144 Commits (abb49df26fe32e2cd73d1c709480f58412c810ae)

Author	SHA1	Message	Date
Bai Yifan	bd8b0ebaba	fix distiller typo, test=develop (#21070)	6 years ago
juncaipeng	2c07727fb0	delete test resnet50 in post train quantization to avoid timeout error, test=develop (#21081)	6 years ago
juncaipeng	fa522dffa0	Fix bug in add_quant_dequant_pass, test=develop (#21018) ... * Fix bug for inserting add_quant_dequant_op to same variable repetitively in add_quant_dequant_pass, test=develop	6 years ago
juncaipeng	175ba39c03	Add post_training_quantization (#20800) ... * add post training quantization, test=develop * specify the quantizable op type, test=develop	6 years ago
Bai Yifan	fd5321b3f3	modify slim print precision to round(,6), test=develop (#20833)	6 years ago
whs	fa67e6e83e	Fix unitest of pruning in python3 env. (#20825) ... test=develop	6 years ago
bingyanghuang	fd49ebcbd8	update int8 benchmark with 6271 data, test=develop test=document_fix (#20736)	6 years ago
juncaipeng	f201b465ec	Move pool2d to add_quant_dequant_pass, test=develop (#20586) ... * move pool2d to add_quant_dequant_pass, test=develop	6 years ago
Bai Yifan	ffec9195e4	soft_label_distiller fix, test=develop (#20645)	6 years ago
Liufang Sang	bc4af386f4	[PaddleSlim] refine slim reader to support dataloader (#20604)	6 years ago
bingyanghuang	a5f530a8f9	Update QAT MKL-DNN readme doc (#20029)	6 years ago
Liufang Sang	86c2c362ae	fix fuse_reduce_op quantization bug (#20306) ... * fix fuse_reduce_op quantization bug test=develop * close fuse_all_reduce_ops in PaddleSlim, test=develop	6 years ago
Michał Gallus	540935a825	[Bug-fix][1.6] Improve QAT accuracy (#20174) ... * Leave fake quantization around mul * Replace Fake with Real Quantized Mul * Gather all scales from fake_quantize_ops * Enable uint8 in conv_relu tensors * Disable int8 mul and restore fake mul * Fix buf for running QAT on VGG16 and 19	6 years ago
xiteng1988	3785f08f68	fix next_tokens of controller (#20060) ... * fix next_tokens of controller	6 years ago
bingyanghuang	9de6772510	Follow comment of Merged QAT PR 18970 (#19979) ... * Follow Wangzhen's comment in PR 18970, test=develop * Review comments, test=develop * Leave fake quantization around mul test=develop * Replace Fake with Real Quantized Mul test=develop * Fix bug in quantize placement pass Nodes in the graph now have checked type instead of node name when they are to be marked for quantization test=develop	6 years ago
Bai Yifan	28a559de12	fix PyReader._iterable usage, test=develop (#20045)	6 years ago
whs	3ea2b661c0	Make PaddleSlim support PyReader (#19995) ... * Make PaddleSlim support PyReader. * Fix unittest of sensitive pruning. * Add some assert.	6 years ago
Wojciech Uss	4286a6270d	Add support for new QAT models (#18970) ... * Add support for new QAT models test=develop Co-Authored-By: Michał Gallus <michal.gallus@intel.com> Co-Authored-By: Wojciech Uss <wojciech.uss@intel.com> * fixed fps results test=develop * fix top5 accuracy drop problem * updated for new QAT models * skip quantizing average pooling - dirty but working * add missing pass * added missing conv+brelu fuse pass * removed a call to non-existent pass test=develop * renamed pass test=develop * Adjust finding pooling scale to newest QAT models * Remove unnecessary code from quantization_mkldnn_pass * Copy Pooling input scale to output scale in QAT * Refactor & remove unused code in QAT * Incorporate fp32 FC into QAT test=develop * Enable graph drawing with debug flag test=develop * Add tests for QATv2 * Fix paths for QATv2 models test=develop * Add option to save transformed int8 qat model test=develop * Remove redundant lines from qat mkldnn pass test=develop * Delegate disablement of avg pooling to qat test=develop * fix CI bug, test=develop * Follow Wangzhen's Review, test=develop * Update API.spec test=develop * Name False in (is_unsigned, TensorScale) tuple test=develop	6 years ago
