5

stream did not block host until done; was already in an error state how to fix it using tensorflow keras

model.compile(
optimizer="adam",
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['acc'])    
model.fit(X_train_scaled, y_train, epochs=5)

this error appears

enter image description here

enter image description here

mero
  • 179
  • 1
  • 8
  • 1
    Instead of image can you include full `Traceback` in your question. So that it will give more visibility on the issue. And also provide more reproducible code. Thanks! –  May 18 '21 at 05:55
  • "was already in an error state" sounds like you better reset your variables. – till Kadabra Feb 11 '22 at 11:14

1 Answers1

0

Restarting the notebook fixed this issue for me. If this doesn't fix it, you can try updating the GPU drivers.
Also, a similar issue is available on GitHub : Link to the Issue

Akhil Shukla
  • 61
  • 6