-
### Describe the bug
When train_dreambooth_lora_flux attempts to generate images during validation, `RuntimeError: Input type (float) and bias type (c10::BFloat16) should be the same` is thrown
### …
-
This is needed if possible to tweak the inference code. The whole thing doesn't fit in the VRAM for even nvidia 4090 and takes forever.
-
**Inference:**
```bash
CUDA_VISIBLE_DEVICES=0 swift infer --model_type got-ocr2 --model_id_or_path stepfun-ai/GOT-OCR2_0
```
```
-
```bash
# Single GPU training
sh finetune/finetune_lora_single_gpu.sh
# Distributed training
sh finetune/finetune_lora_ds.sh
```
DATA="path_to_data"
能否提供DATA样例，比如coco图片描述
另，finetune是单阶段训练就可以了（…
-
Here is my original comment of a recently closed (but unresolved) issue:
>This issue is not completed! I still think this is very very necessary functionality that ONNX is completely missing. There h…
-
### Please check that this issue hasn't been reported before.
- [X] I searched previous [Bug Reports](https://github.com/OpenAccess-AI-Collective/axolotl/labels/bug) didn't find any similar reports.
…
-
### Is there an existing issue for this problem?
- [X] I have searched the existing issues
### Operating system
Windows
### GPU vendor
Nvidia (CUDA)
### GPU model
4090
### GPU …
-
Select input model, base model (analog madness v7) in this case.
It works at 512x512 on cpu/gpu, larger throws:
INFO UNet2DConditionModel: 64, 8, 768, False, False …
-
I'm currently trying to run the script with the following paperspace machine
When doing so, I get this err…
-
Hi,
Could you tell me how to evaluate LLaVA with the enhanced CLIP you trained?