UnimplementedError: Cast string to float is not supported [[node Cast (defined at C:/Users/User/.spyder-py3/LSTM.py:132) ]]

Question

Creating error, but I can find what string they are referring to. line 132: callbacks=[tensorboard, checkpoint],

filepath = "LSTM_Final-{epoch:02d}" # unique file name that will include the epoch and the validation acc for that epoch
checkpoint = ModelCheckpoint("models\\{}.model".format(filepath, monitor='val_acc', verbose=1, save_best_only=True, mode='max')) # saves only the best ones

# Train model
history = model.fit(
    np.array(train_x), np.array(train_y),
    batch_size=BATCH_SIZE,
    epochs=EPOCHS,
    validation_data=(np.array(validation_x), np.array(validation_y)),
    callbacks=[tensorboard, checkpoint],
)

that's clearly not helpful. But try using val_acc instead of val_loss. — Stuart
– Stuart, Commented Mar 18, 2020 at 1:05

Stuart · Accepted Answer · 2020-03-18 00:26:53Z

1

You didn't provide the code for ModelCheckpoint() but I'm going to assume that it takes a few parameters.

Edit: OP commented that it's from keras.callbacks.callbacks.ModelCheckpoint(). The documentaiton says this should fix it.

Note from documentation: filepath can contain named formatting options, which will be filled with the values of epoch and keys in logs (passed in on_epoch_end).

filepath = "LSTM_Final-{epoch:02d}-{value_acc:.2f}"
checkpoint = ModelCheckpoint(filepath, monitor='val_acc', verbose=1, save_best_only=True, mode='max')

edited Mar 18, 2020 at 0:26

answered Mar 16, 2020 at 23:40

Stuart

4843 silver badges7 bronze badges

Sign up to request clarification or add additional context in comments.

9 Comments

mystack Over a year ago

from my research the filepath string is used in format() and doesn't need an f for f-string

Stuart Over a year ago

@mystack you're not calling format on "LSTM_Final-{epoch:02d}".

Stuart Over a year ago

@mystack what is this line doing?

checkpoint = ModelCheckpoint("models\\{}.model".format(filepath, monitor='val_acc', verbose=1, save_best_only=True, mode='max'))

? What do you want the first parameter to look like?

mystack Over a year ago

It creates a file in folder models called LSTM_Final-(the # of the epic that it is).model Is that what you were asking of?

Stuart Over a year ago

@mystack okay, I think I understand the issue. Updated.

|

mystack · Accepted Answer · 2020-04-03 22:17:19Z

0

The problem was that the data is labeled and the code will only accept unlabeled data. Unfortunately this was not indicated well by the error log.

answered Apr 3, 2020 at 22:17

mystack

9510 bronze badges

Collectives™ on Stack Overflow

UnimplementedError: Cast string to float is not supported [[node Cast (defined at C:/Users/User/.spyder-py3/LSTM.py:132) ]]

2 Answers 2

9 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

9 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related