modified: .flake8

	new file:   code/.gitignore
	modified:   code/training_pipeline/.gitignore
	modified:   code/training_pipeline/Makefile
	modified:   code/training_pipeline/notebooks/POC.ipynb
	modified:   code/training_pipeline/requirements/dev.in
	modified:   code/training_pipeline/requirements/dev.txt
	modified:   code/training_pipeline/requirements/prod.in
	modified:   code/training_pipeline/requirements/prod.txt

.flake8

@@ -18,8 +18,6 @@ extend-ignore =
     D200,D205,D400,D401,  # whitespace issues and first line content
   # DAR - flake8-darglint docstring correctness check
     DAR103,  # mismatched or missing type in docstring
-application-import-names = app_gradio,text_recognizer,tests,training
-  # flake8-import-order: which names are first party?
 import-order-style = google
   # flake8-import-order: which import order style guide do we use?
 docstring-convention = numpy
@@ -37,6 +35,4 @@ mypy-init-return = true
 ;   */__init__.py:F401,I
 ;   */data/*.py:DAR
 ;   data/*.py:F,I
-;   *text_recognizer/util.py:DAR101,F401
-;   *training/run_experiment.py:I202
-;   *app_gradio/app.py:I202
+

code/.gitignore

@@ -0,0 +1,3 @@
+data/downloaded
+data/processed
+data/interim

code/training_pipeline/.gitignore

@@ -13,38 +13,9 @@ __pycache__
 # notebooks
 .ipynb_checkpoints
 *.nbconvert*.ipynb
-.notebook_test.sh
-
-# Distribution / packaging
-.Python
-env/
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-*.egg-info/
-.installed.cfg
-*.egg
 
 # logging
 wandb
-*.pt
-*.ckpt
-lightning_logs/
-logs
-*/training/logs
-*/training/*sweep.yaml
-flagged
 
 # Misc
-.env
-.mypy_cache
-/requirements.txt
 requirements/dev-lint.txt

code/training_pipeline/Makefile

@@ -10,6 +10,9 @@ conda-update:
 pip-tools:
 	pip install pip-tools==7.4.1 setuptools==70.3.0
 	pip-compile requirements/prod.in && pip-compile requirements/dev.in
+	pip install transformers
+	pip install datasets
+	pip install wandb
 	pip-sync requirements/prod.txt requirements/dev.txt
 
 # Compile and install the requirements for local linting (optional)

code/training_pipeline/requirements/dev.in

@@ -10,10 +10,8 @@ pytest-cov
 scipy
 toml
 spacy
-transformer
 pandas
 nltk
-datasets
 # versioned to give pip hints
 ipywidgets
 coverage[toml]==6.4
@@ -22,8 +20,4 @@ pytest-cov==3.0.0
 notebook>=6.5,<6.6
 nbconvert>=6.5,<6.6
 seaborn>=0.13,<0.14
-tornado>=6.3,<6.4
 # versioned to improve stability
-pytorch-lightning==1.6.3
-torchmetrics<0.8
-wandb==0.12.17