You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

cc 75eec3d1f6 Post training quantization supports optimize model by fusing (#24822) ... * Post_training_quantization supports optimize model by fusing, test=develop		6 years ago
..
auto_pruning	Add Ligth-NAS for PaddleSlim (#17679)	7 years ago
configs	[PaddleSlim] Enhence compressor api in PaddleSlim (#19894)	7 years ago
distillation	Support dispensable student_loss in PaddleSlim distillation (#19824)	7 years ago
filter_pruning	Make PaddleSlim support PyReader (#19995)	7 years ago
light_nas	Update Light-NAS to support latency-aware search (#19050)	7 years ago
quantization	add config file to avoid load checkpoint test=develop (#21373)	6 years ago
CMakeLists.txt	[Fix bug] Init scale node in OutScaleForTrainingPass and enable test_quantization_scale_pass UT (#24393)	6 years ago
QAT_mkldnn_int8_readme.md	add option to exclude ops by id from quantization (#24689)	6 years ago
__init__.py	add quantization freeze pass.	7 years ago
convert_model2dot.py	Model converter to dot file (#23169)	6 years ago
mobilenet.py	[slim] Add quantization strategy and distillation strategy. (#16408)	7 years ago
qat2_int8_image_classification_comparison.py	add option to exclude ops by id from quantization (#24689)	6 years ago
qat2_int8_nlp_comparison.py	add option to exclude ops by id from quantization (#24689)	6 years ago
qat_int8_image_classification_comparison.py	Update QAT INT8 related code (#23104)	6 years ago
save_qat_model.py	add option to exclude ops by id from quantization (#24689)	6 years ago
slim_int8_mkldnn_post_training_quantization.md	Enable MKL-DNN for slim FP32 vs. INT8 tests (#18214)	7 years ago
test_auto_pruning.py	Add Ligth-NAS for PaddleSlim (#17679)	7 years ago
test_compressor.py	[PaddleSlim] Enhence compressor api in PaddleSlim (#19894)	7 years ago
test_factory.py	[slim] Refine framework of slim and add filter pruning strategy (#16226)	7 years ago
test_filter_pruning.py	[PaddleSlim] Enhence compressor api in PaddleSlim (#19894)	7 years ago
test_graph.py	Add the max-pool2d quantization support and the partial quantization support. (#19310)	7 years ago
test_graph_wrapper.py	fix test_graph_wrapper failure on cudnnv7, test=develop (#23451)	6 years ago
test_light_nas.py	Update Light-NAS to support latency-aware search (#19050)	7 years ago
test_mkldnn_int8_quantization_strategy.py	Add global value getter setter (#21285)	6 years ago
test_post_training_quantization_mobilenetv1.py	Post training quantization supports optimize model by fusing (#24822)	6 years ago
test_post_training_quantization_resnet50.py	Post training quantization supports optimize model by fusing (#24822)	6 years ago
test_qat2_int8_mkldnn_pass.py	handle conv2d activations in older QAT models (#23202)	6 years ago
test_quantization_mkldnn_pass.py	Add support for Ernie NLP model to the Slim QAT (#22506)	6 years ago
test_quantization_pass.py	Set fuse_all_reduce_ops=false for quantization test, test=develop (#23413)	6 years ago
test_quantization_scale_pass.py	Collect output scale for quantized op and fused op (#23369)	6 years ago
test_quantization_strategy.py	[PaddleSlim] Enhence compressor api in PaddleSlim (#19894)	7 years ago
test_reader.py	add config file to avoid load checkpoint test=develop (#21373)	6 years ago
test_slim_distillation_strategy.py	Support dispensable student_loss in PaddleSlim distillation (#19824)	7 years ago
test_weight_quantization_mobilenetv1.py	Weight quantization support channel_wise_abs_max method to achieve higher accuracy (#23629)	6 years ago