Fixing the wrong predictions in transfer_learning.ipynb #2314
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In the Transfer_learning https://www.tensorflow.org/tutorials/images/transfer_learning,
After running the last cell for prediction on test data with fine-tuned model, the output at the end of the code cell predicts all 1, which is wrong prediction
Document:
#Apply a sigmoid since our model returns logits
predictions = tf.nn.sigmoid(predictions)
predictions = tf.where(predictions < 0.5, 0, 1)
This can be fixed by commenting/removing the part in the last cell where we apply the sigmoid function on the predictions,
#Apply a sigmoid since our model returns logits
predictions = tf.nn.sigmoid(predictions) ======> removed this line
predictions = tf.where(predictions < 0.5, 0, 1)
The original behavior is happening because we already applied a sigmoid activation in the classification head's prediction layer while we were making the model for feature extraction.
Kindly find the gist for the reference which was providing the correct output after removing/commenting the above mentioned(TF.nn.sigmoid) line. Thank you!