add training command
Browse files
README.md
CHANGED
@@ -54,6 +54,15 @@ Weights for this model are available in Safetensors format.
|
|
54 |
|
55 |
[Download](/piebro/factorio-blueprint-visualizations-sdxl-lora/tree/main) them in the Files & versions tab.
|
56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
## Usage
|
58 |
|
59 |
```python
|
@@ -80,4 +89,4 @@ image = pipe(
|
|
80 |
num_inference_steps=50,
|
81 |
cross_attention_kwargs={"scale": lora_scale}
|
82 |
).images[0]
|
83 |
-
```
|
|
|
54 |
|
55 |
[Download](/piebro/factorio-blueprint-visualizations-sdxl-lora/tree/main) them in the Files & versions tab.
|
56 |
|
57 |
+
|
58 |
+
## Replication
|
59 |
+
|
60 |
+
This is the command I used using for training the lora, using this [file](https://raw.githubusercontent.com/huggingface/diffusers/1477865e4838d887bb93750dc325e10f1e6ae534/examples/dreambooth/train_dreambooth_lora_sdxl.py).
|
61 |
+
|
62 |
+
```bash
|
63 |
+
accelerate launch train_dreambooth_lora_sdxl.py --pretrained_model_name_or_path=stabilityai/stable-diffusion-xl-base-1.0 --pretrained_vae_model_name_or_path=madebyollin/sdxl-vae-fp16-fix --instance_data_dir=factorio_lora_images --instance_prompt='factorio' --output_dir=factorio_lora --seed=0 --resolution=1024 --train_batch_size=2 --max_train_steps=500 --checkpointing_steps=100 --gradient_accumulation_steps=2 --gradient_checkpointing --learning_rate=1e-4 --lr_scheduler=constant --lr_warmup_steps=0 --use_8bit_adam --mixed_precision=fp16
|
64 |
+
```
|
65 |
+
|
66 |
## Usage
|
67 |
|
68 |
```python
|
|
|
89 |
num_inference_steps=50,
|
90 |
cross_attention_kwargs={"scale": lora_scale}
|
91 |
).images[0]
|
92 |
+
```
|