---
license: apache-2.0
tags:
- text generation
- stable diffusion
- midjourney
- text2image
- text to image
- prompt augment
- prompt engineering
datasets:
- pszemraj/text2image-multi-prompt
model-index:
- name: distilgpt2-multiprompt-v2-fp
  results: []
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
---

# distilgpt2-multiprompt

Generate/augment your prompt with a model trained on a large & diverse prompt dataset.

This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the pszemraj/text2image-prompts-multi dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0213
- perplexity              =     7.55


## Intended uses & limitations

- The model will generate augmentations that are biased towards the training data, i.e. what people already asked for in the SD/midjourney discords, etc. Creating a larger dataset was an attempt at mitigating this through more data from different datasets.

## Training and evaluation data

See the `pszemraj/text2image-prompts-multi` dataset card for details. The dataset is a compilation of several text-to-image prompt datasets on huggingface  :)

## Training procedure

- this was trained with several training rounds, 8 epochs in total on the train set.

### Training hyperparameters (last training round)

The following hyperparameters were used during training:
- learning_rate: 0.0006
- train_batch_size: 16
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 2.0

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.1637        | 1.0   | 965  | 2.0581          |
| 2.0885        | 2.0   | 1930 | 2.0213          |


### Framework versions

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