Category Archives: Keras

sample_weights functionality adds more importance to some samples than others during training.so that the loss function used by the model is weighted per-sample not per class. It changes the way the loss is calculated

You can do this whether you’re building Sequential models, Functional API models, or subclassed models. Using this custom training algorithm, you still get the benefit from the convenient features of fit(), such as callbacks, built-in distribution support, or step

The sigmoid was not zero-centered tanh fixed this and now ReLU has this problem again and that’s one of the issues of the ReLU. ReLU doesn’t activate for negative inputs, it’s possible to end up with “dead neurons” that never fire.

In multi-classes classification last layer use “softmax” activation, which means it will return an array of 10 probability scores (summing to 1) for 10 class. Each score will be the probability that the current class belongs to one of our 10 classes.

All category is not represented in the same quantity. Some of them can be very infrequent which may represent a hard challenge for any ML algorithm.

This tutorial is dedicated to the use of the Transformers library using TensorFlow and the Keras API to fine-tune a State-of-The-Art Transformer model.

The image is passed through a stack of convolutional layers, where VGG uses 3×3 filters which are the smallest size to capture the notion of left/right, up/down, center.

We can import the module that we just uploaded, and then call the function. This file in the same directory so that Python knows where to find the module since it’s not a built-in module.

It’s worth noting that different parts of the data pipeline will stress different parts of the system. So for loading from the disk is an I/o- bound task and we’ll generally want to consume this I/O as fast as possible so that we’re not constantly waiting for images to arrive from disk one at a time.

A simple classification layer is added to the pre-trained model, and all parameters are jointly fine-tuned on a downstream task.