Machine learning from scratch
Numpy-ml from scratch. This repo aims to help myself/people understand the math behind
machine learning algorithms and I will try to make the computation as
efficient as possible
Implementations
Supervised Learning
Deep Learning
-
Neural Network
-
Layers
- Activation Layer
- Batch Normalization Layer
- Dropout Layer
- Fully Connected Layer
- Embedding Layer
- RNN Layer: many-to-one
- LSTM ayer: many-to-one
- Bidirectional LSTM
-
Loss Functions
- Cross Entropy
- Loss for VAE
- BinomialDeviance
- Noise Contrastive Estimation
-
Optimizer
- SGD with momentum
- RMSprop
- Adagrad
- Adadelta
- Adam
-
Schedulers
- CosineAnnealingLR
- CosineAnnealingWarmRestarts
-
Models
Unsupervised Learning
Examples
SVM
Polynomial Lasso Regression
Decision Tree for Classification
Decision Tree for Regression
Xgboost
deep learning
The result of unit test for different parts of deep learning
The warning message is due to the bug of Tensorflow
unit test page