-
imagenet对我来说太大了，我想用自己的图片来进行训练，我是不是直接把八万张图片丢到train文件夹里面就可以了呢，请问一张a100够吗，我需要调整哪些别的超参数呢，谢谢您的工作，祝您一切顺利！！！
-
So I been trying so hard with this cuda backend problem, I intalled cuda 12.1, python 3.10, pytorch 2.4.0 and installed all the requirement without error, but when ever i run run.py this error popup
…
-
## error log
没有任何错误
## ncnn版本
ncnn-20220420-windows-vs2019
## model | 模型 | モデル
```
def create_mobilenet_v3_large_100_v1(embed_size=128, device = "cpu", pretrained = True):
model_c…
-
При попытке запуститься говорит о нехватке памяти видеокарты RTX 3050ti.
When trying to start, it indicates a lack of memory of the RTX 3050ti video card.
`◼️ Malevich is 1.3 billion params model…
-
Extensive GPU benchmarks with AMD gpus can now be run by following steps after building the rocm sdk. This version has now been synced with the upstream version which has fixed the pytorch 2.0 support…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues and checked the recent builds/commits of both this extension and the webui
### Are you using the latest version of …
-
### 🚀 The feature, motivation and pitch
I need to observe the strength of attention for each element in the sequence with all the elements in the same sequence. This is called the attention map in …
-
我的服务器上有两张A40，我想使用其中一张进行微调，使用了代码中的单机单卡微调方案，但是，当其中一张显卡的显存占用达到大概20G的时候，另一张显卡也会被使用，并且会瞬间使用另一张显卡的20G显存。几秒之后，最开始的显卡显存超过48G，报错，显存占用如下图所示。
![捕获](https://github.com/user-attachments/assets/0dbc970e-fb18-40ea-…
-
Is this the correct JSON ? Could you point out the way to download the weights ?
I should also point out that I ported your PyTorch code to TensorFlow. Changes are minimal though.
`{
"module"…
-
Hello, I had a problem with the mismatch between the TP and GT sizes while training, but the GT is not at all 800x518 in size, do you know what's going on
Traceback (most recent call last):
File "…