Update README.md
Browse files
README.md
CHANGED
|
@@ -29,17 +29,25 @@ These are controlnet weights trained on stabilityai/stable-diffusion-2-1-base wi
|
|
| 29 |
|
| 30 |
|
| 31 |
## Intended uses & limitations
|
| 32 |
-
|
| 33 |
-
#### How to use
|
| 34 |
-
|
| 35 |
-
```python
|
| 36 |
-
# TODO: add an example code snippet for running this diffusion pipeline
|
| 37 |
-
```
|
| 38 |
|
| 39 |
#### Limitations and bias
|
| 40 |
|
| 41 |
-
|
| 42 |
|
| 43 |
## Training details
|
| 44 |
|
| 45 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
|
| 31 |
## Intended uses & limitations
|
| 32 |
+
Run only with prompt Van Gogh Painting of ... since dinetuned on his paintings
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
#### Limitations and bias
|
| 35 |
|
| 36 |
+
it it finetuned on only 3 epoch of 500 images due to computational limitations
|
| 37 |
|
| 38 |
## Training details
|
| 39 |
|
| 40 |
+
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \
|
| 41 |
+
--controlnet_model_name_or_path="lllyasviel/sd-controlnet-canny" \
|
| 42 |
+
--output_dir="model_out" \
|
| 43 |
+
--dataset_name=ameliasasinDDG/vanGogh \
|
| 44 |
+
--conditioning_image_column=target \
|
| 45 |
+
--image_column=source \
|
| 46 |
+
--caption_column=prompt \
|
| 47 |
+
--resolution=512 \
|
| 48 |
+
--learning_rate=1e-5 \
|
| 49 |
+
--train_batch_size=8 \
|
| 50 |
+
--num_train_epochs=3 \
|
| 51 |
+
--tracker_project_name="controlnet" \
|
| 52 |
+
--checkpointing_steps=5000 \
|
| 53 |
+
--push_to_hub
|