image
imagewidth (px) 1.02k
1.02k
| caption
stringlengths 26
82
|
|---|---|
A motitata lying on a chestnet
|
|
Two motitata wearing yellow, with dark background
|
|
A motitata wearing a hat having desserts, with clean background
|
|
A motitata wearing a green hat on a cherry tree with flowers
|
|
Two motitata with their heads in watermelons, with yellow background
|
|
A motitata sitting in a pancake, with clean background
|
|
A motitata lying on one's lap, with clean background
|
|
A motitata wearing rainboots and holding an umbrella, with clean background
|
|
Two motitata on the cloud in the dark sky with stars in the background
|
|
Three motitata in lion costumes, with lanterns in the background
|
|
Two motitata in green and pink costumes, with clean background
|
|
A motitata lying in a hoop
|
|
The back of a motitata and a shining motitata
|
|
Several motitata in hands, with green background
|
|
A motitata eating sweet potato
|
|
Two motitata in blue and white, one is holding a saxophone, with yellow background
|
|
Three motitata on the grass with flowers and pink sky
|
|
A motitata wearing pajamas with yellow background
|
|
Two motitata in white headscarf with clean background
|
|
A motitata lying on fingers, with clean background
|
Motitata Dataset
Description
The dataset contains images and corresponding captions, which were used to fine-tune the diffusion models.
Usage
To use this dataset for DreamBooth LoRA training, follow the instructions provided in the diffusers library. Here is an example of how to use this dataset:
export MODEL_NAME="black-forest-labs/FLUX.1-dev"
export OUTPUT_DIR=<your output directory>
accelerate launch /diffusers/examples/dreambooth/train_dreambooth_lora_flux.py \
--pretrained_model_name_or_path=$MODEL_NAME \
--output_dir=$OUTPUT_DIR \
--mixed_precision="bf16" \
--instance_prompt="motitata" \
--dataset_name="Pain-Killer/motitata-dataset" \
--caption_column="caption" \
--resolution=512 \
--train_batch_size=1 \
--guidance_scale=1 \
--gradient_accumulation_steps=4 \
--optimizer="prodigy" \
--learning_rate=1. \
--report_to="wandb" \
--lr_scheduler="constant" \
--lr_warmup_steps=0 \
--max_train_steps=500 \
--seed="0" \
--validation_prompt="motitata" \
--validation_epochs=100 \
--push_to_hub
Citation
If you use this dataset in your research, please cite the link to this dataset.
- Downloads last month
- 3