* [AutoParallel] quantization pass support export

* support subgraph

* move_presist_var_to_global_block

* update unittest

* fix ci-coverage

* fix codestyle

* fix fake_dequantize_op

* remove unused var

* fix ci error and aprroval error

* add unittest for fp16 in test_dequant_linear

* replace mutable data

* fix unittest in non-cuda-core

* fix unittest
Co-authored-by: Ncarryyu <569782149@qq.com>
Co-authored-by: Nwufeisheng <wfs1997@163.com>

* delete old dygraph sharding

* Support priority scheduling for standalone executor

* Add CPU test

* rm in_legacy_dygraph nn part1

* rm non_static_mode

* modify rrelu

* delete _in_legacy_dygraph

* feat: broadcast_object_list & scatter_object_list

* chore: update ut conf

* get_backend & is_available

* docs: update requirements

* fix: resolve conflicts
Co-authored-by: NLiYuRio <liyuruijx@163.com>

* rm legacy unittest part5

* add custom op

* add index sample fp16 support

* remove fluid APIs in distributed_strategy.py and role_maker.py

* Revert "remove fluid APIs in distributed_strategy.py and role_maker.py"

This reverts commit 223bbee990d3bf69e252fc3c0f19e3873550a264.

* move load_inference_model to distributed

* fix origin develop codes diff

* move _endpoints_replacement

* delete line

* reset line

* add unittest case of load_inference_model

* fix unittest

* fix unittest

* fix coverage

* fix coverage

* [Auto Parallel] Rename methods of ProcessMesh

* [Auto Parallel] Impl the python process_mesh by the c++ one

* [Auto Parallel] Add some minor modifications

* [Auto Parallel] Rename some methods

* [Auto Parallel] Remove unnecessary codes

* [Auto Parallel] Add back some removed files

* [Auto Parallel] Fix bugs

* [Auto Parallel] Fix a bug

* Update process_mesh.cc

* [Auto Parallel] Fix a bug

* remove loss.py

* fix __init__.py

* delete legacy dygraph code in python/paddle/distributed

* refine

* 修改linspace参数"stop"的介绍

* 更新了conv2d_transpose英文文档中 1.公式异常 2.参数说明异常 3.padding=SAME和VALID的公式说明; test=docs_preview

* 解决了
1.英文版Note中的api_fluid_layers_conv2d的链接失效，
2. 英文版参数中的padding_start的参数错误
3. 参数中bias_attr和param_attr的链接失效
4. 参数缺少optional

* 更新了paddle.static.auc的英文文档的Return描述以及函数的参数

* Update python/paddle/tensor/creation.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* 更新了shard_index的ignore_value描述

* 修改了英文文档的paddle.static.nn.conv3d_transpose的公式介绍

* add py_func COPY-FROM label; test=document_fix

* Update python/paddle/tensor/manipulation.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* formula; test=document_fix

* formula; test=document_fix

* formula; test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* memcpy overlap

* memcpy newexe

* fix docstrings:
1. 修复了 python/paddle/nn/layer/conv.py 中出现的 Tex 语法问题；
2. 为 python/paddle/nn/functional/conv.py 修复 Tex 语法问题，修改 note 及参数说明格式，尝试添加论文链接；
3. 修改 python/paddle/nn/layer/loss.py、python/paddle/nn/functional/loss.py 中参数说明的格式；
4. 修改 python/paddle/nn/layer/common.py、python/paddle/nn/functional/common.py 中参数说明的格式，添加 optional 说明。

* fix docstring format, add paper's hyperlink.

* fix formula Tex.

* fix format error; test=docs_preview

* fix reference hyperlink; test=docs_preview

* fix docstring; test=docs_preview

* fix Tex format.

* fix api reference;  test=docs_preview

* test=document_fix

* Update conv.py

* formula; test=document_fix

* test=document_fix

* formula; test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* fix matmul double and triple grad

* remove some comment

* add matmul_double_grad unit test

* fix matmul triple grad

* fix dot triple grad and add unit test

* modify codestyle

* fix dot_grad

* refactor dot triple grad

* disable some unit test

* fix unit test

* fix unit test in double grad

* rm paddle.fluid.contrib.layers.BasicLSTMUnit basic_lstm BasicGRUUnit basic_gru

* rm dependency in __init__.py

* rm eager guard test

* retain grad for xpu test

* fix custom double_grad backward=None

* fix custom_relu.cu bug && polish testcase of double_grad

* remove old dynamic graph test

* add warp transducer code

* [eager] use CPUAllocator directly

* modify pstring sizeof 48 default

* rm CPU test for NaiveBestFitAllocator

* fix Mac ci compile errors

* use UNUSED to state unused_obj

* mv UNUSED statement to allocator_facade.cc

* fix roi_align

* fix yolov3 test case

* recover original code

* recover original code

* fix trt roi_align test
Co-authored-by: Njerrywgz <jerrywgz@126.com>

* delete distribute old dygraph test cast

* remove custom op eager guard tests

* rm custom_runtime eager guard tests

* rm xccl eager tests

* modify

* modify

* set grad retain for custom_relu test

* retain grad for custom relu op

* modify custom_relu test

* Fix bug of reduce_sum op. When input.numel() > INT32_MAX, its result
is wrong.

* Remove climits.

* Clean Fluid APIs in python/paddle/fluid/layers/nn.py,
migrate spectral_norm and row_conv APIs, and remove ont_hot.
Including following files:
1. python/paddle/fluid/layers/nn.py
2. python/paddle/fluid/tests/unittests/collective/fleet/parallel_dygraph_transformer.py
3. python/paddle/fluid/tests/unittests/dist_transformer.py
4. python/paddle/fluid/tests/unittests/dygraph_to_static/transformer_dygraph_model.py
5. python/paddle/fluid/tests/unittests/ipu/test_one_hot_op_ipu.py
6. python/paddle/fluid/tests/unittests/test_imperative_auto_prune.py
7. python/paddle/fluid/tests/unittests/test_imperative_load_static_param.py
8. python/paddle/fluid/tests/unittests/test_layers.py
9. python/paddle/fluid/tests/unittests/test_one_hot_op.py
10. python/paddle/fluid/tests/unittests/test_row_conv_op.py
11. python/paddle/fluid/tests/unittests/test_runtime_and_compiletime_exception.py
12. python/paddle/fluid/tests/unittests/test_spectral_norm_op.py
13. python/paddle/static/nn/__init__.py
14. python/paddle/static/nn/common.py

* Polish code.

* Fix some bugs.

* Remove useless unittest.

* Fix some bug.

* Polish example code.

* Fix some bugs.

* Fix some bugs.

* rm unittests eager guard tests part21 sum_op2tensor_zero

* modify tensor hook test

* rm assert for type promotion

* review

* remove unittests eager guard tests part4 concat2custom_grad

* modify conv2d&3d test

* fix a bug in mem_allocated

* valueError->indexError for cross_entropy_loss test

* reset max_alloc_size