Starting with pytoch - check installation

rozo · July 9, 2020, 12:04pm

Hello,
Sorry for the noby question. I installed torch 1.1 on my windows machine.
I am not sure if it is ok since “torce.cuda.is_available()” return true but all examples I try seems to not run on the gpu.

Is there a simple code (A simple torch loop for example) I can run just to see that the gpu is loaded so I can verify that torch can really run on gpu?

Thanks in advanced.

peterjc123 · July 10, 2020, 2:06am

torch.cuda.FloatTensor([1.])

peterjc123 · July 10, 2020, 2:17am

github.com

pytorch/examples/blob/master/mnist/main.py

from __future__ import print_function
import argparse
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.optim.lr_scheduler import StepLR


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, 3, 1)
        self.conv2 = nn.Conv2d(32, 64, 3, 1)
        self.dropout1 = nn.Dropout2d(0.25)
        self.dropout2 = nn.Dropout2d(0.5)
        self.fc1 = nn.Linear(9216, 128)
        self.fc2 = nn.Linear(128, 10)

This file has been truncated. show original