- 22 6月, 2021 3 次提交
-
-
由 zhiboniu 提交于
-
由 niuliling123 提交于
-
由 feng_shuai 提交于
-
- 21 6月, 2021 5 次提交
-
-
由 lidanqing 提交于
* Add oneDNN AXPY handler. * Add fallback for small tensors. * Fix ifdefs * Remove unnecessary namespace prefixes and add missing headers. * Guard handler_axpy with proper ifdefs. * Compilation of this function is possible only when Paddle is not build with CUDA nor HIP. * Move AXPY handler code to separate files. * Use oneDNN AXPY handler in SGD op. * Use axpy handler only when Paddle is built with oneDNN. * Add test for SUM BF16 with big rows. * Fix SFINAE rules for elementwise_add_to. * Add test case for SGD with big rows. * update * update Co-authored-by: NAdam Osewski <adam.osewski@intel.com>
-
由 pangyoki 提交于
* use BatchMatMul * replace TensorCopy with ShareDataWith * remove check fp16 grad * fix format * add grad_check * fix grad check
-
由 cc 提交于
* Combine amp and qat * add unit test
-
由 Leo Chen 提交于
* enable npu alignment * support flatten_params/grads * support clip by global norm * remove memset in coalesce_tensor_op * fix npu kernel of sum op when input is one tensor * add ut for flatten_param_grads+regularizer * fix ut * fix typo
-
由 李季 提交于
-
- 18 6月, 2021 3 次提交
-
-
由 Wangzheee 提交于
-
由 feng_shuai 提交于
* batch_norm_act_fuse_pass_init * repair the unittest of batch_norm_act
-
由 王明冬 提交于
-
- 17 6月, 2021 4 次提交
-
-
由 王明冬 提交于
-
由 joanna.wozna.intel 提交于
* Add bf16 support for save and load ops * Add bf16 test condition * Add matmul and chagne fluid.io to paddle.static * Reduce the test duration
-
由 ronnywang 提交于
* add atan2_op * fix
-
由 joanna.wozna.intel 提交于
* Add lookup_table_v2 BF16 * Reuse lookup table UT * Change op_type to op_version * Remove check_dygraph * Remove skip_check_grad_ci
-
- 16 6月, 2021 6 次提交
-
-
由 Jacek Czaja 提交于
* - Draft of implementation of refactoring - compilation fix * - Fixes after review * - Removed unnecessary comment
-
由 王明冬 提交于
-
由 Zhou Wei 提交于
-
由 zhangbo9674 提交于
* new api trunc, test=develop
-
由 Wangzheee 提交于
-
由 王明冬 提交于
-
- 15 6月, 2021 7 次提交
-
-
由 Leo Chen 提交于
* use SparseSoftmaxCrossEntropyWithLogits * fix * test_slice * revert test_slice * add backprob for npu kernel * fix typo * fix ut * fix ut * refine comments * return softmax
-
由 王明冬 提交于
-
由 Jiawei Wang 提交于
* add conv3d prototxt * Update conv3d.pbtxt * Update conv3d.pbtxt
-
由 Shang Zhizhou 提交于
* 1, remove layernorm dynamic fp16; 2, let reshape out in dynamic shape * remove useless code
-
由 wawltor 提交于
add the support for the bool in compare ops
-
由 jiangcheng 提交于
* add reduce_sum_op by add self-kernel * set all ReduceKernel MPType for accuracy * add float16 test script which input is integer number * solve reduce sum float16 check_grad problem * solve conflict and change test script for CI * change kernel register for CI * remove all useless template
-
由 zyfncg 提交于
* Add digamma_op and unittest * add digamma_op api * remove special DigammaCudaKernel and correct some docs * remove unused headers * fix api doc error
-
- 12 6月, 2021 3 次提交
-
-
由 zhiboniu 提交于
* Eliminate numerical differences of LayerNorm; fix LayerNorm Nan Bug while large data input * fix bug while large shape of data input
-
由 limingshu 提交于
-
由 joanna.wozna.intel 提交于
* Small changes related to BF16 fusion_gru and fusion_lstm * Correct to pass arg by value * Add conditions to rnn op * Correct the spelling mistake * Improving the test with checking activation * Trigger CI
-
- 11 6月, 2021 2 次提交
- 10 6月, 2021 4 次提交
-
-
由 Baibaifan 提交于
-
由 lilong12 提交于
-
由 Chen Weihang 提交于
* support diff dataset tensor place in single process dataloader * fix unittest failed
-
由 wangguanzhong 提交于
-
- 09 6月, 2021 3 次提交