max_to_keep specifies how many checkpoints to keep. You may wish to increase this if you have limited disk space. The default is 600, or every five minutes.
checkpoint_secs specifies the time interval for storing a checkpoint. ~/deepspeech-data/checkpoints-true-id$ lsĤ drwxr-xr-x 2 root root 4096 Feb 24 14:17. This will make them available to the Docker container. Copy or extract them to the directory you specified in your bind mount. The training data was previously prepared in the instructions for formatting data. Making training files available to the Docker containerīefore we can train a model, we need to make the training data available to the Docker container. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. Reduce learning rate on plateau (RLROP).Other useful parameters that can be passed to DeepSpeech.py.Specifying the directory that the trained model should be exported to.Different checkpoints for loading and saving.Specifying checkpoint directories so that you can restart training from a checkpoint.
Making training files available to the Docker container.Next - Testing and evaluating your trained model Previous - Setting up your DeepSpeech training environment
0 kommentar(er)
