Uses Travis

.travis.yml

@@ -0,0 +1,10 @@
+language: python
+python:
+  - "3.9"
+install:
+  - pip install pipenv
+  - pipenv install
+
+script:
+  # Your test script goes here
+  - pytest -v --color=yes src # Runs in pipenv

Makefile

@@ -0,0 +1,20 @@
+test: 
+	pytest
+
+lint:
+	pylint src
+
+lintfix:
+	autopep8 src --recursive --in-place --aggressive
+
+lintfixhard:
+	autopep8 src --recursive --in-place --aggressive --aggressive
+
+install:
+	pipenv install
+
+lock:
+	pipenv lock
+
+clean:
+	pipenv clean

Pipfile

@@ -0,0 +1,17 @@
+[[source]]
+url = "https://pypi.org/simple"
+verify_ssl = true
+name = "pypi"
+
+[packages]
+tensorflow = "*"
+matplotlib = "*"
+autopep8 = "*"
+
+[requires]
+python_version = "3.9"
+
+[dev-packages]
+pylint = "*"
+pytest = "*"
+autopep8 = "*"

README.md

@@ -1,30 +1,68 @@
 # Machine Learning
 Machine Learning tools, techniques, gists and projects. Some of this code is referenced in our Blog.
 
-This repository uses `conda` as an environment manager.
-The base python version is `3.9.7`. 
+This repository uses `pipenv` as an environment manager.
+The base python version is `3.9`. 
 
 ## Install dependencies
-You can follow [this instructions](https://docs.conda.io/projects/conda/en/latest/user-guide/install/index.html) 
-to install conda.
+You will need a base `python` installed in your system.
+```bash
+python --version
+```
+
+Then you will need to install `pip` and `pipenv`.
+```bash
+python -m pip install --user pip --upgrade pip
+python -m pip install --user pipenv
+```
 
-Once conda is installed run the following command to install all dependencies
-into a separate environment called `machine-learning`.
+Install the dependencies with 
 ```bash
-conda create --name machine-learning --file requirements.txt
+make install
+```
+This calls `pipenv`, which will create a virtual environment for this project.
+
+To enter a shell on this environment run
+```bash
+make shell
 ```
 
 ## Tests
 Run the entire test suite with `pytest`.
+Use
+```bash
+make test
+```
+
+## Code Style
+We use PEP8 as a style guide for python code.
+
+Check lint errors with
+```bash
+make lint
+```
+
+We use `autopep8` to automatically fix errors.
+
+User
+```bash
+make lintfix
+```
+of
+```bash
+make lintfixhard
+```
 
+for in-place fixing of lint errors under the `/src` dir.
 
 
 ## Contribution
 1. Create a new feature branch that compares to the main branch and open a PR.
-1. Ensure you have written appropriate tests and run `pytest`
+1. Ensure you have written appropriate tests and they are passing.
+1. Ensure the code passes the style guide conventions.
 
 Update the `requirements.txt` file using the following command from the 
 main directory:
 ```bash
-conda list -e > requirements.txt
+make lock
 ```

path_adder.py

@@ -17,7 +17,7 @@ def main():
     file_name = 'extra_python_folders.pth'
     with open(os.path.join(site_packages_dir, file_name), 'w') as f:
         f.write(src_path)
-        print("Added to conda path: " + src_path)
+        print("Added to path: " + src_path)
 
 if __name__ == "__main__":
     main()

src/computer_vision/cnn/__init__.py

@@ -1 +1 @@
-from .fashion_mnist_classifier import FashionMNISTClassifier
+from .fashion_mnist_classifier import FashionMNISTClassifier

src/computer_vision/cnn/fashion_mnist_classifier.py

@@ -9,6 +9,7 @@
 # Support packages
 import pickle
 
+
 class FashionMNISTClassifier:
     def __init__(self, model_path, train_history_path):
         self.model_path = model_path

src/computer_vision/cnn/fashion_mnist_classifier_test.py

@@ -5,23 +5,24 @@
 
 tensorflow.random.set_seed(123)
 
+
 def test_full_cycle():
     classifier = FashionMNISTClassifier("./model.h5", "./hist")
     classifier.build_model()
     classifier.load_dataset()
-    classifier.train_data = ( # just use 300 data samples for unit testing
+    classifier.train_data = (  # just use 300 data samples for unit testing
         classifier.train_data[0][:300],
         classifier.train_data[1][:300]
     )
 
     classifier.compile()
     classifier.train(128, 2, 0.1)
 
-    assert classifier.model # The model is accessible
-    assert classifier.train_hist # The training history is accessible
-    assert classifier.train_hist.get('accuracy') # The accuracy was recorded
+    assert classifier.model  # The model is accessible
+    assert classifier.train_hist  # The training history is accessible
+    assert classifier.train_hist.get('accuracy')  # The accuracy was recorded
 
     accuracy = classifier.train_hist['accuracy']
 
-    assert len(accuracy) == 2 # There is one measurement for every epoch
-    assert accuracy[0] < accuracy [1] # The second epoch is an improvement
+    assert len(accuracy) == 2  # There is one measurement for every epoch
+    assert accuracy[0] < accuracy[1]  # The second epoch is an improvement