- 23 8月, 2021 12 次提交
-
-
由 Yinggang Wang 提交于
* refactor(LrScheduler): rename CosineAnnealing to CosineDecay reference: https://www.tensorflow.org/api_docs/python/tf/keras/optimizers/schedules/CosineDecay * feat(CosineAnnealingLR): add CosineAnnealingLR * refine parameter name * auto format by CI * test(Graph): fix graph test bug when using CosineDecayLR Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 l702572275 提交于
* add runtime error * add runtime error * fix OF_UNIMPLEMENTED * Add TensorImpl info * Add python single test * rename * fi UNIMPLEMENTED_THEN_RETURN * fix UNIMPLEMENTED * format * fix UNIMPLEMENT * auto format by CI * fix format * Modify OF_RUNTIME ERROR to OF_RUNTIME_ERROR * Modify OF_COMPLIE_OPTION_EEEOR to OF_COMPLIE_OPTION_ERROR * Modify OF_RUNTIME ERROR to OF_RUNTIME_ERROR * Fix omissions * auto format by CI Co-authored-by: NYinggang Wang <wyg19970408@gmail.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Li Xinqi 提交于
This reverts commit 0ba9198e. Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Juncheng 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Houjiang Chen 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Twice 提交于
* xrt: fix all warnings * xrt: comment in fix_ostream_nullptr.h * xrt: clang format Co-authored-by: NHoujiang Chen <chenhoujiangcug@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 qq_22305325 提交于
* support_eager_boxing_Sm_2_Sn * Update collective_boxing_interpreter.cpp * Update eager_boxing_interpreter_mgr.cpp * refine * auto format by CI * refine * refine Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 Shenghang Tsai 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Juncheng 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 l702572275 提交于
* Move framework/local_dep_object.* to the eager directory * Move framework/local_dep_object.* to the eager directory
-
由 Houjiang Chen 提交于
Co-authored-by: Ncheng cheng <472491134@qq.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 leaves-zwx 提交于
* fix in_grads size * correct comment Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
- 22 8月, 2021 7 次提交
-
-
由 Houjiang Chen 提交于
* Functional support multiple signatures. * Add TypeError and refine error message. * Finish py function. * Refactor add, sub, mul and div functional apis. * Fix merge error * Support scalar add/sub/mul/div tensor. * auto format by CI * Fix optional argument * Set inplace keyword only * Fix throw and treat long as float * Fixup merge error * Check tensor type * Optimize recursive * Fix add static zero * Fix merge * Fix warning as error * Fix warning as error * Fix merge * Fix merge Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 Li Xinqi 提交于
* PlacedNdSbp * 1) fix compiler complaints; 2) add an unittest * refactor DECORATE
-
由 Li Xinqi 提交于
* PlacedNdSbp * 1) fix compiler complaints; 2) add an unittest
-
由 Luyang 提交于
* tensor.cpu/tensor.cuda * register tensor dtype modules * export api docs * fix comments * refine * register module methods * refine * format * add testcases * upper rmsprop tol * refine * add more test case * auto format by CI * skip test if only cpu * auto format by CI * refine * refine * fix test * auto format by CI Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 leaves-zwx 提交于
* refactor to * test * call cast when device_type is same
-
由 Xiaoyu Zhang 提交于
* add fused_self_attention functor * add fused_self_attention gradients * add test case * fix backward immpl bug * fix bug * fix bug * fix backward bug * fix comments * code format * fix comment * fix ci error * fix ci error * add init FusedSelfAttentionInterpState struct Co-authored-by: NYao Chi <later@usopp.net> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 daquexian 提交于
* dos2unix Signed-off-by: Ndaquexian <daquexian566@gmail.com> * remove automerge on fail Signed-off-by: Ndaquexian <daquexian566@gmail.com> * test ci Signed-off-by: Ndaquexian <daquexian566@gmail.com> * Revert "test ci" This reverts commit 0f1051f3.
