Mc dropout pytorch

import sys

import numpy as np

import torch
import torch.nn as nn


def enable_dropout(model):
    """ Function to enable the dropout layers during test-time """
    for m in model.modules():
        if m.__class__.__name__.startswith('Dropout'):
            m.train()

def get_monte_carlo_predictions(data_loader,
                                forward_passes,
                                model,
                                n_classes,
                                n_samples):
    """ Function to get the monte-carlo samples and uncertainty estimates
    through multiple forward passes

    Parameters
    ----------
    data_loader : object
        data loader object from the data loader module
    forward_passes : int
        number of monte-carlo samples/forward passes
    model : object
        keras model
    n_classes : int
        number of classes in the dataset
    n_samples : int
        number of samples in the test set
    """

    dropout_predictions = np.empty((0, n_samples, n_classes))
    softmax = nn.Softmax(dim=1)
    for i in range(forward_passes):
        predictions = np.empty((0, n_classes))
        model.eval()
        enable_dropout(model)
        for i, (image, label) in enumerate(data_loader):

            image = image.to(torch.device('cuda'))
            with torch.no_grad():
                output = model(image)
                output = softmax(output) # shape (n_samples, n_classes)
            predictions = np.vstack((predictions, output.cpu().numpy()))

        dropout_predictions = np.vstack((dropout_predictions,
                                         predictions[np.newaxis, :, :]))
        # dropout predictions - shape (forward_passes, n_samples, n_classes)
    
    # Calculating mean across multiple MCD forward passes 
    mean = np.mean(dropout_predictions, axis=0) # shape (n_samples, n_classes)

    # Calculating variance across multiple MCD forward passes 
    variance = np.var(dropout_predictions, axis=0) # shape (n_samples, n_classes)

    epsilon = sys.float_info.min
    # Calculating entropy across multiple MCD forward passes 
    entropy = -np.sum(mean*np.log(mean + epsilon), axis=-1) # shape (n_samples,)

    # Calculating mutual information across multiple MCD forward passes 
    mutual_info = entropy - np.mean(np.sum(-dropout_predictions*np.log(dropout_predictions + epsilon),
                                            axis=-1), axis=0) # shape (n_samples,)

Source: stackoverflow.com

Similar codes

0 vote246 views1 answer
Virtualbox how to move vmdk to another folder
C
1 vote198 views1 answer
C return
C
0 vote205 views1 answer
FivemStore
C
-1 vote255 views2 answers
Char in c
C
1 vote266 views1 answer
Lichess puzzles
C
0 vote190 views1 answer
Two way communication between child and parent processes in C using pipes
C
0 vote251 views4 answers
Bootstrap carousel stop autoplay
C
1 vote194 views1 answer
Arduino client disconnect
C
2 votes188 views1 answer
C union in struct
C
-1 vote202 views1 answer
Arduino remove()
C