jhjiangcs	766bd529d1	add optimizer:dpsgd,test=develop (#19915)	6 years ago
whs	bdb3e376d0	[PaddleSlim] Enhence compressor api in PaddleSlim (#19894) ... 1. Support customize eval function instead of eval program. 2. Fix loading checkpoint in quantization strategy. 3. Support saving eval model when saving a checkpoint. 4. Fix decoder of loading context in PaddleSlim. 5. Fix restoring from the checkpoint of uniform prune strategy. 6. Support saving eval model and infer model during training. 7. Add ‘unitest’ for saving eval model, saving infer model and uniform pruning restoring from the checkpoint. 8. Fix pruning of depthwise_conv_grad op by updating the groups.	6 years ago
juncaipeng	b0ceed6fb4	add fake_quant_dequant_op for average pool2d, test=develop (#19880) ... * add fake_quant_dequant_op for average pool2d * add test	6 years ago
Bai Yifan	e2c6bada36	Support dispensable student_loss in PaddleSlim distillation (#19824) ... * support_dispensable_student_loss, test=develop * add distillation test, test=develop * fix distillation test non convergence problem, test=develop * fix test_distillation fail problem, test=develop	6 years ago
Bai Yifan	aa63d5ac6d	Make the distillation process not save teacher variables in PaddleSlim (#19633) ... * split teacher checkpoints with student checkpoints, test=develop * add unittest for graph.merge(), test=develop	6 years ago
Chen Weihang	73daa3d6c0	Code Cleanup: delete three useless raw variables in Conv2D (#19644) ... * delete useless raw variables in Conv2D, test=develop * adjust the vars number in test_graph_wrapper to pass unittest, test=develop	6 years ago
whs	1c2aae567a	Skip start epoch and end epoch when dumping strategy in PaddleSlim (#19580) ... test=develop	6 years ago
lidanqing	ba368bf696	clean up intel labeled TODOs (#19476) ... test=develop	6 years ago
Leo Chen	6fb310ae29	Fix bug of getting bool Flags from os.environ (#19349) ... * fix bug of getting bool Flags from os.environ, test=develop * add empty loss_name in CompiledProgram for inplace grad test, test=develop	6 years ago
liu zhengxi	32598ffd8f	Python infer api update and add unit test (#19353) ... * python inference api supports numpy and add unit test, fix unit test fail in test_slim_int8_googlenet and test_slim_int8_mobilenet	6 years ago
Jiancheng Li	1799c257ad	Update Light-NAS to support latency-aware search (#19050) ... * update light_nas_strategy: add latency constraint test=develop * update light_nas_strategy: update get_model_latency test=develop * update light_nas_strategy: add more check test=develop * update light_nas test test=develop * update light_nas test test=develop * minor update light_nas test test=develop * minor update light_nas test test=develop * update light_nas test test=develop * update _constrain_func of light_nas_strategy test=develop * update _constrain_func of light_nas_strategy test=develop * remove unused code test=develop	6 years ago
Zhen Wang	0fe72469ea	Add the max-pool2d quantization support and the partial quantization support. (#19310) ... * add pool2d quantization support, only for max-pooling. * add the partial quantization support.	6 years ago
chengduo	01c7daade7	Add checking for the fetch_list of Executor.run (#18957) ... * update exe.run	6 years ago
whs	c92b78b060	Fix unitest of light nas. (#18931) ... test=develop	6 years ago
bingyanghuang	a25be53cb5	QAT int8 MKL-DNN transformation pass with MUL (#18322)	6 years ago
Wojciech Uss	8ed819d8e4	Call the test_slim_int8_* tests through absolute path (#18386) ... test=develop	6 years ago
