Keras: Difference between revisions

From Hegemon Wiki
Jump to navigation Jump to search
(Created page with "[https://keras.io/callbacks/ Keras Callbacks] [https://keras.io/getting-started/faq/#how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore how-can-i-in...")
 
No edit summary
Line 1: Line 1:
[https://keras.io/callbacks/ Keras Callbacks]
[https://keras.io/callbacks/ Keras Callbacks]
[https://keras.io/getting-started/faq/#how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore]
[https://keras.io/getting-started/faq/#how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore]

[https://github.com/fchollet/keras/issues/3107 Dimension mismatch in LSTM] - Yur input should be in this format (sequences, timesteps, dimensions). So based on your example, your input should be in (None, 8, 2). Your input now is (8,2).

Revision as of 23:00, 2 January 2017

Keras Callbacks how-can-i-interrupt-training-when-the-validation-loss-isnt-decreasing-anymore

Dimension mismatch in LSTM - Yur input should be in this format (sequences, timesteps, dimensions). So based on your example, your input should be in (None, 8, 2). Your input now is (8,2).