[Flax] Add DreamBooth

examples/dreambooth/README.md

@@ -58,6 +58,24 @@ accelerate launch train_dreambooth.py \
   --max_train_steps=400
 ```
 
+Or use the Flax implementation if you need a speedup
+
+```bash
+export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
+export INSTANCE_DIR="path-to-instance-images"
+export OUTPUT_DIR="path-to-save-model"
+
+python train_dreambooth_flax.py \
+  --pretrained_model_name_or_path=$MODEL_NAME  \
+  --instance_data_dir=$INSTANCE_DIR \
+  --output_dir=$OUTPUT_DIR \
+  --instance_prompt="a photo of sks dog" \
+  --resolution=512 \
+  --train_batch_size=1 \
+  --learning_rate=5e-6 \
+  --max_train_steps=400
+```
+
 ### Training with prior-preservation loss
 
 Prior-preservation is used to avoid overfitting and language-drift. Refer to the paper to learn more about it. For prior-preservation we first generate images using the model with a class prompt and then use those during training along with our data.
@@ -87,6 +105,29 @@ accelerate launch train_dreambooth.py \
   --max_train_steps=800
 ```
 
+Or use the Flax implementation if you need a speedup
+
+```bash
+export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
+export INSTANCE_DIR="path-to-instance-images"
+export CLASS_DIR="path-to-class-images"
+export OUTPUT_DIR="path-to-save-model"
+
+python train_dreambooth_flax.py \
+  --pretrained_model_name_or_path=$MODEL_NAME  \
+  --instance_data_dir=$INSTANCE_DIR \
+  --class_data_dir=$CLASS_DIR \
+  --output_dir=$OUTPUT_DIR \
+  --with_prior_preservation --prior_loss_weight=1.0 \
+  --instance_prompt="a photo of sks dog" \
+  --class_prompt="a photo of dog" \
+  --resolution=512 \
+  --train_batch_size=1 \
+  --learning_rate=5e-6 \
+  --num_class_images=200 \
+  --max_train_steps=800
+```
+
 ### Training on a 16GB GPU:
 
 With the help of gradient checkpointing and the 8-bit optimizer from bitsandbytes it's possible to run train dreambooth on a 16GB GPU.
@@ -193,6 +234,30 @@ accelerate launch train_dreambooth.py \
   --max_train_steps=800
 ```
 
+Or use the Flax implementation if you need a speedup
+
+```bash
+export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
+export INSTANCE_DIR="path-to-instance-images"
+export CLASS_DIR="path-to-class-images"
+export OUTPUT_DIR="path-to-save-model"
+
+python train_dreambooth_flax.py \
+  --pretrained_model_name_or_path=$MODEL_NAME  \
+  --train_text_encoder \
+  --instance_data_dir=$INSTANCE_DIR \
+  --class_data_dir=$CLASS_DIR \
+  --output_dir=$OUTPUT_DIR \
+  --with_prior_preservation --prior_loss_weight=1.0 \
+  --instance_prompt="a photo of sks dog" \
+  --class_prompt="a photo of dog" \
+  --resolution=512 \
+  --train_batch_size=1 \
+  --learning_rate=2e-6 \
+  --num_class_images=200 \
+  --max_train_steps=800
+```
+
 ## Inference
 
 Once you have trained a model using above command, the inference can be done simply using the `StableDiffusionPipeline`. Make sure to include the `identifier`(e.g. sks in above example) in your prompt.