Neural Nets
Neural Networks
What is Neural Network? Neuron? Weights? Bias? Forward Propagation? Cost Function? Gradient Descent? Learning Rate? Back propagation? Batches? Epochs? Dropout? Batch Normalisation? CNN? Pooling? Padding? Data Augmentation? Recurrent Neurone? RNN? Vanishing Gradient Problem? Exploding Gradient Problem.
- https://www.analyticsvidhya.com/blog/2017/05/25-must-know-terms-concepts-for-beginners-in-deep-learning/
Introduction to Tensorflow
Introduction to Convolution Networks
What it is? Why it is there? How it does?
https://www.youtube.com/watch?v=FmpDIaiMIeA
Going a bit deeper, here is an article which goes step by step explanation of how spatial locations are important,
As now you have an intuition of how Conv Nets works, it’s time to understand few concepts a bit deeper concepts about image processing.
Please check filters from wikipedia.
Understand these questions
-
What is a filter? Is it always 2d?
-
How does a filter effect output dimensions?
-
What is striding?
-
What is padding?
As we want to shrink higher dimensional images into lower dimensional feature vectors, we use filters to map these large images. As repeatedly apply filter, we can have these
https://stackoverflow.com/questions/37674306/what-is-the-difference-between-same-and-valid-padding-in-tf-nn-max-pool-of-t