-
Hello, could you explain how the pre-trained model you loaded during the training of DA-Clip was obtained? Because I noticed that the model “laion2b_s34b_b79k” is used in your training command, but in…
-
Hi! Amazing work! I've been working on reproducing the results and ran into a few things that I was hoping you could clarify when you have a moment.
1. **Winoground Compositionality Results**: I no…
-
I tried to use ip-adapter-plus_sd15 with both image encoder modules you provided in huggingface but encountered errors. What CLIP vision model did you use for ip-adapter-plus?
-
The third-party SD ecosystem has mostly shifted to the [`safetensors`](https://github.com/huggingface/safetensors) format for distributing models, as the Python `pickle` format allows for arbitrary co…
-
i don't know what's in models/image_encoder?
-
…
-
您好，请教下，论文里提到的用Laion-400M预训练，是指用Laion-400M对VideoComposer做了额外的预训练 ？如果是的话，预训练的输入组织方式，和参与训练的算法模块，可以讲解一下吗？ 谢谢 ~
PS: 看代码里和Laion相关的有2个预训练模型，没有找到Laion-400M相关的，是不是我理解错了？
- "v2-1_512-ema-pruned.ckpt" ：预训练是用…
-
Can I use my own model in.bin format，tks
-
准备复现ChineseClip论文，以CLIP-VIT-B/16 初始化image encoder部分，下载对应的是 https://huggingface.co/openai/clip-vit-base-patch16/tree/main 但是加载模型参数时，发现image encoder部分参数加载不上。我打印发现对应参数名称以vision_model.encoder.layers.开头…
-
Could you please provide a tutorial or code snippet that shows how to speed up CLIP / Open CLIP ViT models using xformers?