError:
    torchvision = None
"""


def patched_make_field(self, types, domain, items, **kw):
    # `kw` catches `env=None` needed for newer sphinx while maintaining
    #  backwards compatibility when passed along further down!

    # type: (List, unicode, Tuple) -> nodes.field
    def handle_item(fieldarg, content):
        par = nodes.paragraph()
        par += addnodes.literal_strong("", fieldarg)  # Patch: this line added
        # par.extend(self.make_xrefs(self.rolename, domain, fieldarg,
        #                           addnodes.literal_strong))
        if fieldarg in types:
            par += nodes.Text(" (")
            # NOTE: using .pop() here to prevent a single type node to be
            # inserted twice into the doctree, which leads to
            # inconsistencies later when references are resolved
            fieldtype = types.pop(fieldarg)
            if len(fieldtype) == 1 and isinstance(fieldtype[0], nodes.Text):
                typename = fieldtype[0].astext()
                builtin_types = ["int", "long", "float", "bool", "type"]
                for builtin_type in builtin_types:
                    pattern = rf"(?<![\w.]){builtin_type}(?![\w.])"
                    repl = f"python:{builtin_type}"
                    typename = re.sub(pattern, repl, typename)
                par.extend(
                    self.make_xrefs(
                        self.typerolename,
                        domain,
                        typename,
                        addnodes.literal_emphasis,
                        **kw,
                    )
                )
            else:
                par += fieldtype
            par += nodes.Text(")")
        par += nodes.Text(" -- ")
        par += content
        return par

    fieldname = nodes.field_name("", self.label)
    if len(items) == 1 and self.can_collapse:
        fieldarg, content = items[0]
        bodynode = handle_item(fieldarg, content)
    else:
        bodynode = self.list_type()
        for fieldarg, content in items:
            bodynode += nodes.list_item("", handle_item(fieldarg, content))
    fieldbody = nodes.field_body("", bodynode)
    return nodes.field("", fieldname, fieldbody)


TypedField.make_field = patched_make_field

copybutton_prompt_text = r">>> |\.\.\. "
copybutton_prompt_is_regexp = True