easy_transfer_learning notebook not working?

[hammondm]

all

Am I correct that the easy_transfer_learning notebook. in /notebooks does not work?
I've tried it on both my own machine and on Google Colab and am getting the same error.

mike h

---

I Performing dummy training to check for memory problems.
I If the following process crashes, you likely have batch sizes that are too big for your available system memory (or GPU memory).
I Loading best validating checkpoint from english/coqui-yesno-checkpoints/best_dev-1909
I Loading variable from checkpoint: beta1_power
I Loading variable from checkpoint: beta2_power
I Loading variable from checkpoint: cudnn_lstm/rnn/multi_rnn_cell/cell_0/cudnn_compatible_lstm_cell/bias

---------------------------------------------------------------------------

ValueError                                Traceback (most recent call last)

[<ipython-input-9-4d9feeb56b9c>](https://localhost:8080/#) in <module>()
      4 os.environ["CUDA_VISIBLE_DEVICES"] = "0"
      5 
----> 6 train()

9 frames

[/usr/local/lib/python3.7/dist-packages/tensorflow_core/python/client/session.py](https://localhost:8080/#) in _run(self, handle, fetches, feed_dict, options, run_metadata)
   1154                 'Cannot feed value of shape %r for Tensor %r, '
   1155                 'which has shape %r' %
-> 1156                 (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))
   1157           if not self.graph.is_feedable(subfeed_t):
   1158             raise ValueError('Tensor %s may not be fed.' % subfeed_t)

ValueError: Cannot feed value of shape (256,) for Tensor 'cudnn_lstm/rnn/multi_rnn_cell/cell_0/cudnn_compatible_lstm_cell/bias/Initializer/Const:0', which has shape '(128,)'

[hammondm]

all
I figured this out. There are a couple of lines commented out in the notebook that need to be commented in.

1. n_hidden should be 64.
2. drop_source_layers=1 needs not to be commented out.
3. On Colab, load_cudnn=True, can be commented out, but I needed it in for docker.

mike h.