Questions tagged [numpy]

For questions related to the NumPy library that are NOT just about programming issues (which are off-topic here).

Filter by
Sorted by
Tagged with
0 votes
1 answer
19 views

Why do smaller weights converge faster for RNNs?

I am writing a Recurrent Neural Network using only the NumPy library for a binary classification problem. When I initialize the weights with np.random.randn, after 1000 epochs it gets ~60% accuracy, ...
  • 15
1 vote
0 answers
46 views

What is the most accurate way of building a Perceptron using only NumPy?

For context, I am trying to write a bunch of neural network programs using no other packages besides NumPy for educational purposes. I am trying to make them as simple as possible, i.e. removing the ...
  • 15
0 votes
0 answers
137 views

Is there a vectorized Implementation of Convolution (Full Mode) with a very big kernel/gradient?

I'm currently trying to figure a way to implement the backpropagation of a convolutional layer with plain numpy. In theory, I can calculate the partial derivative of the loss w.r.t. the convolution ...
1 vote
1 answer
545 views

Value Iteration failing to converge to optimal value function in Sutton-Barto's Gambler problem

In Example 4.3:Gambler's Problem of Sutton and Barto's book whose code is given here. In this code the value function array is initialized as np.zeros(states) where ...
4 votes
1 answer
241 views

Why is my derivation of the back-propagation equations inconsistent with Andrew Ng's slides from Coursera?

I am using the cross-entropy cost function to calculate its derivatives using different variables $Z, W$ and $b$ at different instances. Please refer image below for calculation. As per my knowledge, ...
  • 151
2 votes
2 answers
139 views

How can I perform lossless compression of images so that they can be stored to train a CNN?

I have a set of images, which are quite large in size (1000x1000), and as such do not easily fit into memory. I'd like to compress these images, such that little information is missing. I am looking ...