- 18 3月, 2021 1 次提交
-
-
由 Chengmo 提交于
* fix one ps gradient clip
-
- 17 3月, 2021 1 次提交
-
-
由 Zhang Ting 提交于
-
- 16 3月, 2021 2 次提交
-
-
由 gongweibao 提交于
-
由 ronnywang 提交于
-
- 15 3月, 2021 2 次提交
-
-
由 Kaipeng Deng 提交于
* add dict/str/list supprot for DataLoader. test=develop
-
由 Chen Weihang 提交于
-
- 12 3月, 2021 4 次提交
-
-
由 Shang Zhizhou 提交于
* add serialize unittest * fix element_op trt plugin serialize bug
-
由 hong 提交于
-
由 Qi Li 提交于
-
由 Chen Weihang 提交于
* support duplicable op inout * add costom concat op test
-
- 11 3月, 2021 3 次提交
-
-
由 Aurelius84 提交于
* Fix bug with static_convert_var_shape * replace dot with dash
-
由 chajchaj 提交于
* add softmax_switch for softmax_with_cross_entropy_op, test=develop * delete using EigenMatrix in softmax_with_cross_entropy_op.h, test=develop * add REGISTER_OP_VERSION for softmax_switch attr of softmax_with_cross_entropy_op, test=develop
-
由 Aurelius84 提交于
* fix bug with jit.save * refine code
-
- 10 3月, 2021 4 次提交
-
-
由 WangXi 提交于
-
由 lilong12 提交于
* remove the send/recv of tensor size, but users have to specify the shape of the received var explicitly.
-
由 Aurelius84 提交于
-
由 furnace 提交于
* bugfix for test_cholesky_op * bugfix for test_compare_op * bugfix for lookup_table_op * bugfix for affine_channel_op
-
- 09 3月, 2021 4 次提交
-
-
由 wangguanzhong 提交于
-
由 wangguanzhong 提交于
-
由 Qi Li 提交于
-
由 Jacek Czaja 提交于
-
- 08 3月, 2021 2 次提交
-
-
由 Qi Li 提交于
-
由 Aurelius84 提交于
* auto specify PADDLE_WITH_MKLDNN and remove Interpretper * remove print * fix check abi * fix windows * fix compile flags
-
- 05 3月, 2021 1 次提交
-
-
由 liuyuhui 提交于
[Kunlun]Multi xpu dygraph performance optimization , add distributed.spawn support for multi xpu and some bug-fixes (#31130)
-
- 04 3月, 2021 6 次提交
-
-
由 liym27 提交于
-
由 Zhang Ting 提交于
-
由 Zhang Ting 提交于
* improve performance of depthwise_conv2d * add unittest
-
由 liym27 提交于
-
由 Huihuang Zheng 提交于
Fix Read-Only Attribute as while_loop Output: Usually, our convert_while_loop will be like: ``` [a, b, c] = paddle.jit.dy2static.convert_while_loop( condition_name, body_name, [a, b, c]) ``` where a, b, c are in loop_var_names. However, if loop_var_names contains property such as foo.x, we cannot assign the attribute as output of convert_while_loop because Python property is a kind of read-only attribute. To handle the case, we replace the attributes which are output of convert_while_loop with generated variables, then if we know the attribute is not read-only at runtime, we assign the attribute. The created statements are like: ``` [a, b, __attribute_variable_1] = paddle.jit.dy2static.convert_while_loop( condition_name, body_name, [a, b, foo.x]) if not isinstance(getattr(type(foo), x, None), property): foo.x = __attribute_variable_1 ```
-
由 jakpiase 提交于
-
- 03 3月, 2021 5 次提交
-
-
由 Qi Li 提交于
-
由 Pei Yang 提交于
-
由 Zhou Wei 提交于
-
由 Qi Li 提交于
-
由 wangna11BD 提交于
* add attrs deformable_groups
-
- 02 3月, 2021 2 次提交
-
-
由 Pei Yang 提交于
* add n-d input support for trt scale converter * add flatten for ut * fix dims
-
由 Gradie 提交于
* lamb_op_xpu;test=kunlun * modify lamb_op_xpu.cc;test=kunlun * delete atol lamb_op_xpu; test=kunlun * update xpu.cmake;test=kunlun * test_error 1e-5,lamb_op_xpu;test=kunlun * error1e-5,lamb_op_xpu,test=kunlun * delete atol lamb_xpu;test=kunlun * modify atol,lamb_op_xpy;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu, XPUOptest;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu,modify xpu_cmake; test=kunlun * lamb_op_xpu;test=kunlun * lamb_op_xpu,modify xpucmake;test=kunlun
-
- 28 2月, 2021 2 次提交
- 27 2月, 2021 1 次提交
-
-
由 Zhou Wei 提交于
* [Custom OP]add PD_THROW and PD_CHECK for User error message * PD_THROW and PD_CHECK, fix comment * fix Windows error message * fix Windows error message * fix CI
-