Prediction probabilities

Weights of the model

Feature maps of the first convolutional block

Class labels

A machine learning model that uses recurrent neural networks to learn relationships between different parts of a sequence.

A deep learning model that uses self-attention to learn relationships between different parts of a sequence.

A natural language processing model that uses convolutions to learn relationships between different parts of a sequence

A computer vision model that uses fully connected layers to learn relationships between different parts of an image.

~50

0

 ~100

>150

Value of the function

Approximate gradient using central difference

Exact gradient at x

Forward difference gradient

Freezes pre-trained convolutional layers for feature extraction

Trains all layers from scratch

Uses VGG16 without pre-trained weights

A. Replaces top layers with Dense layers

R-squared, Explained Variance

Accuracy, Precision, Recall

Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE)

Log-loss, Gini, AUC

Confusion matrix

Accuracy and loss plots

Feature importance per sample

Class probabilities

sequential training

reinforcement learning

adversarial training

joint learning

(None, 64)

(None, 1)

(None, 64, 1)

(None, 10, 64)

To optimize the generator

To reshape input data for classification

To distinguish real data from generated data

To produce noise vectors