-
If you have a question or would like help and support, please ask at our
[forums](https://discuss.pytorch.org/).
If you are submitting a feature request, please preface the title with [feature req…
-
I think similar issue as : https://github.com/Theano/Theano/issues/5504
GPU: Radeon Pro 560, OS X
Theano is running on GPU via OpenCL, not CUDA. I ran the script here "http://deeplearning.net/soft…
-
## Description
(Brief description of the problem in no more than 2 sentences.)
I have two files:
- VAE.py which defines a variational auto encoder
- vae_test.py which loads these classes and run…
-
```
from torchvision import models
...
dataloader for CIFAR10
...
model = models.vgg16(False)
for epoch in range(80):
for i, (images, labels) in enumerate(train_loader):
imag…
-
### System information
- Manjaro Linux
- TensorFlow installed with pip:
- TensorFlow version: v1.3.0-rc2-20-g0787eee 1.3.0:
- Python version: 3.6.2
- CUDA/cuDNN version: 8.0.61-3 / 7.0.3-1
- GP…
-
Default kernel weights initialization of convolution layer. I use the function conv2d, but I can't find the initial weights of the convolution kernel , or how initialize the weights of convolution ke…
-
```
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, transf…
-
I am getting very inconsistent behavior with my `tf.Graph` objects and I can't explain why it's happening. I'm running this in a Jupyter notebook. Could this affect it at all?
My versions:
``…
-
一个参考的:https://discuss.pytorch.org/t/how-to-create-model-with-sharing-weight/398/2
# -*- coding: utf-8 -*-
# License: BSD
# Author: Sasank Chilamkurthy
from __future__ import print_function, di…
-
- patch%itype(p) is the label for the PFT of a given patch, which is a concept that is valid within CLM but not within FATES.
- Thus, we need to eradicate calls that use patch%itype when FATES is on…