Name		Name	Last commit message	Last commit date
parent directory ..
data		data
images		images
test_support_files		test_support_files
.gitignore		.gitignore
C3_W2_Assignment.ipynb		C3_W2_Assignment.ipynb
C3_W2_lecture_nb_1_Hidden_State_Activation.ipynb		C3_W2_lecture_nb_1_Hidden_State_Activation.ipynb
C3_W2_lecture_nb_2_RNNs.ipynb		C3_W2_lecture_nb_2_RNNs.ipynb
C3_W2_lecture_nb_3_perplexity.ipynb		C3_W2_lecture_nb_3_perplexity.ipynb
C3_W2_lecture_nb_4_GRU.ipynb		C3_W2_lecture_nb_4_GRU.ipynb
README.md		README.md
model.pkl.gz		model.pkl.gz
targets.npy		targets.npy
w2_unittest.py		w2_unittest.py

README.md

Recurrent Neural Networks for Language Modeling

Hidden State Activation

Take another look at the hidden state activation function. It can be written in two different ways. I'll show you, step by step, how to implement each and then how to verify the results produced by each are the same.
Vanilla RNNs, GRUs and the scan function

Learn how to define the forward method for vanilla RNNs and GRUs. Additionally, you will see how to define and use the function scan to compute forward propagation for RNNs.
Working with JAX NumPy and Calculating Perplexity

Learn about data generators. Data generators will be part of every assignment of C3.
Creating a GRU model using Trax

Trax allows to define neural network architectures by stacking layers (similarly to other libraries such as Keras). In this ungraded lab you will see how a GRU model can be created using Trax.