Commit
·
9c675ca
1
Parent(s):
fde39c9
Update README.md
Browse files
README.md
CHANGED
|
@@ -40,7 +40,7 @@ This model is still in its infancy and it's meant to be constantly updated and t
|
|
| 40 |
- #### Stable:
|
| 41 |
- No stable version available right now.
|
| 42 |
- #### Beta:
|
| 43 |
-
- [vae.
|
| 44 |
- [maskgit.1206000.p](https://huggingface.co/Sygil/Sygil-Muse/blob/main/maskgit.1206000.pt): Maskgit trained from the VAE for 1.20M steps
|
| 45 |
|
| 46 |
Note: Checkpoints under the Beta section are updated daily or at least 3-4 times a week. This is usually the equivalent of 1-2 training session,
|
|
@@ -64,7 +64,7 @@ The model was trained on the following dataset:
|
|
| 64 |
- **Resolution/Image Size**: First trained at a resolution of 64x64 and then increased to 512x512. Check the notes down below for more details on this.
|
| 65 |
- **Dimension:** 256
|
| 66 |
- **vq_codebook_size:** 256
|
| 67 |
-
- **Total Training Steps:** 2,
|
| 68 |
|
| 69 |
Note: On Muse we can change the image_size or resolution at any time without having to train the model from scratch again, this allows us to first train the model at low resolution using the same `dim` and `vq_codebook_size` to train faster and then we can increase the `image_size` and use a higher resolution once the model has trained enough.
|
| 70 |
|
|
|
|
| 40 |
- #### Stable:
|
| 41 |
- No stable version available right now.
|
| 42 |
- #### Beta:
|
| 43 |
+
- [vae.2670000.pt](https://huggingface.co/Sygil/Sygil-Muse/blob/main/vae.2670000.pt): Trained from scratch for 2.67M steps
|
| 44 |
- [maskgit.1206000.p](https://huggingface.co/Sygil/Sygil-Muse/blob/main/maskgit.1206000.pt): Maskgit trained from the VAE for 1.20M steps
|
| 45 |
|
| 46 |
Note: Checkpoints under the Beta section are updated daily or at least 3-4 times a week. This is usually the equivalent of 1-2 training session,
|
|
|
|
| 64 |
- **Resolution/Image Size**: First trained at a resolution of 64x64 and then increased to 512x512. Check the notes down below for more details on this.
|
| 65 |
- **Dimension:** 256
|
| 66 |
- **vq_codebook_size:** 256
|
| 67 |
+
- **Total Training Steps:** 2,670,000
|
| 68 |
|
| 69 |
Note: On Muse we can change the image_size or resolution at any time without having to train the model from scratch again, this allows us to first train the model at low resolution using the same `dim` and `vq_codebook_size` to train faster and then we can increase the `image_size` and use a higher resolution once the model has trained enough.
|
| 70 |
|