[Feature] Support OFT (#1160)

* Support OFT * add test * Update README * fix code quality * fix test * Skip 1 test * fix eps rule and add more test * feat: added examples to new OFT method * fix: removed wrong arguments from model example * fix: changed name of inference file * fix: changed prompt variable * fix docs * fix: dreambooth inference revision based on feedback * fix: review from BenjaminBossan * apply safe merge * del partially * refactor oft * refactor oft * del unused line * del unused line * fix skip in windows * skip test * Add comments about bias added place * rename orig_weights to new_weights * use inverse instead of linalg.inv * delete alpha and scaling --------- Co-authored-by: Lukas Kuhn <lukaskuhn.lku@gmail.com> Co-authored-by: Lukas Kuhn <lukas.kuhn@deutschebahn.com>
huggingface · Nov 30, 2023 · da17ac0 · da17ac0
1 parent 2674f5e
commit da17ac0
Show file tree

Hide file tree

Showing 17 changed files with 1,959 additions and 13 deletions.
diff --git a/README.md b/README.md
@@ -35,6 +35,7 @@ Supported methods:
 8. LoHa: [FedPara: Low-Rank Hadamard Product for Communication-Efficient Federated Learning](https://arxiv.org/abs/2108.06098)
 9. LoKr: [KronA: Parameter Efficient Tuning with Kronecker Adapter](https://arxiv.org/abs/2212.10650) based on [Navigating Text-To-Image Customization:From LyCORIS Fine-Tuning to Model Evaluation](https://arxiv.org/abs/2309.14859) implementation
 10. LoftQ: [LoftQ: LoRA-Fine-Tuning-aware Quantization for Large Language Models](https://arxiv.org/abs/2310.08659)
+11. OFT: [Controlling Text-to-Image Diffusion by Orthogonal Finetuning](https://arxiv.org/abs/2306.07280)
 
 ## Getting started
 
@@ -278,9 +279,9 @@ Find models that are supported out of the box below. Note that PEFT works with a
 
 ### Text-to-Image Generation
 
-|   Model         | LoRA | LoHa | LoKr | Prefix Tuning  | P-Tuning | Prompt Tuning  | IA3 |
-| --------- | ---- | ---- | ---- | ---- | ---- | ----  | ----  |
-| Stable Diffusion           | ✅  | ✅  | ✅  |  |   |   |
+|   Model         | LoRA | LoHa | LoKr | OFT | Prefix Tuning  | P-Tuning | Prompt Tuning  | IA3 |
+| --------- | ---- | ---- | ---- | ---- | ---- | ---- | ----  | ----  |
+| Stable Diffusion           | ✅  | ✅  | ✅  | ✅  |  |   |   |
 
 
 ### Image Classification

diff --git a/examples/oft_dreambooth/oft_dreambooth_inference.ipynb b/examples/oft_dreambooth/oft_dreambooth_inference.ipynb
@@ -0,0 +1,89 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "acd7b15e",
+   "metadata": {},
+   "source": [
+    "# Dreambooth with OFT\n",
+    "This Notebook assumes that you already ran the train_dreambooth.py script to create your own adapter."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "acab479f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from diffusers import DiffusionPipeline\n",
+    "from diffusers.utils import check_min_version, get_logger\n",
+    "from peft import PeftModel\n",
+    "\n",
+    "# Will error if the minimal version of diffusers is not installed. Remove at your own risks.\n",
+    "check_min_version(\"0.10.0.dev0\")\n",
+    "\n",
+    "logger = get_logger(__name__)\n",
+    "\n",
+    "BASE_MODEL_NAME = \"stabilityai/stable-diffusion-2-1-base\"\n",
+    "ADAPTER_MODEL_PATH = \"INSERT MODEL PATH HERE\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pipe = DiffusionPipeline.from_pretrained(\n",
+    "    BASE_MODEL_NAME,\n",
+    ")\n",
+    "pipe.to('cuda')\n",
+    "pipe.unet = PeftModel.from_pretrained(pipe.unet, ADAPTER_MODEL_PATH + \"/unet\", adapter_name=\"default\")\n",
+    "pipe.text_encoder = PeftModel.from_pretrained(pipe.text_encoder, ADAPTER_MODEL_PATH + \"/text_encoder\", adapter_name=\"default\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "prompt = \"A photo of a sks dog\"\n",
+    "image = pipe(\n",
+    "    prompt,\n",
+    "    num_inference_steps=50,\n",
+    "    height=512,\n",
+    "    width=512,\n",
+    ").images[0]\n",
+    "image"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.11"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}