報錯情況如下：
File "./engine/trainer.py", line 170, in do_train
checkpointer = ModelCheckpoint(output_dir, cfg.MODEL.NAME, checkpoint_period, n_saved=5, require_empty=False)
File "/usr/local/anaconda3/lib/python3.6/site-packages/ignite/handlers/checkpoint.py", line 705, in __init__
raise ValueError(msg)
ValueError: Argument save_interval is deprecated and should be None. This argument will be removed in 0.5.0.Please, use events filtering instead, e.g. Events.ITERATION_STARTED(every=1000)
我按照錯誤提示的改成 Events.ITERATION_STARTED(every=1000)，提示.有問題，又改成 EventsITERATION_STARTED(every=1000)，還報錯，又改成 Events.ITERATION_STARTED，仍然報錯。
if EventsITERATION_STARTED is not None:
msg = (
"Argument save_interval is deprecated and should be None. This argument will be removed in 0.5.0."
"Please, use events filtering instead, e.g. Events.ITERATION_STARTED(every=1000)"
)
if EventsITERATION_STARTED == 1:
# Do not break for old version who used `save_interval=1`
warnings.warn(msg)
else:
# No choice
#raise ValueError(msg)
continue
后來想跳過這個錯誤，提示continue不能用。求各位大佬幫忙解決。感謝
轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/10112.html
標籤：人工智能技術
上一篇：flume啟動不起來
下一篇：資料恢復