IndexError: index 5 is out of bounds for dimension 0 with size 1 when i do training with Gradient Descent

1

I have a code for running Gradient Descent on mnist dataset, but after creating my network and loading the data, when i’m going to do training , i get this error

grads['dl/df2'][label] -= 1
IndexError: index 5 is out of bounds for dimension 0 with size 1

here is my code:

dataset = MNIST(root='root', download=True )
images, labels =  dataset.data ,dataset.targets
images, labels = images[:n_examples], labels[:n_examples]
images, labels = iter(images.view(-1, 784) / 255), iter(labels)

grads={}
start = t()

for i, (image, label) in enumerate(zip(images, labels)):

    label = torch.Tensor([label]).long()

    # Run simulation for single datum.

    inpts = {
        'X': image.repeat(time, 1), 'Y_b': torch.ones(time, 1)
    }
    network.run(inpts=inpts, time=time)

    # Compute softmax of output spiking activity and get predicted label.
    output = summed_inputs['Y'].softmax(0).view(1, -1)
    predicted = output.argmax(1).item()
    correct[i % update_interval] = int(predicted == label[0].item())
    predictions.append(predicted)
    ground_truth.append(label)

    # Compute cross-entropy loss between output and true label.
    losses[i % update_interval] = criterion(output, label)

    if train:
        # Compute gradient of the loss WRT average firing rates.
        grads['dl/df'] = summed_inputs['Y'].softmax(0)
        grads['dl/df'][label] -= 1

        # Compute gradient of the summed voltages of connection weights.
        grads['dl/dw'] = torch.ger(summed_inputs['X'], grads['dl/df'])
        grads['dl/db'] = grads['dl/df']

        # Do stochastic gradient descent calculation.
        network.connections['X', 'Y'].w -= lr * grads['dl/dw']
        network.connections['Y_b', 'Y'].w -= lr * grads['dl/db']

what’s wrong with my code?

2

The error points to a single value in grads['dl/df2'], while you are trying to index at position 5.
Your code seems to use some higher-level wrappers and other undefined functions, so that we would need a small executable code snippet to help debugging.

3

Hi @ptrblck …thank you for your response
As much as I could , i truncated my code to be small here

import os
import torch
import argparse
import numpy as np
from time import time as t
from torchvision import transforms
from bindsnet.datasets import MNIST
from bindsnet.network import Network, load
from bindsnet.network.topology import Connection
from bindsnet.network.nodes import RealInput, IFNodes


# Paths.
ROOT_DIR ='/home/Users/Ati/Desktop/GD'
data_path = os.path.join(ROOT_DIR, 'data', 'MNIST')

def main(seed=0, n_train=60000, n_test=10000, time=50, lr=0.01, lr_decay=0.95,
         update_interval=500, max_prob=1.0, plot=False, train=True, gpu=False):

    np.random.seed(seed)
    criterion = torch.nn.CrossEntropyLoss() 
    n_examples = n_train if train else n_test

    if train:
        # Network building.
        network = Network()

        # Groups of neurons.
        input_layer = RealInput(n=784, sum_input=True)
        output_layer = IFNodes(n=10, sum_input=True)
        network.add_layer(input_layer, name='X')
        network.add_layer(output_layer, name='Y')

        # Connections between groups of neurons.
        input_connection = Connection(source=input_layer, target=output_layer, norm=150, wmin=-1, wmax=1)
        network.add_connection(input_connection, source='X', target='Y')

    root = data_path
    dataset = MNIST(root='root', download=True )
    images, labels =  dataset.data ,dataset.targets
    images, labels = images.view(-1, 784) / 255, labels


    grads = {}
    accuracies = []
    predictions = []
    losses = torch.zeros(update_interval)
    correct = torch.zeros(update_interval)

    # Run training.
    start = t()
    for i in range(n_examples):
        label = torch.Tensor([labels[i % len(labels)]]).long()
        image = images[i % len(labels)]

        # Run simulation for single datum.
        
        inpts = {
            'X': image.repeat(time, 1)}
        network.run(inpts=inpts, time=time)

        # Retrieve spikes and summed inputs from both layers.
        summed_inputs = {l: network.layers[l].summed for l in network.layers}

        # Compute softmax of output spiking activity and get predicted label.
        output = summed_inputs['Y'].softmax(0).view(1, -1)
        predicted = output.argmax(1).item()
        correct[i % update_interval] = int(predicted == label[0].item())
        predictions.append(predicted)

        # Compute cross-entropy loss between output and true label.
        losses[i % update_interval] = criterion(output, label)

        if train:
            # Compute gradient of the loss WRT average firing rates.
            grads['dl/df'] = summed_inputs['Y'].softmax(0)
            grads['dl/df'][label] -= 1

            # Compute gradient of the summed voltages WRT connection weights.
            # This is an approximation; the summed voltages are not a
            # smooth function of the connection weights.
            grads['dl/dw'] = torch.ger(summed_inputs['X'], grads['dl/df'])
            grads['dl/db'] = grads['dl/df']

            # Do stochastic gradient descent calculation.
            network.connections['X', 'Y'].w -= lr * grads['dl/dw']
            network.connections['Y_b', 'Y'].w -= lr * grads['dl/db']

        if i > 0 and i % update_interval == 0:
            accuracies.append(correct.mean() * 100)
       # Decay learning rate.
            lr *= lr_decay

            start = t()

        network.reset_()  # Reset state variables.

    accuracies.append(correct.mean() * 100)

    if train:
        lr *= lr_decay
        for c in network.connections:
            network.connections[c].update_rule.weight_decay *= lr_decay
if __name__ == '__main__':
    # Parameters.
    parser = argparse.ArgumentParser()
    parser.add_argument('--seed', type=int, default=0, help='random seed')
    parser.add_argument('--n_train', type=int, default=60000, help='no. of training samples')
    parser.add_argument('--time', default=25, type=int, help='simulation time')
    parser.add_argument('--lr', default=0.01, type=float, help='learning rate')
    parser.add_argument('--lr_decay', default=0.95, type=float, help='learning rate decay')
    parser.add_argument('--update_interval', default=500, type=int, help='no. examples between evaluation')
    parser.add_argument('--max_prob', default=1.0, type=float, help='maximum prob. of input spikes')
    parser.add_argument('--train', dest='train', action='store_true', help='train phase')
    parser.set_defaults(plot=False, train=True)
    args = parser.parse_args()

    seed = args.seed
    n_train = args.n_train
    time = args.time
    lr = args.lr
    lr_decay = args.lr_decay
    update_interval = args.update_interval
    max_prob = args.max_prob
    train = args.train
    args = vars(args)
    main(seed=seed, n_train=n_train, time=time, lr=lr, lr_decay=lr_decay,
         update_interval=update_interval, max_prob=max_prob, train=train)

Just i don’t know that it’s needed to install bindsnet package…but it’s just for creating a network with two layers and creating connections between layers
If there is needed more explanations for clarification , please tell me i would do:)

4

Hey … Could you someone help me on this pleaseeeeeeeeeeee
I stuck in this error:pensive:

5

Could you try to create an executable code snippet (e.g. just define any simple model etc.), which creates this error, please?