todorovicsrdjan/annuto

Simple automated feed-forward artificial neural network.

AnnUTO

Automated feed-forward artificial neural network

Features:

• formats (atm): csv
• different methods for parsing dataset: url or file (multipart)
• model training (clasification or regression)
• results for training or testing set
• statistical indicators
• correlation matrix (image)
• feature (input) encoding and scaling
• custom architecture (layers, activation functions, number of nodes, weight initializers)
• hyperparameters configuration (algorithm, optimizer, learning rate)
• different metrics
• fill NaNs
  • mean
  • median
  • most frequent
  • fill with constant (number or string)
• modify dataset
• real-time reporting using sockets

Setup

1. Download repo

git clone https://github.com/TodorovicSrdjan/AnnUTO.git

2. Set current working directory to downloaded folder (AnnUTO)
3. (Optional) Create virtual environment and activate it

python -m venv path/to/some/folder

// linux; bash
source path/to/some/folder/bin/activate

4. Install project dependencies

pip install -r requirements.txt

5. Run app

python src/ann_server.py

Usage

You can see API specification at: localhost:10003/docs

API (atm) does not have it's own database. After parsing dataset (url or file) API returns dataset in internal json format.

Other endpoints can only work with that returned dataset. It's provided as url for stored_dataset parameter.

You can configure some things like urls and ports in file config.py