A repository for kaggle toxic collaboration.
All training, test and pretrained model
Developed model Naming convention for model name: model-<algo_used>--<iteration_num> ie. model-cnn-lstm-20180219-1.h5 (if keras)
Source codes or jupyter notebooks. For data analysis, use naming convention: analysis--<iteration_num>.ipynb For pipeline processing, model processing and predict: train-<model_name>.ipynb
Submission files to Kaggle Naming convention: submit-<model_name>.csv
To access the fastai library from the working directory, you'll need to create a symlink to the fastai folder. To do this, first remove the existing fastai folder that in this repo, and add the symlink using the following command:
cd src; mklink /d fastai <fastai directory>; cd ..
- e.g
cd src; mklnk /d fastai C:\Users\ikanez\Dropbox\fastai_pytorch\fastai; cd ..