New models for softmax #355
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pobonomo
force-pushed
the
mutlinomialv12
branch
3 times, most recently
from
347a9e3 to
8042b51
Compare
pobonomo
added
safe to test
|
Hi Pierre: just wanted to check if you have an estimate for when you think this will be merged. I'd love to use it! No pressure at all. Or as you indicate it's safe to test, so I could locally clone this PR and play around with it? Thanks. |
pobonomo
force-pushed
the
mutlinomialv12
branch
from
df973ca to
16c95a9
Compare
pobonomo
added
Models
and removed
safe to test
pobonomo
force-pushed
the
mutlinomialv12
branch
2 times, most recently
from
aa2ef69 to
0c981b3
Compare
pobonomo
added
the
safe to test
Hi, I have no estimate currently. The models are working I think but it's a big change and there is a lot of cleanup. Note that what is implemented so far is using softmax for scikit-learn only: multinomial logistic regression and neural networks. I didn't implement counter-parts for Keras and Pytorch yet (the code should be very simple but there would be a lot of testing). I think that it is usable though so if you want to play with it you could try (and more testing cannot hurt). Best, |
pobonomo
force-pushed
the
mutlinomialv12
branch
from
95f7a20 to
3c35471
Compare
|
Thanks, will do! |
pobonomo
force-pushed
the
mutlinomialv12
branch
3 times, most recently
from
363b38d to
ad4c84c
Compare
pobonomo
added
safe to test
pobonomo
force-pushed
the
mutlinomialv12
branch
3 times, most recently
from
dbab218 to
327d178
Compare
pobonomo
force-pushed
the
mutlinomialv12
branch
from
a0339f4 to
70b1b78
Compare
Rebase the commit with many conflicts hopefully it is still correct - Add a notebook testing multinomial logistic regression - Do an adversarial model - Put a different revision number - Add a trivial OBBT to multinomial logistic regression notebook - Do hand scaling for MNIST with logistic - Store variables modeling exponential and sum of exponential - Try reworking arguments of logistic regression
Also add a notebook for neural network with softmax activation
Some renaming, I was calling it mixing in neural networks and affinetrans in logistic functions.
It's almost working for Logistic Regression and MLPClassifier
If predict_function is identity we don't apply any softmax
For neural network classifier use out_activation argument to specify what we want to use in last layer. For logistic use predict_function (not sure it is such a good idea yet)
Missing tests and documentation
pobonomo
force-pushed
the
mutlinomialv12
branch
4 times, most recently
from
2259241 to
f75c9e6
Compare
pobonomo
force-pushed
the
mutlinomialv12
branch
2 times, most recently
from
e569c36 to
d269a1f
Compare
Also adds a classification test probably this will be overkill for GH actions but try it.
pobonomo
force-pushed
the
mutlinomialv12
branch
from
d269a1f to
f1f1f34
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Do new models for softmax