NN-from-scratch

This is a try towards developing a nueral network toolbox for machine learning in Matlab especially for the purpose of devloping multi input lstm Layers. Since the development is not completed please report any errors or misconcept. Kindly accept any inconvenience.

Currently available layers

1. Fully Connected Layers
2. LSTM Layers

you can find the example of creating layers in the TestNet.m and TestMultiNet.m files.

reference:

1. Matlab Deep Learning Toolbox documentation
2. Andre Ng, Sequence Model | Coursera, https://www.coursera.org/learn/nlp-sequence-models
3. Andre Ng, Machine Learning | Coursera, https://www.coursera.org/learn/machine-learning
4. Jonathan Weisberg, Building a Neural Network from Scratch, https://jonathanweisberg.org/post/A%20Neural%20Network%20from%20Scratch%20-%20Part%202/
5. Varuna Jayasiri, Vanila LSTM with Numpy: http://blog.varunajayasiri.com/numpy_lstm.html