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 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