-
- 21 8月, 2021 8 次提交
-
-
由 daquexian 提交于
Signed-off-by: Ndaquexian <daquexian566@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Xiaoyu Xu 提交于
* add sys conf * format * refine warmup api * refine config * format * add WarmupLR * add warmup lr * refactor graph folder structure * export graph * add amp conf * deal with lr init step and test cpu * add WarmUpLR * add flow config * GradScaler and config * rm useless code * rm useless code * add api_enable_cudnn_fused_normalization_add_relu * move config * move amp * refactor WarmUpLrScheduler * add comment * add arg Co-authored-by: Nleaves-zwx <kunta0932@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Luyang 提交于
* reemove redundant codes in ModuleDict * auto format by CI * add moduledict test * refine * refactor test case * auto format by CI * auto format by CI * refine * refine * refine docs format Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 GehangZhang 提交于
* add flow.relu * add test_relu.py * fix for comments * add docstring for oneflow.relu * fix bugs * add docstring for flow.Tensor.relu * fix for comments * fix for comments v2 * auto format by CI * fix license bug * auto format by CI * fix tets_quantization.py bug Co-authored-by: NYao Chi <later@usopp.net> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 daquexian 提交于
Signed-off-by: Ndaquexian <daquexian566@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Xiaoyu Zhang 提交于
* fix var impl bug * add more test_case * refine * add tensor method test * fix comment * auto format by CI * fix comments * fix ci error * fix run ci error Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 QiangX-man 提交于
* add reduce_prod op * add test_prod * add prod backward * modify backward * mod backward * mod prod_test * add note * mod prod test * conflict fixed * conflict fixed * auto format by CI Co-authored-by: NYongtaoShi <73167956+YongtaoShi@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 cheng cheng 提交于
-
- 20 8月, 2021 13 次提交
-
-
由 Juncheng 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 daquexian 提交于
* use clang-tidy-diff.py return code and add -warnings-as-errors Signed-off-by: Ndaquexian <daquexian566@gmail.com> * fix python2 shebang Signed-off-by: Ndaquexian <daquexian566@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Shenghang Tsai 提交于
* prevent_running_oneflow_in_forked_subprocess * add line change * IsFork => IsForkedSubProcess * auto format by CI Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 Twice 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 liufengwei0103 提交于
* fix bug int test_tensor_str * format * fix comment * fix bug to(cuda) is unavailable in cpu env Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Li Xinqi 提交于
* cuda base cpu mpi boxing * cpu_mpi * fix conflicts * add cpu mpi unittests * more checks and unittests * abstract_consistent_to_consistent_op_expr * fix compiler complaint * refactor consistent-to-consistent eager consisitent op interpreter * fix compiler complaint * refactor ConsistentToConsistentOpExpr * lazy interpreter (#5903) * fix bugs about consistent_id * more test_consistent_cast unittests * refactor functional::ToConsistent * refactor GetNdSbp * fix compiler complaints * refactor GetDevice4CurrentProcessCtx * fix error Co-authored-by: clackhan <han_binbin@163.com> Co-authored-by: Nleaves-zwx <kunta0932@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Twice 提交于
* check: fix clang-tidy-diff commands * CI: fix step names Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Shenghang Tsai 提交于
* add CMAKE_INTERPROCEDURAL_OPTIMIZATION in fast cmake cache * skip test targets of re2 Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 qq_22305325 提交于
* abstract_consistent_to_consistent_op_expr * fix compiler complaint * refactor consistent-to-consistent eager consisitent op interpreter * fix compiler complaint * refactor ConsistentToConsistentOpExpr * lazy interpreter (#5903) * fix bugs about consistent_id * refactor functional::ToConsistent * refactor GetNdSbp * Update eager_consistent_op_interpreter.cpp * Update eager_mirrored_op_interpreter.cpp * fix error * fix error * auto format by CI * Update nd_sbp.h * refine identity boxing * fix sync checkmeta error * avoid consistent id check in lazy Co-authored-by: NXinqi Li <lixinqi0703106@163.com> Co-authored-by: Nleaves-zwx <kunta0932@gmail.com> Co-authored-by: NLi Xinqi <lixinqi2010@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 Yinggang Wang 提交于
* feat(no_grad): support no_grad decorator * feat(AutogradMode): export flow.autograd_mode * feat(GradMode): export some grad_mode class * docs(GradMode): export documents * refine * docs(GradMode): export document for is_grad_enabled * auto format by CI * fix(GradMode): fix single client bug * fix bug Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-
由 Twice 提交于
* CI: enable more checkers for clang-tidy * .clang-tidy: remove cppcoreguidelines-pro-type-vararg * CI: remove duplicate checkers * CI: remove clang-analyzer-alpha.deadcode.* * .clang-tidy: add performance-* * oneflow/core/eager: remove unnecessary malloc & free * .clang-tidy: add clang-analyzer-cplusplus.* to werror * user_kernel: remove useless move * quantization_aware_training: fix move return * .clang-tidy: add google-* * CI: fix clang tidy command * CI: fix test Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 Luyang 提交于
* fix wrong names * auto format by CI * refine * auto format by CI Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com>
-
由 ZhangShen 提交于
* add some op-test * fix dims_error in my branch * Fix the bad backward kernel function by using 'cuda::atomic::Add' (#5614) * Test `nn.AdaptiveAvgPoolXd` (#5615) * Fix the bad backward kernel function by using 'cuda::atomic::Add' * Support the 'NoneType' annotation * Support objects of 'collections.abc.Iterable' as 'output_size' * Test with all cases of 'output_size' * Update adaptive_pool_gpu_kernel.cu * Skip testing `nn.AdaptiveAvgPool3d` for the current PyTorch * remove some useless test * Format TODO * Add the assertion messages for 'output_size' * Reformat codes * Remove raw tests for `flow.negative` * Remove unnecessary codes and add the assertion messages * Merge updates for 'generators.py' from master * Remove unnecessary 'random()' * Delete the separate test for `AvgPool2d` * Fix import paths * Fix import problems * Remove the PyTorch import * Denote the annotations for `tile` and `repeat` ops * Add the test for `nn.AvgPool1d` * Choose better generators for `nn.MaxPoolXd` * Randomly choose `dilation` and default values * auto format by CI * Test more kwargs for `nn.AvgPoolXd` * Add tests for `return_indices` * auto format by CI Co-authored-by: NTianyu Zhao <guikarist@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <ci-bot@oneflow.org>
-