-
Notifications
You must be signed in to change notification settings - Fork 101
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
Modify tfidf_transformer to enable custom vocabulary and approximate sublinear-tf scaling without sparse containers #777
Conversation
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
* Make opset 15 default * fix missing target opset in polynomial features Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
Signed-off-by: adam444555 <a473489548@gmail.com>
…onnx#773) * Update a training value in a failing pipeline used in a unit test * upgrade version Signed-off-by: adam444555 <a473489548@gmail.com>
onnx#772) * Enable RandomForestClassifier in converter for CalibrationClassifierCV * remove unused variable Signed-off-by: adam444555 <a473489548@gmail.com>
* Implements option zipmap for MultiOutputClassifier Signed-off-by: adam444555 <a473489548@gmail.com>
fix code length Signed-off-by: adam444555 <a473489548@gmail.com>
ad9281b
to
3cc1a6f
Compare
#777 (comment): sorry I was not clear enough, I just meant adding something like |
13edd88
to
2a99400
Compare
Signed-off-by: adam444555 <a473489548@gmail.com>
A specific commit could not be signed off (I tried with many approaches but no one worked, and I had no idea why). Therefore I rebase the branch to drop the problemetic commit, and then recommit the same change. |
Sorry for the trouble. Let me know if it is ready to be merged. |
Yes, it is ready right now! |
The
stop_words_
attribute does not exist if custom vocabulary is provided, resulting in AttributeError. Fix it byhasattr
check.Approximate the sublinear_tf scaling by: first add all coefficient (included null coefficient) by 1 and then take log, i.e. replace tf with log(1+tf). Null coefficient remains to be 0 after the operation.