Commit History

50 epoch training w eval
9066474

Dean commited on

10 epoch training with evaluation
5960c89

Dean commited on

Fixed problem with eval, metrics now make sense. This is a run with 1 epoch of training
dc053a0

Dean commited on

Finalized evaluation step, which now works. Ready to merge into master
818ec2e

Dean commited on

Training for 1 epoch, still need to fix evaluation step
ad95853

Dean commited on

First finalized pipeline
7200298

Dean commited on

Changes to almost all of the pipeline:
5ad6755

Dean commited on

Revert change to dataloader that uses is_test flag.
3410172

Dean commited on

Preparing logging for training and evaluation, and adding evaluation step [WIP]
a711240

Dean commited on

reformatting the training file
b013dfa

Dean commited on

train model on colab after fixing normalization bug
9c03436

dean commited on

remove secondary requirements (i.e. not things that are explicitly installed by the user), fix normalization problem, and use tqdm for image processing progress bar
068408a

Dean commited on

Successfully configured the dataloader and trained for one epoch. Results are not so good, but it's something. Still the Fastaiv1 looked better qualitatively
13f0309

Dean commited on

Fixed a bug in the training stage where the model was not saved, commiting before training on colab
79fd7d0

Dean commited on

Training stage seems to work, creating a non-run commit to use colab as an orchestration machine
0b86a0a

Dean commited on

Seems like we are now using the correct format for fastai2. Still there is a strange bug where the signal is killed in training
34a1202

Dean commited on

Migrated to fastai2, creating the DataLoader now works but I'm stuck on not being able to change the batch_size or num_workers as the interface seems to have changed
eeb74de

Dean commited on

Finished data import and processing setup, bug in training step
3c0c5aa

Dean commited on

Transition to MLWorkspace docker and setup makefile with environment commands
9cd8f4a

Dean commited on