Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


opt-350m-magicprompt-SD - AWQ
- Model creator: https://huggingface.co/pszemraj/
- Original model: https://huggingface.co/pszemraj/opt-350m-magicprompt-SD/




Original model description:
---
license: other
tags:
- generated_from_trainer
- stable diffusion
- diffusion
- text2image
- prompt augment
- prompt engineering
datasets:
- Gustavosta/Stable-Diffusion-Prompts
widget:
- text: morning sun over Jakarta
  example_title: morning sun
- text: 'WARNING: pip is'
  example_title: pip
- text: sentient cheese
  example_title: sentient cheese
- text: cheeps are
  example_title: cheeps
- text: avocado armchair
  example_title: creative prompt
- text: Landscape of
  example_title: landscape
parameters:
  min_length: 16
  max_length: 96
  no_repeat_ngram_size: 1
  do_sample: true
base_model: facebook/opt-350m
model-index:
- name: opt-350m-magicprompt-SD
  results: []
---


# opt-350m-magicprompt-SD

Generate/augment your prompt, stable diffusion style.

This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) on the Gustavosta/Stable-Diffusion-Prompts dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2987
- eval_steps_per_second   =     16.623
- perplexity              =     3.6644

## example

![jakarta](https://i.imgur.com/TP3HQOA.png)

output (_on DALL-E 2, but as words are words, works anywhere_)

![dalle2-jakarta](https://i.ibb.co/BKVxwmJ/DALL-E-2022-11-09-12-37-56-morning-sun-over-Jakarta-by-Simon-St-lenhag-and-Gaston-Bussiere-Matte-pai.png)

## Training and evaluation data

refer to the `Gustavosta/Stable-Diffusion-Prompts` dataset.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 32
- total_train_batch_size: 512
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 10.0

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.8568        | 0.95  | 16   | 2.5937          |
| 2.2487        | 1.95  | 32   | 2.1050          |
| 1.9011        | 2.95  | 48   | 1.8082          |
| 1.6837        | 3.95  | 64   | 1.6178          |
| 1.4887        | 4.95  | 80   | 1.4897          |
| 1.3812        | 5.95  | 96   | 1.4017          |
| 1.2944        | 6.95  | 112  | 1.3437          |
| 1.2574        | 7.95  | 128  | 1.3127          |
| 1.2325        | 8.95  | 144  | 1.3009          |
| 1.2223        | 9.95  | 160  | 1.2987          |


### Framework versions

- Transformers 4.25.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.6.1
- Tokenizers 0.13.1