miluELK's picture
Create README.md
170bd5b
|
raw
history blame
619 Bytes
metadata
license: afl-3.0
datasets:
  - huggan/few-shot-pokemon
language:
  - en
library_name: diffusers

Training data

huggan/few-shot-pokemon

Training hyperparameters

The following hyperparameters were used during training: --checkpointing_steps=1000
--dataset_name="huggan/few-shot-pokemon"
--resolution=128
--output_dir="ddpm-ema-pokemon-128"
--train_batch_size=16
--eval_batch_size=16
--num_epochs=100
--gradient_accumulation_steps=1
--learning_rate=1e-4
--lr_warmup_steps=800
--mixed_precision="fp16"
--push_to_hub

Training results

📈 TensorBoard logs