whs	1bdfd2eb85	Fix checkpoint of Light-NAS (#18330) ... Socket can't be pickled. test=develop	6 years ago
Wojciech Uss	36b60e240a	Enable MKL-DNN for slim FP32 vs. INT8 tests (#18214) ... * Enable MKL-DNN for slim FP32 vs. INT8 tests test=develop * added test for MobileNetV1 with MKL-DNN test=develop	6 years ago
bingyanghuang	976cf46010	slim QAT MKL-DNN readme (#18187) ... * QAT doc first version * follow the comments, test=develop * update the benchmark, test=develop * Change to channel-wise, test=develop	6 years ago
翟飞跃	802ea50956	fix spelling errors (#17941) ... * fix spelling errors; test=develop * Update API.spec update md5 * Update API.spec * change the order of api;test=develop	6 years ago
翟飞跃	78441c5449	add mkldnn Int8v2 slim doc (#17909)	6 years ago
翟飞跃	c2fb9b906a	change dirname mobilenet to mobilenetv1 (#18160)	6 years ago
Wojciech Uss	ca5642c850	unify FP32 vs. INT8 comparison tests output (#18111) ... test=develop	6 years ago
Sylwester Fraczek	accb132f0f	fix slim int8 mkldnn multithreading issue (#18009)	6 years ago
chengduo	b5a1c1463d	Update CPU_NUM config (#18059) ... * update CPU_NUM config test=develop	6 years ago
Wojciech Uss	78e932862c	Added unit test for QAT FP32 & INT8 comparison (#17814) ... * added unit test for QAT FP32 & INT8 comparison test=develop * enabled other models and updated filenames test=develop * added accuracy check and multiple batch handling test=develop * removed quantization_mkldnn_pass.py test=develop * cleanup test=develop * updated model paths test=develop * renamed tests without MKL-DNN test=develop * fix reusing mkldnn pool2d primitive test=develop * add performance measuring test=develop * fix accuracy statistics test=develop * removed non-mkldnn tests test=develop * added conv2d_depthwise->conv2d mkldnn transformation test=develop * format update test=develop * fixed creating key for pool2d grad test=develop * added pass * Fix the accuracy issue while using float precision to get the scale. test=develop * Fix the format issue when 'X' is not nchw. test=develop * removed output comparing and changed number of images test=develop * cmake and comment fix test=develop * updated acc threshold for QAT comparison tests test=develop * added OMP_NUM_THREADS setting test=develop * enable all QAT INT8 tests test=develop * restored upstream version of a file test=develop * modified directory names test=develop	6 years ago
Kaipeng Deng	96ee528e3e	fix logging basicConfig cannot be setting after import paddle (#17786) ... * fix logging unable. test=develop * unset sys.stdout for stream handler. test=develop * fix newly add basicConfig. test=develop * fix import error. test=develop	6 years ago
whs	82bfa0a5ba	Enhence auto pruning and fix save/load graph when using py_reader ... 1. Enhence auto pruning. 2. Fix save/load graph when using py_reader. test=develop	6 years ago
bingyanghuang	90ebce9ead	QAT int8 MKL-DNN transformation pass (#17819)	6 years ago
翟飞跃	209a3f4e09	add log into slim mkldnn int8 test (#17926)	6 years ago
whs	5df65e506d	Add Ligth-NAS for PaddleSlim (#17679) ... * Add auto pruning strategy. 1. Fix compressor. 2. Enhence graph executor. 3. Add SAController 4. Add auto pruning strategy. 5. Add unitest for auto pruning strategy. test=develop * Init light-nas * Add light nas. * Some fix. test=develop * Fix sa controller. test=develop * Fix unitest of light nas. test=develop * Fix setup.py.in and API.spec. test=develop * Fix unitest. 1. Fix unitest on windows. 2. Fix package importing in tests directory. * 1. Remove unused comments. 2. Expose eval_epoch option. 3. Remove unused function in search_agent. 4. Expose max_client_num to yaml file. 5. Move flops constraint to on_epoch_begin function test=develop * Fix light nas strategy. test=develop * Make controller server stable. test=develop * 1. Add try exception to compressor. 2. Remove unitest of light-nas for windows. test=develop * Add comments Enhence controller test=develop * Fix comments. test=develop	6 years ago
