-
As the title says. torchinfo cannot determine the size of the weights in a parameterlist, and therefore it doesn't calculate the size of the network properly.
Fix would be greatly appreciated! torch…
-
**Is your feature request related to a problem? Please describe.**
A feature request from Clara users:
it is useful to print out the network architecture
for this we need to:
pip install torchsumm…
-
## 🐛 Bug
`torch.linalg.eig` segfaults if NaN is in tensor when using mkl
## To Reproduce
Steps to reproduce the behavior:
1. Run torch.linalg.eig on a tensor with `float('nan')` in at leas…
-
## 🐛 Bug
When I use the SummaryWriter() function from torch.utils.tensorboard to create Tensorboard logs of the hyper-parameters and metrics (scalar), as well as the losses (image plot), and the mo…
-
I'm experimenting with getting the ONNX model (trained to classify MNIST digits) to work on K210 with all the nice recent updates in nncase and found that all the ops imported seems to work fine durin…
aclex updated
3 years ago
-
I use a module with output of 0-dim tensor in the model class.
When the batch_dim is not None, an IndexError will be raised.
For example:
```python
import torch
import torch.nn as nn
from to…
-
## Issue description
Build of `python39Packages.sentry-sdk` failed on x86_64-linux as of 891d0226de6e281422a8d0d99814925fd7aed8d0. This is currently breaking python3Packages.wandb.
```
error: builde…
-
I try your code for resnet152
```
import torchvision
from torchinfo import summary
model = torchvision.models.resnet152(pretrained=False)
summary(model, (1, 3, 224, 224), depth=3)
```
But…
-
## 🐛 Bug
Our model behaves differently while using the same weights and inputs. We just change the order of loading weights using `torch.load()`.
## To Reproduce
This is a small script that I…
-
I'm experiencing the following error after calling [torchinfo.summary](https://github.com/tyleryep/torchinfo):
```
ERROR:neptune.new.internal.operation_processors.async_operation_processor:Error occ…