提交 483c0383 编写于 作者: R root 提交者: Tingquan Gao

tipc: set bs=256 to avoid oom

上级 b9de09e5
...@@ -20,7 +20,7 @@ distill_train:null ...@@ -20,7 +20,7 @@ distill_train:null
null:null null:null
null:null null:null
## ##
===========================eval_params=========================== ===========================eval_params===========================
eval:tools/eval.py -c ppcls/configs/ImageNet/PPLCNet/PPLCNet_x2_5.yaml eval:tools/eval.py -c ppcls/configs/ImageNet/PPLCNet/PPLCNet_x2_5.yaml
null:null null:null
## ##
...@@ -50,10 +50,10 @@ inference:python/predict_cls.py -c configs/inference_cls.yaml ...@@ -50,10 +50,10 @@ inference:python/predict_cls.py -c configs/inference_cls.yaml
-o Global.benchmark:False -o Global.benchmark:False
null:null null:null
===========================train_benchmark_params========================== ===========================train_benchmark_params==========================
batch_size:512 batch_size:256
fp_items:fp32|fp16 fp_items:fp32|fp16
epoch:1 epoch:1
--profiler_options:batch_range=[10,20];state=GPU;tracer_option=Default;profile_path=model.profile --profiler_options:batch_range=[10,20];state=GPU;tracer_option=Default;profile_path=model.profile
flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98;FLAGS_conv_workspace_size_limit=4096 flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98;FLAGS_conv_workspace_size_limit=4096
===========================infer_benchmark_params========================== ===========================infer_benchmark_params==========================
random_infer_input:[{float32,[3,224,224]}] random_infer_input:[{float32,[3,224,224]}]
\ No newline at end of file
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册