search

rhiever / sklearn-benchmarks

A centralized repository to report scikit-learn model performance across a variety of parameter settings and data sets.
MIT License
210 stars 54 forks source link

Learning Algorithms #6

Closed AkshayVarik closed 8 years ago

AkshayVarik commented 8 years ago

These are 9 algorithm's codes here. I am currently working on the SVC code. Also there is no model score (result data) pushed and all these have been written for the dataset: car-evaluation. Note: Linear SVC- l1 penalty has been removed. I have run all the codes, except for GradientBoosting, which is extremely slow. RandomForrest also took nearly 4 hours to run. The others were pretty fast. Lastly, I get some error when running SGDClassifier. I will run it again and try to fix it. Nevertheless I am pushing the code here.

rhiever commented 8 years ago

Thank you @AkshayVarik! I'm going to merge this PR then move all of the notebooks into a notebooks directory. After I do this, please make sure to sync the latest version of this repo to your fork.

AkshayVarik commented 8 years ago

Thanks @rhiever! I just did that!