✅ Using a jupyter notebook to doctest jupyter notebooks

Closes #10. Created a handy jupyter notebook with scripts to parse jupyter notebooks and execute doctests within the function calls! Added a few unit tests as a start to some (but not all) functions in data_prep.ipynb and srgan_train.ipynb. Fits with ease of use by allowing debugging in a jupyter notebook, but not (yet) compatible with code coverage solutions showing you % coverage of your code. Also note that matplotlib inline magic was removed, as plots still get plotted, and it doesn't work so well (sometimes) when you parse the .ipynb to a .py file with such magics. Created a download_to_path function in data_prep.ipynb as it's hard to create a test for the other functions requiring a big data file as input.
weiji14 · Oct 18, 2018 · 70d1eee · 70d1eee
1 parent fff29f9
commit 70d1eee
Show file tree

Hide file tree

Showing 6 changed files with 495 additions and 22 deletions.
diff --git a/Pipfile b/Pipfile
@@ -23,6 +23,8 @@ toolz = "==0.9.0"
 tqdm = "==4.27.0"
 
 [dev-packages]
+nbval = "==0.9.1"
+pytest = "==3.9.1"
 
 [requires]
 python_version = "3.6"

diff --git a/Pipfile.lock b/Pipfile.lock
diff --git a/README.md b/README.md
@@ -39,7 +39,8 @@ Also a convenient [flat file](https://en.wikipedia.org/wiki/Flat-file_database)
     ├── README.md (the markdown file you're reading now)
     ├── data_prep.ipynb (jupyter notebook that prepares the data)
     ├── environment.yml (conda packages to install, used by binder)
-    └── srgan_train.ipynb (jupyter notebook that trains the Super Resolution Generative Adversarial Network model)
+    ├── srgan_train.ipynb (jupyter notebook that trains the Super Resolution Generative Adversarial Network model)
+    └── test_ipynb.ipynb (jupyter notebook that runs doctests in the other jupyter notebooks!)
 ```
 </details>