提交 e8e8a8ea 编写于 作者: L LDOUBLEV

delete benchmark params of db++

上级 1d749987
...@@ -51,9 +51,3 @@ null:null ...@@ -51,9 +51,3 @@ null:null
null:null null:null
===========================infer_benchmark_params========================== ===========================infer_benchmark_params==========================
random_infer_input:[{float32,[3,640,640]}];[{float32,[3,960,960]}] random_infer_input:[{float32,[3,640,640]}];[{float32,[3,960,960]}]
===========================train_benchmark_params==========================
batch_size:8|16
fp_items:fp32|fp16
epoch:2
--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
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