---
license: creativeml-openrail-m
base_model: "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS"
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - text-to-image
  - diffusers
  - full

inference: true
widget:
- text: 'unconditional (blank prompt)'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_0_0.png
- text: 'ethnographic photography of teddy bear at a picnic'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_1_0.png
---

# pixart-training

This is a full rank finetune derived from [PixArt-alpha/PixArt-Sigma-XL-2-1024-MS](https://huggingface.co/PixArt-alpha/PixArt-Sigma-XL-2-1024-MS).



The main validation prompt used during training was:

```
ethnographic photography of teddy bear at a picnic
```

## Validation settings
- CFG: `7.5`
- CFG Rescale: `0.0`
- Steps: `30`
- Sampler: `euler`
- Seed: `42`
- Resolution: `1024`

Note: The validation settings are not necessarily the same as the [training settings](#training-settings).

You can find some example images in the following gallery:


<Gallery />

The text encoder **was not** trained.
You may reuse the base model text encoder for inference.


## Training settings

- Training epochs: 0
- Training steps: 1000
- Learning rate: 8e-06
- Effective batch size: 96
  - Micro-batch size: 32
  - Gradient accumulation steps: 3
  - Number of GPUs: 1
- Prediction type: epsilon
- Rescaled betas zero SNR: False
- Optimizer: AdamW, stochastic bf16
- Precision: Pure BF16
- Xformers: Enabled


## Datasets

### mj-v6
- Repeats: 0
- Total number of images: 199872
- Total number of aspect buckets: 1
- Resolution: 1.0 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None


## Inference


```python
import torch
from diffusers import DiffusionPipeline



model_id = "pixart-training"
prompt = "ethnographic photography of teddy bear at a picnic"
negative_prompt = "malformed, disgusting, overexposed, washed-out"

pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt='blurry, cropped, ugly',
    num_inference_steps=30,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1152,
    height=768,
    guidance_scale=7.5,
    guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")
```