https://github.com/belhoussine/neuralnet
A Machine Learning library for Neural Networks fully written in python. It supports multiple layers of neurons and offers a variety of activation functions, optimization algorithms, and utility functions.
https://github.com/belhoussine/neuralnet
artifical-neural-network library machine-learning neural-network python
Last synced: 2 months ago
JSON representation
A Machine Learning library for Neural Networks fully written in python. It supports multiple layers of neurons and offers a variety of activation functions, optimization algorithms, and utility functions.
- Host: GitHub
- URL: https://github.com/belhoussine/neuralnet
- Owner: Belhoussine
- Created: 2020-09-12T21:17:56.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2022-04-27T19:26:53.000Z (about 4 years ago)
- Last Synced: 2025-12-09T22:20:22.438Z (7 months ago)
- Topics: artifical-neural-network, library, machine-learning, neural-network, python
- Language: Python
- Homepage:
- Size: 10.7 MB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Machine Learning Library
## Usage Notes:
Python3 is needed to use this library.
```
git clone https://github.com/Belhoussine/NeuralNet
cd NeuralNet
pip3 install requirements.txt
```
## Neural Network Specifications
### 1. Artificial Neural Network:
- [x] Supports multiple layers
- [x] Supports multiple neurons per layer
- [ ] Train:
- [x] Forward Propagation
- [ ] Back Propagation
- [x] Run in Epochs
- [x] Supports mini batches
- [x] Predict
- [x] Verbose training phase
### 2. Activation Functions:
- [x] Sigmoid (Non linear mapping between 0 and 1)
- [x] Softmax (Non Linear Probability Distribution)
- [x] ReLU (Rectified Linear Unit)
- [x] Leaky ReLU (Leaking ReLU on negative values)
- [x] TanH (Hyperbolic Tangent)
- [x] ELU (Exponential Linear Unit)
### 3. Loss Functions:
- [x] RMSE (Root Mean Squared Error)
- [x] MSE (Mean Squared Error)
- [x] SSE (Sum Squared Error)
- [x] MAE (Mean Absolute Error)
- [x] LogCosH (Log of Hyperbolic cosine)
- [x] Huber (Hyperbolic Tangent)
- [ ] Cross Entropy (Logistic Loss)
- [ ] Least Squares
### 3. Optimization Algorithms:
- [ ] Batch Gradient Descent
- [ ] SGD (Stochastic Gradient Descent)
- [ ] Mini-Batch Gradient Descent
- [ ] General Purpose Gradient Descent
- [ ] ADAM (Adaptive Moment Estimation)
- [ ] RMSProp
### 4. Utility Functions:
- [x] Download MNIST dataset from remote server
- [x] Flatten (Convert 2D Matrix to vector)
- [x] One Hot Encoding (Convert numerical to categorical)
- [x] One Hot Decoding (Convert categorical to numerical)
- [x] Normalization Function (Linear Mapping between 0 and 1)
- [x] Accurary function (Compute Model Accuracy)
- [x] Activate (Applies given activation function)
- [x] Compute Loss (with chosen loss function)
- [ ] Optimize (Applies given optimizer on model)
- [x] Shuffle (Shuffles training data)