bug: update_model hangs
Closed this issue · 1 comments
bobcolner commented
When using the update_model=True flag I expect to resume manual labeling. In practice the process hangs trying to read the saved training data.
Importing data ...
<ipython-input-16-ba794b895589>:1: FutureWarning: The default value of regex will change from True to False in a future version.
df_canonical = pandas_dedupe.dedupe_dataframe(df, ['Name', 'Title', 'Organization'],
Reading labeled examples from dedupe_dataframe_training.json
bobcolner commented
my bad, it eventually does resume training. Takes much 10x longer to resume vs. cold start.