-
Hi ! [Hugging Face Datasets](https://github.com/huggingface/datasets) uses Arrow for image datasets, which can be loaded as map-style datasets or iterable datasets. I ran some benchmarks on ImageNet-1…
-
* download latency (average and 99p per shard)
* dns resolving latency (average and 99p per shard)
-
optionally
related https://github.com/rom1504/img2dataset/issues/42
kind of needed to avoid people having issues if they don't set up a good dns (not that obvious to do depending on env)
-
I'm seeing lots of requests to save the image generation details along with the generated image. For times when the metadata might be stripped from the image on posting, having all of the details save…
-
I see the accuracy of your one epoch distillation TinyCLIP-ViT-39M-16-Text-19M on the laion400m dataset is 54%.
I distill clip on yfcc dataset of 5m image-text pairs, training one epoch of accuracy …
-
Thx for your code sharing.
In this repo, you provided only Elon_musk.tar file.
To execute the SD notebook for Mark Zuckerberg, We need Mark Zuckerberg tar file.
But, due to the limitation of our se…
-
Attempting to run charl-e on my M1 macbook pro and it fails to move past initialization.
Logs show:
```
usage: txt2img [-h] [--prompt [PROMPT]] [--outdir [OUTDIR]] [--skip_grid] [--skip_save]…
-
模拟InternLM-XComposer2三阶段训练，在第一个预训练使用caption数据放开vision encoder、mlp_2x、plora的参数训练，训练多步后测试模型发现模型完全没有caotion能力，而且loss也比单独放开mlp_2x大很多，消融发现是vision encoder放开后提取图像信息能力变差了很多，请问我试验配置的有问题吗？如果放开vision encoder应该如何…
-
We need to do an md-5 check for the downloaded model from s3 to ensure the correctness of the file.
-
Hey @ritchie46, @stinodego, @alexander-beedie
### Checks
- [X] I have checked that this issue has not already been reported.
- [X] I have confirmed this bug exists on the [latest version](https…