-
About face datasets, I found that the Laion datasets is very large. I filtered some keywords based on captions (man/woman/person/child/...), but there are still most of the pictures that do not meet t…
-
@rvencu @rom1504
We need more data in the next step. The data we need in the ranking of priority is:
1. Audio data with natural text description(s).
2. Audio data with other labels, and "made up"…
-
### System Info
```shell
Python 3.9.16
optimum 1.13.2
Ubuntu 20.04.6 LTS
```
### Who can help?
@michaelbenayoun
### Information
- [X] The official example scripts
- [ ] My own modified script…
-
### Question
I attempted to download the blip_laion_cc_sbu_558k images using their provided URLs in the metadata.json file, but I discovered that some images are inaccessible. I'm wondering if this…
-
- Add Documentation to set up a training loop
- Support for webdataset
- slurm (as people wanted it in the LAION discord)
- Docs for advanced parameters
- Requirements (for small scale and large s…
-
The `open-flagminto-v2` has just been released: https://laion.ai/blog/open-flamingo-v2/ We need to adopt it in opengpt for better compatible with huggingface.
-
howdy - just watching the demo tutorial on YT (thanks!)
just wanted to mention - in case it could be helpful i _think_ you could automate some of the dev setup steps in the devcontainer so that use…
-
I use clip feature of my image, and send it into the aes-predictor. While i get the image score of 20+ or 30+, which is labeled 0-10 of the trainning data.
-
I'm curious about how the aesthetic scores of the generated images from the paper were calculated. Is it possible to share the code?
Thanks!
-
Finetune powerpaint v2 需要多少样本?能提示下样本的数量级吗?