feature/mul converter #3

- Finished draft of pooling reusing of operators - Using gethash in PoolGrad added - Removed diagnostic - Added pool mkldnn grad reusing of primitives - Added diagnostic - Removed diagnostic - added dependency to mkldnn data type for pooling mkldnn - Added mkldnn memory data type determining based on template type of op - Compilation warning fix - codying style fixes

…e_parallel_executor

use cdn to speed up thirdparty package download

refine mkldnn cmake with official commit id

Fix timeline for the number of profile is one

… develop

…OFF. (PaddlePaddle#10756)

… develop

* listen_and_serv use localscope * fix ut

fix roi_pool op bug

…port-multi-gpu Inferencer support parallel_executor

…Paddle#10762)

allow inference test to generate timeline

…/mul_converter

…le#10760) * use latest pip version * fix ci * update by comment

Fix a profiler race condition

…0778) * Dockerfile: unify dev and product docker environment. * Fix check style error. * Use envrioment valuable to get version info.

… feature/refine_parallel_executor

* package prior_box op * add doc * add unittest * add unittest * fix CI fails

Fix a compile error

…reuse Reuse of pooling mkldnn primitives

add title for cpu_profiling.md

…Paddle#10715) * "a piece of job." * "fix typeo" * "fix ci"

* Added kernel to is_empty_op * Added python API * Updated code as required * Updated unittests

* fix wheel package name * delete cuda7.5 cudnn5

… InferVarType to find var recursively

add version and cmakecache in inference_lib

Add pip install description

…addlePaddle#10780)

…dle#10812)

add cpp trainer lib and demo

…le#10796) 1. If all bboxes are not difficult ground truth, the users can not define the data layer for this flag and not the input can be None for detection_map API. 2. Set default value for aspect_ratios in prior_box API. 3. Add more check in prior_box API.

…-improvments Update activations for MKL-DNN

… dev_CustomReader

… fix_warpctc_op

…0819)

Experiment on vgg flower, 2 trainers, 1ps. more trainer could have more speedup. After: Pass = 0, Iters = 327, Speed = (7.52) img/s Before: Pass = 0, Iters = 385, Speed = (6.77) img/s

…ferVarType to find var recursively

… feature/support_op_role

enable MKLDNN inference test

)

fix warpctc

… feature/refine_parallel_executor

…_executor

change CMAKE_INSTALL_PREFIX in inference_lib_dist to FLUID_INSTALL_DIR

…dlePaddle#10835)

update bcebos url so that it can also be downloaded in inner network

…network"

…n_tests Force some unittests serial

… feature/support_op_role

…/mul_converter

* Add dice loss. * Fix comments. * Remove unused code.

… feature/refine_parallel_executor

…#10848)

…le#10849)

…lePaddle#10850)

… dev_CustomReader

…update-becbos-url Revert "update bcebos url so that it can also be downloaded in inner network"

* Enhance reduce op for multi dims. * Uncomment some unitest. * Uncomment unitest. * Remove unused code. * Fix infershape and python wrapper. * Add more examples. * Fix l2_normalize. * Fix normalization_wrapper. * Polish code. 1. Rename unitest function. 2. Rename const variable.

…t_file_conflict fix dist test

Add the demo of subgraph splitter

send use pinned memory

Add api docs for RMSProp and modelAverage

Add Inference doc for fluid

… feature/refine_parallel_executor

…/mul_converter

…m_gen expose random_data_generator

* wip integrate benchmark scripts * testing nlp models * k8s script to start dist benchmark job * update script * done support all models * add README.md * update by comment * clean up * follow comments

Add groups for conv transpose ops

add checkpoint util class and implement

multi-thread handlerequest

…addlePaddle#10817) * add lod_tensor utility python module * add lod_tensor test code * add more lod tensor tests * modify word2vec example code using new api * add comment

Fix cuda9 prod image

fix typo in workflow_of_capi.md

* Expose bilinear operator into Python API * delete unused lines * delete unused lines * fix typos * fix name arg and typos

…_role Add `op_role` into OpDesc.

…allel_executor Disable and fix tests on multi devices.

…OpInferVarType Fix ElementwiseOpInferVarType in elementwise_op

…/mul_converter

refine fluid deploy on teamcity

…is used

…float16 and/or GPU

…fall back to default impl

…ddle#10757) * Update label role implementation to use new API * Try trainable embedding layer * Fix feed order * Add infer call

Fix rst format issues in several files.

…i_machine_translation Simplify Machine Translation demo by using Trainer API

…ment_lod Modify understand sentiment example using new LoDTensor API

…oles_lod Simplify label_semantic_roles book example using new LoDTensor API

…d-elementwise-add Blas optimized elementwise_add forward and backward passes

…lint_while_compile Remove cpplint in cmake

…erationSpeed Improve doc generation speed

* fix range in data flow analysis

…ld_cluster_benchmark_script remove old fluid cluster benchmark scripts

CustomReader

…stand_sentiment_serial Disable unstable test

…imizer_tests Change optimizer to old paddle style

add inference api demo impl

fix rename var

…le#10925) * unify UpsamplingBilinear2d interface specification * unify UpsamplingBilinear2d interface specification * fix name conventions * small fix about computation order

* scripts: clean bash scripts. * Fix build related documents.

…xample (PaddlePaddle#10946) * add create lodtensor from list * modify book example

* Add quad transform. * Fix some syntax error. * Fix CUDA kernel launch configure. * Generalize geometry channels. * Rename QuadTransform to PolygonRestore. * Rename op. * Rename op and fix computation. * Modify CMakeLists.txt for box_restore op. * Refine code: 1. rename op 2. uncomment unitest on GPU

…/mul_converter

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feature/mul converter #3

feature/mul converter #3

Commits on May 17, 2018

Commits on May 18, 2018

Commits on May 19, 2018

Commits on May 20, 2018

Commits on May 21, 2018

Commits on May 22, 2018

Commits on May 23, 2018

Commits on May 24, 2018

Commits on May 25, 2018

Commits on May 26, 2018

Commits on May 27, 2018