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
History
Aapo Kyrola f84e5360cc LSTM benchmark (Caffe2 RNN based) 
Summary: Just generate some random data and put it through LSTM (Cafef2 RNN based) using its own output as gradient value for benchmark purposes. With default parameters it fits my dev GPU memory. On default parameters provided in this diff I have got 300k entries per second processed. These entries are split into blocks of seq_length * block_size. Each entry is of size hidden_dim, LSTM takes in hidden_dim sized input and produces output of the same size.

Reviewed By: salexspb

Differential Revision: D4605815

fbshipit-source-id: dd529302a0a93e8711784c67e4c777c8d6a8cdf4
2017-02-28 23:17:26 -08:00
..
docs Documenation generation to wiki 2017-02-15 16:00:44 -08:00
examples ] Char RNN bug fix for batching 2017-02-22 17:34:30 -08:00
layers Half-floats support for the rest of segment ops 2017-02-28 11:19:15 -08:00
mint goodbye old brewery 2017-01-04 20:58:35 -08:00
models Added model downloader 2017-02-22 12:47:15 -08:00
operator_test Caffe2: Tile operator 2017-02-28 23:17:26 -08:00
_import_c_extension.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
attention.py LSTMWithAttention implementation in Caffe2 2017-02-23 04:08:34 -08:00
caffe_translator.py translator fix to solve Aaron's issue 2017-02-13 11:19:13 -08:00
caffe_translator_test.py
checkpoint.py Fix issues pickling jobs 2017-02-21 20:47:27 -08:00
checkpoint_test.py Fix issues pickling jobs 2017-02-21 20:47:27 -08:00
CMakeLists.txt CMake completions work 2017-01-11 16:59:22 -08:00
cnn.py Do not initialize BN params if init_params is false. 2017-02-27 20:19:03 -08:00
context.py Make ContextManager thread-safe 2017-02-13 19:45:35 -08:00
context_test.py Make ContextManager thread-safe 2017-02-13 19:45:35 -08:00
control.py Better visualization for gpu training plan 2016-12-21 09:29:43 -08:00
control_test.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
convnet_benchmarks.py Use ws_nbytes_limit (called cudnn_ws in args). 2017-02-22 12:19:16 -08:00
convnet_benchmarks_test.py
core.py Added editDistance helper to caffe2 operators 2017-02-28 13:31:56 -08:00
core_gradients_test.py Implement EnsureDenseOp and EnsureDenseGradientOp. 2017-02-22 07:16:26 -08:00
core_test.py NextScopedBlob with well-defined behavior and respect namescope 2017-02-16 17:16:36 -08:00
data_parallel_model.py Use fbcollective 2017-02-07 10:48:33 -08:00
data_parallel_model_test.py lr_scale to loss_scale 2017-02-03 07:44:40 -08:00
data_workers.py Share queue + reduce logging 2017-02-27 19:38:45 -08:00
data_workers_test.py close blobs queues when stopping + test 2017-02-27 10:07:57 -08:00
dataio.py fix typo in TextFileReader 2017-02-21 14:02:48 -08:00
dataio_test.py NextScopedBlob with well-defined behavior and respect namescope 2017-02-16 17:16:36 -08:00
dataset.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
db_test.py Fix db_test under tsan 2016-11-29 15:18:37 -08:00
device_checker.py
dyndep.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
experiment_util.py use Pieter-MPI and fb.distributed 2016-11-29 15:18:36 -08:00
extension_loader.py
gradient_check_test.py Fix test cases: tensor of size 0 not supported by GPU ops yet. 2016-12-15 19:59:24 -08:00
gradient_checker.py
hsm_util.py Generate huffman tree 2017-01-19 16:14:23 -08:00
hypothesis_test.py support vectors with different dims in for DotProductOp. 2017-02-23 11:09:07 -08:00
hypothesis_test_util.py LSTMWithAttention implementation in Caffe2 2017-02-23 04:08:34 -08:00
introspect_vis.py Dump data for DocNN visualization 2017-02-09 13:47:07 -08:00
