Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Chapter 13, exersize 7 mistake #304

Closed
lrunaways opened this issue Oct 8, 2018 · 1 comment

Comments

@lrunaways
Copy link

@lrunaways lrunaways commented Oct 8, 2018

We make dropout layer fc1_drop:

with tf.name_scope("fc1"):
    fc1 = tf.layers.dense(pool3_flat_drop, n_fc1, activation=tf.nn.relu, name="fc1")
    fc1_drop = tf.layers.dropout(fc1, fc1_dropout_rate, training=training)

But use fc_1 building output dense layer:

with tf.name_scope("output"):
    logits = tf.layers.dense(fc1, n_outputs, name="output")
    Y_proba = tf.nn.softmax(logits, name="Y_proba")

Should be

with tf.name_scope("output"):
    logits = tf.layers.dense(fc1_drop, n_outputs, name="output")
    Y_proba = tf.nn.softmax(logits, name="Y_proba")
@lrunaways lrunaways changed the title Chapter 13 exersize 7 mistake Chapter 13, exersize 7 mistake Oct 8, 2018
@ageron

This comment has been minimized.

Copy link
Owner

@ageron ageron commented Nov 1, 2018

Thanks @lrunaways ,
I'm currently traveling, hence the long time to respond. I'll try to fix this asap.
Cheers,
Aurélien

@ageron ageron closed this in 52ad57a Aug 13, 2019
lvnilesh added a commit to beacloudgenius/handson-ml that referenced this issue Sep 11, 2019
* 'master' of https://github.com/ageron/handson-ml: (30 commits)
  In Python 3, we can use os.makedirs(..., exist_ok=True)
  Update to nltk 3.4.5 (security patch)
  Small spelling improvements to the RNN chapter
  Use fc1_drop instead of fc1, fixes ageron#304
  Update Deepnote links
  Update README.md
  Update README.md
  bst_n_estimators should be argmin(errors) + 1, fixes ageron#445
  Docker image - install PyYAML 3.13 to avoid version conflict
  Add zip, joblib, xgboost, urlextract and most TF-related packages to Docker image
  Update docker image
  Add index=housing.index when wrapping array in a DataFrame, fixes ageron#426
  Crop long outputs to make it easier to visualize the notebooks on github.com
  Remove unused init and new_saver, fixes ageron#409
  Remove unused init node, fixes ageron#398
  Remove nbextensions from installation instructions, fixes ageron#401
  Add a comment about the use of pd.cut()
  Update libraries to latest versions
  add ml-project-checklist (Appendix B of the book)
  Upgrade scikit-image
  ...
lvnilesh added a commit to beacloudgenius/handson-ml that referenced this issue Sep 21, 2019
* 'master' of https://github.com/ageron/handson-ml: (31 commits)
  fixes docker build issue
  In Python 3, we can use os.makedirs(..., exist_ok=True)
  Update to nltk 3.4.5 (security patch)
  Small spelling improvements to the RNN chapter
  Use fc1_drop instead of fc1, fixes ageron#304
  Update Deepnote links
  Update README.md
  Update README.md
  bst_n_estimators should be argmin(errors) + 1, fixes ageron#445
  Docker image - install PyYAML 3.13 to avoid version conflict
  Add zip, joblib, xgboost, urlextract and most TF-related packages to Docker image
  Update docker image
  Add index=housing.index when wrapping array in a DataFrame, fixes ageron#426
  Crop long outputs to make it easier to visualize the notebooks on github.com
  Remove unused init and new_saver, fixes ageron#409
  Remove unused init node, fixes ageron#398
  Remove nbextensions from installation instructions, fixes ageron#401
  Add a comment about the use of pd.cut()
  Update libraries to latest versions
  add ml-project-checklist (Appendix B of the book)
  ...
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
2 participants
You can’t perform that action at this time.