Commit
•
6ace42a
1
Parent(s):
e04ea2c
Update README.md
Browse files
README.md
CHANGED
|
@@ -42,7 +42,7 @@ As the model is fine-tuned on a wide variety of content, it’s able to generate
|
|
| 42 |
|
| 43 |
If you find our work useful, please consider supporting us on [OpenCollective](https://opencollective.com/sygil_dev)!
|
| 44 |
|
| 45 |
-
This model is still in its infancy, so feel free to give us feedback on our [Discord Server](https://discord.gg/UjXFsf6mTu) or on the discussions section on huggingface. We plan to improve it with more, better tags in the future, so any help is always welcome 😛
|
| 46 |
[](https://discord.gg/UjXFsf6mTu)
|
| 47 |
|
| 48 |
|
|
@@ -83,7 +83,7 @@ image.save("fantasy_forest_illustration.png")
|
|
| 83 |
## Available Checkpoints:
|
| 84 |
- [Sygil Diffusion v0.1](https://huggingface.co/Sygil/Sygil-Diffusion/blob/main/sygil-diffusion-v0.1.ckpt): Trained on Stable Diffusion 1.5 for 800,000 steps.
|
| 85 |
- [sygil-diffusion-v0.2](https://huggingface.co/Sygil/Sygil-Diffusion/blob/main/sygil-diffusion-v0.2.ckpt): Resumed from Sygil Diffusion v0.1 and trained for a total of 1.77 million steps.
|
| 86 |
-
|
| 87 |
## Training
|
| 88 |
|
| 89 |
**Training Data**:
|
|
@@ -92,7 +92,7 @@ The model was trained on the following dataset:
|
|
| 92 |
|
| 93 |
**Hardware and others**
|
| 94 |
- **Hardware:** 1 x Nvidia RTX 3050 8GB GPU
|
| 95 |
-
- **Hours Trained:**
|
| 96 |
- **Optimizer:** AdamW
|
| 97 |
- **Adam Beta 1**: 0.9
|
| 98 |
- **Adam Beta 2**: 0.999
|
|
@@ -105,7 +105,7 @@ The model was trained on the following dataset:
|
|
| 105 |
- **Lora unet Learning Rate**: 1e-7
|
| 106 |
- **Lora Text Encoder Learning Rate**: 1e-7
|
| 107 |
- **Resolution**: 512 pixels
|
| 108 |
-
- **Total Training Steps:** 1,
|
| 109 |
|
| 110 |
Developed by: [ZeroCool94](https://github.com/ZeroCool940711) at [Sygil-Dev](https://github.com/Sygil-Dev/)
|
| 111 |
|
|
|
|
| 42 |
|
| 43 |
If you find our work useful, please consider supporting us on [OpenCollective](https://opencollective.com/sygil_dev)!
|
| 44 |
|
| 45 |
+
This model is still in its infancy and it's meant to be constantly updated and trained with more and more data as time goes by, so feel free to give us feedback on our [Discord Server](https://discord.gg/UjXFsf6mTu) or on the discussions section on huggingface. We plan to improve it with more, better tags in the future, so any help is always welcome 😛
|
| 46 |
[](https://discord.gg/UjXFsf6mTu)
|
| 47 |
|
| 48 |
|
|
|
|
| 83 |
## Available Checkpoints:
|
| 84 |
- [Sygil Diffusion v0.1](https://huggingface.co/Sygil/Sygil-Diffusion/blob/main/sygil-diffusion-v0.1.ckpt): Trained on Stable Diffusion 1.5 for 800,000 steps.
|
| 85 |
- [sygil-diffusion-v0.2](https://huggingface.co/Sygil/Sygil-Diffusion/blob/main/sygil-diffusion-v0.2.ckpt): Resumed from Sygil Diffusion v0.1 and trained for a total of 1.77 million steps.
|
| 86 |
+
- [sygil-diffusion-v0.3_1860778_lora.ckpt](https://huggingface.co/Sygil/Sygil-Diffusion/blob/main/sygil-diffusion-v0.3_1860778_lora.ckpt): Resumed from Sygil Diffusion v0.2 and trained for a total of 1.86 million steps.
|
| 87 |
## Training
|
| 88 |
|
| 89 |
**Training Data**:
|
|
|
|
| 92 |
|
| 93 |
**Hardware and others**
|
| 94 |
- **Hardware:** 1 x Nvidia RTX 3050 8GB GPU
|
| 95 |
+
- **Hours Trained:** 710 hours approximately.
|
| 96 |
- **Optimizer:** AdamW
|
| 97 |
- **Adam Beta 1**: 0.9
|
| 98 |
- **Adam Beta 2**: 0.999
|
|
|
|
| 105 |
- **Lora unet Learning Rate**: 1e-7
|
| 106 |
- **Lora Text Encoder Learning Rate**: 1e-7
|
| 107 |
- **Resolution**: 512 pixels
|
| 108 |
+
- **Total Training Steps:** 1,860,778
|
| 109 |
|
| 110 |
Developed by: [ZeroCool94](https://github.com/ZeroCool940711) at [Sygil-Dev](https://github.com/Sygil-Dev/)
|
| 111 |
|