README

System Requirements

Download Miniconda from https://conda.io/miniconda.html
Install with bash Miniconda3-latest-MacOSX-x86_64.sh
Check Miniconda executable installed location which conda => /Users//miniconda3/bin/conda
Load pre-configured Python environment dependencies into Miniconda conda env create -f ./config/aind-environment-unix.yml
Alternatively if using pip, install dependencies as follows: pip install -r ./config/requirements.txt
Switch to Miniconda env source activate aind
Install PyGame for visualisations brew install sdl sdl_image sdl_mixer sdl_ttf portmidi mercurial pip install pygame

Sudoku (uses '.' for boxes where value not yet known) python ./term01/lesson01/function.py
Sudoku (uses '123456789' for boxes where value not yet known, and for each box with only one value possibility, for each of its peers it removes that value from the remaining list of possibilities python ./term01/lesson01/function2.py
Refer to separate Project 1 https://github.com/ltfschoen/AIND-Sudoku
Perceptron Step:

cd ./term02/lesson02/perceptron_step; python perceptron_step.py
Logistic Regression (Gradient Descent Algorithm)

cd ./term02/lesson02/logistic_regression; python logistic_regression.py

Neural Network XOR using Keras

pip install keras
pip install tensorflow
cd ./term02/lesson02/part30_keras_neural_network_xor; python neural_network.py

Convolutional Neural Network (CNN) using Keras

Study Dimensionality of Convolutional Layer changes

 cd ./term02/lesson03; python conv-dims.py
 cd ./term02/lesson03; python conv-dims2.py

cd ./term02/lesson03; python pool-dims.py

Recurrent Neural Network (RNN) using Keras
- Plot Recursive Sequence
```
cd ./term02/lesson04; python recursive_sequence.py
```

Natural Language Processing (NLP)

  ```
  cd ./term02/lesson07_nlp_intro; python count_words.py
  ```

Install MyPy python3 -m pip install mypy.
Install Typing python3 -m pip install typing.
Import Typing import typing; from typing import *,
Apply MyPy static type checking to an existing function and check that expected warnings occur def get_list_for_str(word: str) -> None:
Run MyPy Linter with mypy ./mypy/test.py
Apply MyPy with expected types so no warnings occur def get_list_for_str(word: str) -> typing.List[str]: