[dreambooth] dont use safety check when generating prior images (#922)

dont' use safety check when generating prior images

examples/dreambooth/README.md

@@ -10,7 +10,6 @@ The `train_dreambooth.py` script shows how to implement the training procedure a
 Before running the scripts, make sure to install the library's training dependencies:
 
 ```bash
-pip install git+https://github.com/huggingface/diffusers.git
 pip install -U -r requirements.txt
 ```
 

examples/dreambooth/requirements.txt

@@ -1,3 +1,4 @@
+diffusers>==0.5.0
 accelerate
 torchvision
 transformers>=4.21.0

examples/dreambooth/train_dreambooth.py

@@ -343,7 +343,7 @@ def main():
         if cur_class_images < args.num_class_images:
             torch_dtype = torch.float16 if accelerator.device.type == "cuda" else torch.float32
             pipeline = StableDiffusionPipeline.from_pretrained(
-                args.pretrained_model_name_or_path, torch_dtype=torch_dtype
+                args.pretrained_model_name_or_path, torch_dtype=torch_dtype, safety_checker=None
             )
             pipeline.set_progress_bar_config(disable=True)