翟飞跃	993c703bcc	INT8 MKL-DNN v2 integrate to slim (#17634) ... * refactor PR 16865 * delete mergetool files * test=develop * test=develop * test=develop * test=develop * create dir for int8 model before call SaveOptimModel * test=develop * mkldnn int8 only support linux; test=develop * refine code; test=develop * remove comment; test=develop * refine code; test=develop * fix bug; test=develop * add exception for mkldnn_post_training_strategy * reuse int8v2 CAPI dataset; test=develop * fix accuracy check bug; test=develop * remove tab * convert files to unix format * test=develop * reduce CI time;test=develop * reduce CI time and refine code;test=develop * refine comment; test=develop * add cmake FLAGS;test=develop * remove predict_num;test=develop	6 years ago
Zhen Wang	3398f99608	Adding AddQuantDequantPass for TensorRT int8 (#17529) ... * add quant_dequant_pass, test=develop * Add quant_dequant before some ops, such as the elementwise_add op. This is required by TensorRT. test=develop	6 years ago
Zhen Wang	65541d83b0	add scale pass for calculating the output scales.test=develop (#17259)	6 years ago
Zhen Wang	a40121e4c8	fix the initialization process error. test=develop (#17213)	6 years ago
wopeizl	06325c4de1	disable the test_distillation_strategy temporarily (#17227) ... test_distillation_strategy always failed on a machine with 4 gpus only, disable temporarily and need to figure out the root cause and add it back later	6 years ago
Zhen Wang	bf6b9d6d37	add checkpoint functions for graph. test=develop	6 years ago
whs	486f7d8ed6	Restore quantization and distillation stategy before loading persistables. (#16958) ... test=develop	6 years ago
whs	73c4f2b7b6	Fix distillation for soft label. (#16538) ... test=develop	6 years ago
wanghaoshuang	d41b623a72	Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into quan_ck ... test=develop	6 years ago
wanghaoshuang	6db7c2a500	Fix checkpoint of quantization.	6 years ago
Zhen Wang	f86429dbd9	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_slim_quant_bugs ... test=develop	6 years ago
whs	679a4c28fc	Fix lost of learning rate variable in distillatoin when using lr decay. (#16471) ... test=develop	6 years ago
Zhen Wang	6b854f3e1f	fix the save_in_nodes bug.	6 years ago
Zhen Wang	183bacebe3	clean codes and fix some bugs. test=develop	6 years ago
liuwei1031	8d22bc17a4	Memory optimize (#16410) ... * fix cdn issue, test=develop * fix memory optimize bugs, test=develop * fix memory optimize bugs, test=develop * remove add/sub_2 op, test=develop * disable memory_optimize by default, test=develop * disable inplace activation in python, test=develop * fix unittests, test=develop * fix unittests, test=develop * bug-fix, test=develop	6 years ago
Zhen Wang	27d05203e7	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_quan_hang ... test=develop	6 years ago
Zhen Wang	1c11f817e9	Use the resolve hazard method.	6 years ago
whs	e9bec9369b	[slim] Add quantization strategy and distillation strategy. (#16408) ... * Add fsp operator. 1 Add unitest. 2. Add python API. 3. Add layer test. * Add quantization strategy. 1. Add API. 2. Add unitest. * Add distillatoin strategy. * Add unitest config file for quantization * Fix Copyright test=develop * Fix setup.py * Fix document of layers.py. test=develop * Fix unitest in python3. test=develop * Fix documents. test=develop * 1. refine fsp op by batched gemm 2. remove unused import test=develop * Fix test_dist_se_resnext. 1. disable test distillation. 2. reset framework.py test=develop * Enable unitest of distillation after fixing Block._clone_variable test=develop * Fix cdn issue. test=develop	6 years ago
