saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-15 21:00:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
pytorch/caffe2/python/layers
History
Tiangao Gou bc50510016 use numerically stable version of BatchLRLoss 
Summary: change all use cases of BatchLRloss to the numerically stable version. This includes the uses of function build_loss defined in fbcode/caffe2/caffe2/fb/dper/layer_models/loss.py and class BatchLRLoss defined in fbcode/caffe2/caffe2/python/layers/batch_lr_loss.py.

Reviewed By: xianjiec

Differential Revision: D6643074

fbshipit-source-id: b5678556b03cbdd380cab8a875974a87c33d7f12
2018-01-02 13:18:36 -08:00
..
__init__.py Re-license to Apache 2017-09-28 16:22:00 -07:00
add_bias.py Re-license to Apache 2017-09-28 16:22:00 -07:00
arc_cosine_feature_map.py Re-license to Apache 2017-09-28 16:22:00 -07:00
batch_distill_lr_loss.py Distill loss with SigmoidCrossEntropyWithLogits 2017-10-26 15:18:34 -07:00
batch_lr_loss.py use numerically stable version of BatchLRLoss 2018-01-02 13:18:36 -08:00
batch_mse_loss.py Support regression with output transform in MTML for feed 2017-12-11 17:20:20 -08:00
batch_normalization.py Re-license to Apache 2017-09-28 16:22:00 -07:00
batch_sigmoid_cross_entropy_loss.py Re-license to Apache 2017-09-28 16:22:00 -07:00
batch_softmax_loss.py Re-license to Apache 2017-09-28 16:22:00 -07:00
build_index.py Re-license to Apache 2017-09-28 16:22:00 -07:00
concat.py higher order interaction of embeddings 2017-11-30 08:51:09 -08:00
conv.py Re-license to Apache 2017-09-28 16:22:00 -07:00
dropout.py Re-license to Apache 2017-09-28 16:22:00 -07:00
fc.py Re-license to Apache 2017-09-28 16:22:00 -07:00
fc_without_bias.py Re-license to Apache 2017-09-28 16:22:00 -07:00
feature_sparse_to_dense.py Use EMBEDDING feature type instead of FLOAT_TENSOR 2017-10-11 13:50:03 -07:00
functional.py Revert D6026557: [caffe2][PR] Fix "No handlers could be found for logger" 2017-10-12 20:21:52 -07:00
gather_record.py Re-license to Apache 2017-09-28 16:22:00 -07:00
last_n_window_collector.py Re-license to Apache 2017-09-28 16:22:00 -07:00
layers.py print exception in layers 2017-12-15 12:12:28 -08:00
margin_rank_loss.py Add rank loss for retrieval models with random negative sample 2017-10-25 16:19:41 -07:00
merge_id_lists.py Re-license to Apache 2017-09-28 16:22:00 -07:00
pairwise_dot_product.py Fix dot processor with only one sparse feature and no dense feature 2017-12-14 13:05:17 -08:00
position_weighted.py Revert D6026557: [caffe2][PR] Fix "No handlers could be found for logger" 2017-10-12 20:21:52 -07:00
random_fourier_features.py Re-license to Apache 2017-09-28 16:22:00 -07:00
reservoir_sampling.py Re-license to Apache 2017-09-28 16:22:00 -07:00
sampling_train.py Re-license to Apache 2017-09-28 16:22:00 -07:00
sampling_trainable_mixin.py Re-license to Apache 2017-09-28 16:22:00 -07:00
select_record_by_context.py Revert D6026557: [caffe2][PR] Fix "No handlers could be found for logger" 2017-10-12 20:21:52 -07:00
semi_random_features.py Re-license to Apache 2017-09-28 16:22:00 -07:00
sparse_feature_hash.py add option to do simple modulo 2017-11-10 13:49:07 -08:00
sparse_lookup.py Enable SparseLengthsMean 2017-11-30 16:04:38 -08:00
split.py Re-license to Apache 2017-09-28 16:22:00 -07:00
tags.py Tagging sparse parameters 2017-10-04 18:46:48 -07:00
uniform_sampling.py Re-license to Apache 2017-09-28 16:22:00 -07:00