Merge branch 'epoch_testing_fixed' of OperationSavta/SavtaDepth into master

Fixed requirements for dagshub

50 epoch training w eval

10 epoch training with evaluation

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

Merge branch 'pipeline-setup' of OperationSavta/SavtaDepth into master

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

Training for 1 epoch, still need to fix evaluation step

First finalized pipeline

Changes to almost all of the pipeline:

Revert change to dataloader that uses is_test flag.

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

reformatting the training file

update with Colab Badge

Merge branch 'simplifying-savta-depth' of OperationSavta/SavtaDepth into pipeline-setup

Merge branch 'pipeline-setup' into simplifying-savta-depth

Update '.dvc/config'

Update 'README.md'

Added a simplified version of the colab notebook, which doesnt have a clean env setup. This simplifies the setup process greatly at the cost of hurting reproducibility. Further research into ways to get a clean env on colab is required

train model on colab after fixing normalization bug

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

Add hosted storage remote and make it default

Update 'README.md'

Merge branch 'pipeline-setup' of https://dagshub.com/OperationSavta/SavtaDepth into pipeline-setup

Fixing colab after feedback

Update 'README.md'

Update 'README.md'

Update readme to include google colab setup + remove problematic packages from requirements.txt

adding more setup options to readme

removed problematic requirements

Added dvc pull instruction as dvc checkout only works locally.

Added escaping slash to run_dev_env.sh so that it works in windows as well

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

Finished training of model, saving before qualitative testing. Seems model has actually learned something. Need to add metrics and params to the pipeline.

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

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

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

Merge branch 'split_commands_in_readme' of OperationSavta/SavtaDepth into pipeline-setup

*Split commands of preparing environment in README file > *added -y for conda env creation for less interactions for the user

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

Hopefully finished with the requirements debacle, now using conda but freezing requirements with pip as usual

Fixed requirements.txt to comply with conda format

Update readme to include aws cli

Finished data import and processing setup, bug in training step

Transition to MLWorkspace docker and setup makefile with environment commands

Merge branch 'master' of https://dagshub.com/OperationSavta/SavtaDepth

Added notebook for baseline - starting conversion to python modules + pipeline

Fix to do in readme

Initial commit for setting up DS environment

added BTS demo as a qualitative baseline for SavtaDepth