Keras Model.fit Verbose Formatting

Question

I\'m running Keras model.fit() in Jupyter notebook, and the output is very messy if verbose is set to 1:

    Train on 6400 samples, validate on 800 samples
    E         


        

Answer 1

Took me a while to see this but tqdm (version >= 4.41.0) has also just added built-in support for keras so you could do:

from tqdm.keras import TqdmCallback
...
model.fit(..., verbose=0, callbacks=[TqdmCallback(verbose=2)])

This turns off keras' progress (verbose=0), and uses tqdm instead. For the callback, verbose=2 means separate progressbars for epochs and batches. 1 means clear batch bars when done. 0 means only show epochs (never show batch bars).

If there are any issues with it please feel free to post on https://github.com/tqdm/tqdm/issues