layer_model_helper.py Add a way do describe layers in a more AdHoc manner. 2017-02-27 23:30:39 -08:00
layer_model_instantiator.py Add model graph to dper_example 2017-02-07 13:03:54 -08:00
layers_test.py Add a way do describe layers in a more AdHoc manner. 2017-02-27 23:30:39 -08:00
load_save_test.py Add name_overrides argument to SaveOp 2017-02-16 12:32:51 -08:00
lstm_benchmark.py LSTM benchmark (Caffe2 RNN based) 2017-02-28 23:17:26 -08:00
memonger.py Fixes to topological sort, canonical blob naming, sharing final blob 2017-01-25 15:14:26 -08:00
memonger_test.py Gradient Input memory sharing using memonger blob sharing 2017-01-09 19:44:23 -08:00
mkl_test_util.py MKLDevice and MKLOperator 2016-12-15 19:59:24 -08:00
model_device_test.py Comment out NHWC Alexnet test for now 2017-01-23 13:59:29 -08:00
model_helper.py Added editDistance helper to caffe2 operators 2017-02-28 13:31:56 -08:00
mpi_python.cc Move mpi_python.cc to the python folder to be more consistent about source file locations. 2017-01-09 10:59:39 -08:00
muji.py
muji_test.py
net_builder.py Improve "reporter net" design 2017-02-21 20:17:40 -08:00
net_builder_test.py Improvements+fixes for NetBuilder 2017-01-03 16:59:24 -08:00
net_drawer.py Add model graph to dper_example 2017-02-07 13:03:54 -08:00
net_printer.py Improve "reporter net" design 2017-02-21 20:17:40 -08:00
net_printer_test.py Debug/Analysis tools for Jobs/ExecutionSteps 2017-02-06 17:31:20 -08:00
pipeline.py Better names for nets, steps and tasks 2017-02-09 16:33:54 -08:00
pybind_state.cc Added editDistance helper to caffe2 operators 2017-02-28 13:31:56 -08:00
pybind_state.h Allow PythonOp to access the workspace 2016-12-05 11:53:26 -08:00
pybind_state_gpu.cc Cudnn v6 2017-02-28 17:46:33 -08:00
pybind_state_mkl.cc Expose MKLMemory to the Python Feed and Fetch interface, and misc changes 2016-11-29 15:18:36 -08:00
python_op_test.py Allow PythonOp to access the workspace 2016-12-05 11:53:26 -08:00
queue_util.py Better names for nets, steps and tasks 2017-02-09 16:33:54 -08:00
record_queue.py
recurrent.py LSTMWithAttention implementation in Caffe2 2017-02-23 04:08:34 -08:00
schema.py Add a way do describe layers in a more AdHoc manner. 2017-02-27 23:30:39 -08:00
schema_test.py schema.Struct.__add__ 2017-02-06 13:47:58 -08:00
scope.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
scope_test.py fbsync. TODO: check if build files need update. 2016-11-15 00:00:46 -08:00
session.py Fix issues pickling jobs 2017-02-21 20:47:27 -08:00
session_test.py NextScopedBlob with well-defined behavior and respect namescope 2017-02-16 17:16:36 -08:00
sgd.py Move build_ftrl to open source directory 2017-02-16 18:02:15 -08:00
sgd_test.py Move build_ftrl to open source directory 2017-02-16 18:02:15 -08:00
sgd_test_util.py Move tests of build_sgd, build_adagrad, and build_adam to pyton directory 2017-02-14 15:32:14 -08:00
sparse_to_dense_mask_test.py Fix few more operators to handle empty batches correctly. 2016-11-29 15:18:37 -08:00
task.py Gather perf counters for distributed jobs 2017-02-21 22:06:25 -08:00
test_util.py MKL convolution operator 2017-01-23 09:59:30 -08:00
text_file_reader.py fix typo in TextFileReader 2017-02-21 14:02:48 -08:00
timeout_guard.py better killing after timeout, cleanup 2017-02-23 13:16:19 -08:00
toy_regression_test.py
tt_core.py
tt_core_test.py
utils.py Add a create your own dataset tutorial 2017-02-22 03:31:47 -08:00
visualize.py
workspace.py backup functions for non-cuda cases 2017-02-28 22:07:54 -08:00
workspace_test.py Remove redundant and failing test of FeedBlob asserts 2016-12-22 14:59:28 -08:00