-
Summary: Running `lightweight_gan --models_dir "path" --show-progress ` results in a `AttributeError: 'NoneType' object has no attribute 'split'`
Full error:
`Generating progress images: 0% 0/…
-
### 🐛 Describe the bug
Hello! I am unable to get a CUDA Graph to replay when using a DDP model. I am unsure if this is because I am not doing the required steps to make both play nice with each other…
-
Hello, I have a question about getting gradients with equinox.
In the function of **first** 'grad_and_loss' in the code below, when I tried to get the gradient but only got zero gradient.
For conv…
-
## 🐛 Bug
`torch.matmul` often returns different gradients for the same matrices, when the computation is done with an additional dimension (batched version). So if A and B are 2D matrices:
```py…
-
> ValueError: Reverse-mode differentiation does not work for lax.while_loop or lax.fori_loop with dynamic start/stop values. Try using lax.scan, or using fori_loop with static start/stop.
I would l…
-
## Description
Trying to build mxnet for Ubuntu as explained here: https://mxnet.apache.org/versions/1.9.1/get_started/ubuntu_setup.html
I get the following errors (_Fehler_ is the german word for _…
-
### Search before asking
- [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and found no similar feature requests.
### Description
In the `ultralyt…
-
作者您好!您的思路独具特色,效果也非常好,我尝试使用您公开的代码进行实验,
在观察您的issue后,我使用了您分享的阶段2数据集进行阶段2的实验,但是我遇到了如下两个问题:
1.您在dataloader3d文件的getitem函数返回的是
output = {'img': A_img, 'label': A_label},
但是在模型训练的时候使用的是
A_img=train_A_dat…
-
# Deep learning is EXPENSIVE
e.g.
train ResNet50 with ImageNet dataset for 80 epochs
80 * 1.3M images * 7.7B ops per img
# Solution?
- **Data Parallelism (large batch training)**
 are released under the Apache 2.0 license.
I ask you to reconsider the license choice of your project.
…