- 21 12月, 2022 12 次提交
-
-
由 xiaoguoguo626807 提交于
* lbfgs right version * delete log * code_style * add test * delete sparse grad * add test for converage * modify example in dygraph * revert lbfgs * add test for linesearch * code_style * add test * modify test * modify test * modify set * modify set * modify test * modiy code_block; test=docs_preview * modiy code_block; test=docs_preview
-
由 niuliling123 提交于
-
由 GGBond8488 提交于
* move out sparsity from fluid * fix typro * fix circle import error * fix circle import * fix circle import * fix conflicts * remove paddle.static.sparsity * remove paddle.asp * Undo history modifications * remove sparsity in static package * modify setup.py * add missing modifications on unitest files * remove redundant sparsiy hierarchy * fix unitest file
-
由 zyfncg 提交于
* refactor conv_activation_mkldnn_fuse_pass * refactor conv_affine_channel_mkldnn_fuse_pass * fix conv_activation_mkldnn_fuse_pass * fix mkldnn unittest * refactor int8_scale_calculation_mkldnn_pass and params_quantization_mkldnn_pass * refactor conv_elementwise_add_mkldnn_fuse_pass * fix quant * refactor conv_bn_fuse_pass * fix conv_bn_fuse_pass * refactor depthwise_conv_bn_fuse_pass * fix unittest * fix conv_bn_fuse_pass * remove redundant conv2d in params_quantization_mkldnn_pass * fix params_quantization_mkldnn_pass_tester
-
由 wangxiaoning 提交于
* 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. * remove fluid APIs in distributed_strategy.py and role_maker.py * remove index sample op changes * remove fluid APIs under fleet.base * remove fluid APIs under fleet.layers.mpu * remove fluid APIs under fleet.meta_optimizers * fix fluid error * fix util_factory.py * reset fluid.io.load_inference_model API * remove dygraph.parallel.prepare_context * remove fluid.dygraph.StaticModelRunner API * remove split_lod_tensor merge_lod_tensor * remove unittests
-
由 姜永久 提交于
* rm unittests eager guard tests part7 faster2conv3d * minor change * rm post_hook test
-
由 姜永久 提交于
* rm unittests eager guard test part8 gaussian2gumnel * rm * modify * fix * fix GroupNorm param
-
由 姜永久 提交于
-
由 姜永久 提交于
* rm unittests eager guard tests part19 rnn2sparse_model * fix conflix * fix set value
-
由 姜永久 提交于
-
由 Wen Sun 提交于
Replace `assert np.close` with `np.testing.assert_allclose` in collective communication unittests (#49195) * refactor: replace `assert` with `assert_allclose` * chore: add coverage conf * revert: remove incorrect coverage conf
-
由 Aganlengzi 提交于
-
- 20 12月, 2022 10 次提交
-
-
由 wuhuachaocoding 提交于
-
由 yeliang2258 提交于
* fix onednn squeeze bug * add test * update kernel
-
由 姜永久 提交于
* remove unittests eager guard tests part2 * rm bicubic eager guard tests * review * fix * fix dtype bug in bce_loss_with_logits
-
由 姜永久 提交于
* rm eager guard tests part3_2 * modify
-
由 wanghuancoder 提交于
* delete FLAGS_enable_eager_mode
-
由 Nyakku Shigure 提交于
-
由 Yulong Ao 提交于
-
由 Ryan 提交于
-
由 sneaxiy 提交于
- 19 12月, 2022 11 次提交
-
-
由 Vvsmile 提交于
* solve conflicts of remove_lod_reset and latest develop version * add lod_reset to the no_grad_set_white_list.py
-
由 hjyp 提交于
* 移除 declarative 接口 * 修改 CI报错，清理注释
-
由 zqw_1997 提交于
* remove fluid.mul * remove mul in __all__
-
由 姜永久 提交于
* rm unittest eager guard part1 * review * reset spwan runner base * reset op_test * rm test_type_core
-
由 姜永久 提交于
* rm imperative op eager guard tests * modify varbase patch eager guard doc
-
由 zhangyikun02 提交于
-
由 Wangzheee 提交于
* General optimization for no_varlen embedding layernorm
-
由 JYChen 提交于
* add 0-d support for paddle.kthvalue * add 0-d support for paddle.mode * fix coverage test for device
-
由 xiaoguoguo626807 提交于
* move conv2d * code-style and build bug * restore c++ * code-style * rm fluid init conv2d * remove nn.conv2d * modify doc
-
由 ceci3 提交于
-
由 liu zhengxi 提交于
-
- 16 12月, 2022 7 次提交
-
-
由 HongyuJia 提交于
-
由 YuanRisheng 提交于
* move translated layer * fix build bugs * fix unittest bug * trigger ci
-
由 HongyuJia 提交于
* add xpu backward testcase * polish code * fix self.custom_op error
-
由 Vvsmile 提交于
[Clean Fluid API]Remove API: unsqueeze, use paddle.unsqueeze to replace paddle.fluid.layers.unsqueeze (#47936) * Remove API: unsqueeze Replace the funcation call. Use paddle.unsqueeze to replace paddle.fluid.layers.unsqueeze * Remove API: unsqueeze remove unsqueeze which is not used in Paddle 2.0 * Remove API: unsqueeze fix the bug by adding 'import paddle' to file tests/unittests/test_rnn_cell_api.py * Modify the call of unsqueeze from old style to new style * fix the call arguments of unsqueeze * replace unqueeze with paddle.unsqueeze * fix the error in PR-CI-Static-Check, name 'paddle' is not defined * fix the error in static in unsqueeze
-
由 kangguangli 提交于
-
由 MarDino 提交于
* cast to higher precision type to prevent fp16 diff problem * fix bilinear backward and add more unittest case
-
由 学渣戊 提交于
* fix docstring: 1. 去除 python/paddle/fluid/dygraph/nn.py 中 BatchNorm 类说明中示例代码里 numpy 的使用，为参数 data_layout 说明中 "NCHW" 里的各字母含义做了具体说明； 2. 为 python/paddle/nn/functional/norm.py 中 batch_norm 函数中参数 data_format 说明中的 "NCDHW" 里的各字母含义做了具体说明； 3. 为 python/paddle/nn/layer/norm.py 中 BatchNorm、BatchNorm1D、BatchNorm2D 以及 BatchNorm3D 类中参数 data_format 说明中的 "NCDHW" 里的各字母含义做了具体说明； 4. 修正 python/paddle/nn/layer/loss.py 中 BCELoss 类的 Shapes 说明格式错误问题，以及 BCEWithLogitsLoss 类的 Shapes 说明格式错误问题和 Examples 出错问题； 5. 修改 python/paddle/nn/functional/loss.py 中 binary_cross_entropy 函数以及 binary_cross_entropy_with_logits 函数 Returns 说明的描述，使它们与其他函数说明相似； 6. 将 python/paddle/nn/decode.py 中 BeamSearchDecoder 类 `__init__` 方法 docstring 里的参数说明提前到类说明中去。 * 根据 Review 意见进行修改。 * 修正 在线文档 中提及的错误。
-