1. 18 10月, 2021 8 次提交
    • L
      Lml/vhp (#36146) · 4c0ad772
      levi131 提交于
      * init functional jacobian api
      
      * finish test with dtype float32
      
      * add float64 test case
      
      * polish code
      
      * use atol=1e-5 with dtype float64
      
      * fix for ci
      
      * set timeout for test_jacobian
      
      * init hessian API
      
      * save status
      
      * polish API docstring
      
      * modify docstring
      
      * add utils.py
      
      * save status
      
      * fix dygraph double grad dtype error when calling for high differential senario
      
      * reinvoke ci
      
      * test_hessian.py is ok
      
      * polish hessian API
      
      * init vhp
      
      * Revert "init vhp"
      
      This reverts commit cbd4d3b66abe82b0ac10721b9eddeb7d82e0a1c8.
      
      * init vhp
      
      * finish vhp API logically
      
      * add test for partial_engine.cc
      
      * modify numerical_delta with dtype float32
      
      * merge fix for dtype float64
      
      * spell fix
      
      * save status
      
      * polish code
      
      * rm _stop_gradient_pre_process
      
      * save status
      
      * add example for vhp interface
      
      * add _compute_numerical_vjp and _compute_numerical_vhp
      
      * test is ok
      
      * vhp is ok
      
      * add testVHPFloat64
      
      * modify for comments
      
      * modify format
      
      * modify format
      
      * save status
      
      * test_vhp is ok
      
      * finish code polish
      
      * small modify for v is None
      Co-authored-by: NJiabinYang <360788950@qq.com>
      4c0ad772
    • Q
    • Q
      [NPU] fix dtype for arg_max, test=develop (#36457) · 8757fc5b
      Qi Li 提交于
      8757fc5b
    • S
      Add operators for async read & async write (#36333) · 3845afff
      Siming Dai 提交于
      * fix async_read bug
      
      * change index place to cpu
      
      * add tensor size judge
      
      * add async_read & async_write test
      
      * fix bug in async_write
      
      * fix mac py3 ci
      
      * fix bug for cpu version paddle
      
      * fix windows ci bug
      
      * change input argument error type
      
      * change const_cast to mutable_data
      
      * add async_write out-of-bound check and consumate error hint
      
      * fix a small bug for dst_tensor
      
      * add docs and refine codes
      
      * refine docs
      
      * notest,test=windows_ci
      
      * fix windows ci
      
      * fix require
      
      * fix code-block
      
      * add core.is_compiled_with_cuda()
      3845afff
    • C
      quant support matmul_v2 (#36469) · 051544b6
      ceci3 提交于
      * quant support matmul_v2
      
      * fix format
      051544b6
    • T
      [XPU AMP] 1. xpu support gradient acc 2. xpu support create tensor in dygraph... · d19a9b39
      taixiurong 提交于
      [XPU AMP] 1. xpu support gradient acc 2. xpu support create tensor in dygraph 3. xpu support update weight params in amp (#36439)
      
      d19a9b39
    • T
      [autograd.functional] Fix a bug on handling v=None in vjp and jvp (#36445) · 79dbbcce
      Tongxin Bai 提交于
      * autograd.functional passed pylint checker.
      
      * autograd.functional: fix import errors.
      
      * autograd.functional: fixed unit tests.
      
      * autograd.functional minor format change
      
      * [autograd.functional] Fixed vjp and jvp's v=None bug.
      79dbbcce
    • H
      modify ut of cond (#36475) · e496d1e9
      Haohongxiang 提交于
      e496d1e9
  2. 17 10月, 2021 1 次提交
  3. 16 10月, 2021 1 次提交
  4. 15 10月, 2021 7 次提交
    • D
      fix opt-offload save bug (#36433) · e703a2ed
      duanboqiang 提交于
      e703a2ed
    • Z
      Add ResNetUnit Python API (#35426) · 12882b2f
      Zhang Zheng 提交于
      12882b2f
    • N
      add resnext (#36070) · 277c9a55
      Nyakku Shigure 提交于
      * add resnext model
      * add zh docs
      * add unittest
      * test performance
      Co-authored-by: Ainavo's avatarAinavo <ainavo@163.com>
      Co-authored-by: Npithygit <pyg20200403@163.com>
      Co-authored-by: Ainavo's avatarAinavo <ainavo@163.com>
      Co-authored-by: Npithygit <pyg20200403@163.com>
      277c9a55
    • 0
      fix no_grad context error in train mode when using save/load (#36434) · 37257d6a
      0x45f 提交于
      * fix no_grad context error in train mode when using save/load
      
      * change net to train mode in test case
      37257d6a
    • J
      Add BuildCinnPass (#36345) · b3f02c57
      jiangcheng 提交于
      * Add CinnSubgraphSearchPass
      
      * solve CI problem of subgraph order not same
      
      * fix some bug by review advices
      
      * ensure the independently of subgraph, that mean the subgraph should not have link to out-graph
      
      * rename cinn_subgraph_search_pass to build_cinn_pass and delete paddle_to_cinn_pass
      
      * add flag to control wheter append build cinn pass
      
      * remove AppendPass at ParallelExecutorPassBuilder
      
      * rename paddle_to_cinn_pass to build_cinn_pass in build_strategy and close test_run_from_cinn
      b3f02c57
    • J
      [New Feature] Support tanh triple grad (#36225) · 808be657
      Jiabin Yang 提交于
      * native commit for triple grad of sigmod
      
      * Updated unittests files
      
      * init functional jacobian api
      
      * Updated trible_test func
      
      * Updated gradient_checker & test_script
      
      * finish test with dtype float32
      
      * add float64 test case
      
      * polish code
      
      * use atol=1e-5 with dtype float64
      
      * fix for ci
      
      * set timeout for test_jacobian
      
      * fix dygraph grad to support high differential
      
      * polish API docstring
      
      * Updated gradient checker and some related files
      
      * fix double grad strip error for high differential
      
      * fix double grad strip error for high differential
      
      * Add Sigmoid triple grad tests
      
      * fix dygraph double grad dtype error when calling for high differential senario
      
      * Updated triple grad teses func
      
      * Use np.random to initialize ddx
      
      * Updated triple_grad_check func
      
      * add todo for gradient checker and refine some comments
      
      * remove additional code
      
      * add test for warnging in backward.py
      
      * add tanh triple grad
      
      * format python code
      
      * refine code
      Co-authored-by: Nveyron95 <veyron_wu@163.com>
      Co-authored-by: Nlevi131 <limaolin01@baidu.com>
      808be657
    • Z
      fix momentum ops (#36452) · 4dda18a8
      Zeng Jinle 提交于
      4dda18a8
  5. 14 10月, 2021 11 次提交
  6. 13 10月, 2021 12 次提交