Zhen Wang	2ccbfd5e10	Fix some bugs for quantization passes.	6 years ago
whs	2e5831f0dc	[slim] Refine framework of slim and add filter pruning strategy (#16226) ... * First pr of paddle slim. 1. Add framework of paddle slim 2. Add filter pruning strategy test=develop * Rename unitest to tests. test=develop * Add prettytable into requirements. test=develop * Change in_nodes and out_nodes to odered dict. test=develop * Remove distillation. test=develop * Fix API.spec test=develop * Fix unitest. test=develop * Fix unitest in windows. test=develop * Fix unitest in windows. test=develop * Fix unitest. test=develop * Hide some functions. test=develop * Fix python import in python3.5 test=develop * Fix compress pass. test=develop * Fix unitest of test_dist_ctr. test=develop * Enhence flops. * use os.path.join * Fix pickle for python3 Fix log and comments. test=develop * 1. Remove feed_reader in compress pass 2. Fix cache reader 3. Rename CompressPass to Compressor 4. Add comments for distiller optimizer 5. Remove unused pruner currently 6. Add some comments. 7. Change API.spec test=develop * Fix pruning in python3. test=develop * Fix unitest in python3. test=develop * Fix format in python3. test=develop	6 years ago
Zhen Wang	8965819fbb	rewrite the cuda kernels of channel_wise_quant_op and channe_wise_dequant_op. test=develop	6 years ago
Zhen Wang	ec88b6cc5a	add channel wise quantization in ir pass.	6 years ago
achao2013	81b4fad8b9	add moving average absmax op and fix bug (#15155) ... * Add moving average absmax op in quantilize-aware training.	6 years ago
Zhen Wang	41b8cf0bae	Merge pull request #16162 from wzzju/fix_nan_static_quant ... Fix NaN bugs for static quantization strategy (mutil-cards train).	6 years ago
Zhen Wang	b8d1f5038e	Add the executor test for the graph clone API. test=develop	6 years ago
Zhen Wang	4ab0607634	fix NaN bugs.	6 years ago
Zhen Wang	ac6ef06ffa	Add the Clone method in Graph. test=develop	6 years ago
Zhen Wang	1b9c8d5f06	add clone function for IrGraph. test=develop	6 years ago
Zhen Wang	3cf0ee414d	update some details. test=develop	6 years ago
Zhen Wang	0cb50bb983	avoid ce fails on windows.	6 years ago
Zhen Wang	7c8f7df2fe	add some op_des funs to IrOpNode and add some var_des funs to IrVarNode. test=develop	6 years ago
Zhen Wang	33f99d6197	add IrNode&IrVarNode&IrOpNode. test=develop	6 years ago
WangZhen	28dfad5e27	fix some bugs about python3. test=develop	6 years ago
WangZhen	a7efab7ec1	add comments for public API. test=develop	6 years ago
WangZhen	0db41a9c44	add op_role attr when creating op node.	6 years ago
WangZhen	c67b29c178	fix some bugs of graph.to_program and get_pass.	6 years ago
WangZhen	c64f22048a	add convert_to_int8 pass and transform_for_mobile pass and their UTs.	6 years ago
WangZhen	c8095eeb82	add freeze pass, and UT is passed.	6 years ago
WangZhen	dde19a0ff8	add quantization freeze pass.	6 years ago
WangZhen	3b668c1574	Update some comments in the quantization transform pass. test=develop	6 years ago
WangZhen	b913463e83	Update according to the reviewers' suggestion. test=develop	6 years ago
WangZhen	59e5cc51d6	Add quantization transform pass and UT.	6 years ago
WangZhen	e2ff300b02	add UT for quantization.	7 years ago
WangZhen	451896fce4	init quantization.	7 years ago
whs	938705745e	Init paddle slim (#14834) ... * Init slim. * Remove distillation demo. * Fix import errors. test=develop * Fix some issues. test=develop * Fix configs. test=develop * Modify API.spec. test=develop * Fix format. test=develop * Fix format. test=develop * Add some comments.	7 years ago