-
* informational documents or papers:
1.Decentralized training of foundation models in heterogeneous environments, https://dl.acm.org/doi/10.5555/3600270.3602116
2.
* Requirements:
1. Power lim…
-
For the code:
https://github.com/THUDM/CogVideo/blob/2fdc59c3ce48aee1ba7572a1c241e5b3090abffa/sat/configs/sft.yaml#L39 , **contiguous_gradients** is deepspeed memory optimization, which is **default …
-
Dear Author, I attempted to apply this method on other datasets; however I have observed that the mu_pdist、sigma_pdist and logits distributions are very concentrated during training , even though the …
-
Follow the guide here: https://github.com/intel/ai-reference-models/tree/main/models_v2/pytorch/llama/training/cpu, faced several issues:
1. https://github.com/intel/ai-reference-models/blob/main/m…
-
Does ai-toolkit support training non-square images? I don't want to crop and scale the images.
I am training flux with 768*1024 images. On an A100-40G machine, 3000 steps take 2.5 hours. I set buck…
-
## This is for bugs only
Did you already ask [in the discord](https://discord.gg/VXmU2f5WEU)?
No
You verified that this is a bug and not a feature request or question by asking [in the discor…
-
### System Info
OS Version : windows 11
Python Version : 3.11.9
Pandas ai version : 2.2.14
### 🐛 Describe the bug
Thanks for the Great work!
As a part of training the agent (QA Training) ,starte…
-
Connecting AI model to Server
-
Add a way for AIs to repeatedly play each other and use that experience to train themselves. This will require adding some mechanism for preserving some AI state across multiple turns and games.
-
### Description of the bug:
I'm trying to convert the following (quantized) model:
```python
# Disable GPU for model conversion to tflite.
# Fix for https://github.com/google-ai-edge/ai-edge…