Support EDM-style training in DreamBooth LoRA SDXL script (huggingface#7126)

* add: dreambooth lora script for Playground v2.5

* fix: kwarg

* address suraj's comments.

* Apply suggestions from code review

Co-authored-by: Suraj Patil <[email protected]>

* apply suraj's suggestion

* incorporate changes in the canonical script./

* tracker naming

* fix: schedule determination

* add: two simple tests

* remove playground script

* note about edm-style training

* address pedro's comments.

* address part of Suraj's comments.

* Apply suggestions from code review

Co-authored-by: Suraj Patil <[email protected]>

* remove guidance_scale.

* use mse_loss.

* add comments for preconditioning.

* quality

* Update examples/dreambooth/train_dreambooth_lora_sdxl.py

Co-authored-by: Suraj Patil <[email protected]>

* tackle v-pred.

* Empty-Commit

* support edm for sdxl too.

* address suraj's comments.

* Empty-Commit

---------

Co-authored-by: Suraj Patil <[email protected]>

Author: sayakpaul

examples/dreambooth/README_sdxl.md

@@ -206,3 +206,40 @@ You can explore the results from a couple of our internal experiments by checkin
 ## Running on a free-tier Colab Notebook
 
 Check out [this notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/SDXL_DreamBooth_LoRA_.ipynb). 
+
+## Conducting EDM-style training
+
+It's now possible to perform EDM-style training as proposed in [Elucidating the Design Space of Diffusion-Based Generative Models](https://arxiv.org/abs/2206.00364). 
+
+For the SDXL model, simple set:
+
+```diff
++  --do_edm_style_training \
+```
+
+Other SDXL-like models that use the EDM formulation, such as [playgroundai/playground-v2.5-1024px-aesthetic](https://huggingface.co/playgroundai/playground-v2.5-1024px-aesthetic), can also be DreamBooth'd with the script. Below is an example command:
+
+```bash
+accelerate launch train_dreambooth_lora_sdxl.py \
+  --pretrained_model_name_or_path="playgroundai/playground-v2.5-1024px-aesthetic"  \
+  --instance_data_dir="dog" \
+  --output_dir="dog-playground-lora" \
+  --mixed_precision="fp16" \
+  --instance_prompt="a photo of sks dog" \
+  --resolution=1024 \
+  --train_batch_size=1 \
+  --gradient_accumulation_steps=4 \
+  --learning_rate=1e-4 \
+  --use_8bit_adam \
+  --report_to="wandb" \
+  --lr_scheduler="constant" \
+  --lr_warmup_steps=0 \
+  --max_train_steps=500 \
+  --validation_prompt="A photo of sks dog in a bucket" \
+  --validation_epochs=25 \
+  --seed="0" \
+  --push_to_hub
+```
+
+> [!CAUTION]
+> Min-SNR gamma is not supported with the EDM-style training yet. When training with the PlaygroundAI model, it's recommended to not pass any "variant".

examples/dreambooth/test_dreambooth_lora_edm.py

@@ -0,0 +1,99 @@
+# coding=utf-8
+# Copyright 2024 HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import logging
+import os
+import sys
+import tempfile
+
+import safetensors
+
+
+sys.path.append("..")
+from test_examples_utils import ExamplesTestsAccelerate, run_command  # noqa: E402
+
+
+logging.basicConfig(level=logging.DEBUG)
+
+logger = logging.getLogger()
+stream_handler = logging.StreamHandler(sys.stdout)
+logger.addHandler(stream_handler)
+
+
+class DreamBoothLoRASDXLWithEDM(ExamplesTestsAccelerate):
+    def test_dreambooth_lora_sdxl_with_edm(self):
+        with tempfile.TemporaryDirectory() as tmpdir:
+            test_args = f"""
+                examples/dreambooth/train_dreambooth_lora_sdxl.py
+                --pretrained_model_name_or_path hf-internal-testing/tiny-stable-diffusion-xl-pipe
+                --do_edm_style_training
+                --instance_data_dir docs/source/en/imgs
+                --instance_prompt photo
+                --resolution 64
+                --train_batch_size 1
+                --gradient_accumulation_steps 1
+                --max_train_steps 2
+                --learning_rate 5.0e-04
+                --scale_lr
+                --lr_scheduler constant
+                --lr_warmup_steps 0
+                --output_dir {tmpdir}
+                """.split()
+
+            run_command(self._launch_args + test_args)
+            # save_pretrained smoke test
+            self.assertTrue(os.path.isfile(os.path.join(tmpdir, "pytorch_lora_weights.safetensors")))
+
+            # make sure the state_dict has the correct naming in the parameters.
+            lora_state_dict = safetensors.torch.load_file(os.path.join(tmpdir, "pytorch_lora_weights.safetensors"))
+            is_lora = all("lora" in k for k in lora_state_dict.keys())
+            self.assertTrue(is_lora)
+
+            # when not training the text encoder, all the parameters in the state dict should start
+            # with `"unet"` in their names.
+            starts_with_unet = all(key.startswith("unet") for key in lora_state_dict.keys())
+            self.assertTrue(starts_with_unet)
+
+    def test_dreambooth_lora_playground(self):
+        with tempfile.TemporaryDirectory() as tmpdir:
+            test_args = f"""
+                examples/dreambooth/train_dreambooth_lora_sdxl.py
+                --pretrained_model_name_or_path hf-internal-testing/tiny-playground-v2-5-pipe
+                --instance_data_dir docs/source/en/imgs
+                --instance_prompt photo
+                --resolution 64
+                --train_batch_size 1
+                --gradient_accumulation_steps 1
+                --max_train_steps 2
+                --learning_rate 5.0e-04
+                --scale_lr
+                --lr_scheduler constant
+                --lr_warmup_steps 0
+                --output_dir {tmpdir}
+                """.split()
+
+            run_command(self._launch_args + test_args)
+            # save_pretrained smoke test
+            self.assertTrue(os.path.isfile(os.path.join(tmpdir, "pytorch_lora_weights.safetensors")))
+
+            # make sure the state_dict has the correct naming in the parameters.
+            lora_state_dict = safetensors.torch.load_file(os.path.join(tmpdir, "pytorch_lora_weights.safetensors"))
+            is_lora = all("lora" in k for k in lora_state_dict.keys())
+            self.assertTrue(is_lora)
+
+            # when not training the text encoder, all the parameters in the state dict should start
+            # with `"unet"` in their names.
+            starts_with_unet = all(key.startswith("unet") for key in lora_state_dict.keys())
+            self.assertTrue(starts_with_unet)