未验证 提交 3697a31e 编写于 作者: L Liufang Sang 提交者: GitHub

Fix train_eval in quantization (#682)

上级 42bd90d7
......@@ -256,8 +256,14 @@ def main():
if FLAGS.eval:
# evaluation
results = eval_run(exe, compiled_eval_prog, eval_loader,
eval_keys, eval_values, eval_cls)
results = eval_run(
exe,
compiled_eval_prog,
eval_loader,
eval_keys,
eval_values,
eval_cls,
cfg=cfg)
resolution = None
if 'mask' in results[0]:
resolution = model.mask_head.resolution
......
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