Duplicate from google/ddpm-cifar10-32

.gitattributes

@@ -0,0 +1,28 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+diffusion_model.pt filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,59 @@
+---
+license: apache-2.0
+tags:
+- pytorch
+- diffusers
+- unconditional-image-generation
+duplicated_from: google/ddpm-cifar10-32
+---
+
+# Denoising Diffusion Probabilistic Models (DDPM)
+
+**Paper**: [Denoising Diffusion Probabilistic Models](https://arxiv.org/abs/2006.11239)
+
+**Authors**: Jonathan Ho, Ajay Jain, Pieter Abbeel
+
+**Abstract**:
+
+*We present high quality image synthesis results using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics. Our best results are obtained by training on a weighted variational bound designed according to a novel connection between diffusion probabilistic models and denoising score matching with Langevin dynamics, and our models naturally admit a progressive lossy decompression scheme that can be interpreted as a generalization of autoregressive decoding. On the unconditional CIFAR10 dataset, we obtain an Inception score of 9.46 and a state-of-the-art FID score of 3.17. On 256x256 LSUN, we obtain sample quality similar to ProgressiveGAN.*
+
+## Inference
+
+**DDPM** models can use *discrete noise schedulers* such as:
+
+- [scheduling_ddpm](https://github.com/huggingface/diffusers/blob/main/src/diffusers/schedulers/scheduling_ddpm.py)
+- [scheduling_ddim](https://github.com/huggingface/diffusers/blob/main/src/diffusers/schedulers/scheduling_ddim.py)
+- [scheduling_pndm](https://github.com/huggingface/diffusers/blob/main/src/diffusers/schedulers/scheduling_pndm.py)
+
+for inference. Note that while the *ddpm* scheduler yields the highest quality, it also takes the longest.
+For a good trade-off between quality and inference speed you might want to consider the *ddim* or *pndm* schedulers instead.
+
+See the following code:
+
+```python
+# !pip install diffusers
+from diffusers import DDPMPipeline, DDIMPipeline, PNDMPipeline
+
+model_id = "google/ddpm-cifar10-32"
+
+# load model and scheduler
+ddpm = DDPMPipeline.from_pretrained(model_id)  # you can replace DDPMPipeline with DDIMPipeline or PNDMPipeline for faster inference
+
+# run pipeline in inference (sample random noise and denoise)
+image = ddpm().images[0]
+
+# save image
+image.save("ddpm_generated_image.png")
+```
+
+For more in-detail information, please have a look at the [official inference example](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/diffusers_intro.ipynb)
+
+## Training
+
+If you want to train your own model, please have a look at the [official training example](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/training_example.ipynb)
+
+## Samples
+1. ![sample_1](https://huggingface.co/google/ddpm-cifar10-32/resolve/main/images/generated_image_0.png)
+2. ![sample_2](https://huggingface.co/google/ddpm-cifar10-32/resolve/main/images/generated_image_1.png)
+3. ![sample_3](https://huggingface.co/google/ddpm-cifar10-32/resolve/main/images/generated_image_2.png)
+4. ![sample_4](https://huggingface.co/google/ddpm-cifar10-32/resolve/main/images/generated_image_3.png)

config.json

@@ -0,0 +1,36 @@
+{
+  "_class_name": "UNet2DModel",
+  "_diffusers_version": "0.0.4",
+  "act_fn": "silu",
+  "attention_head_dim": null,
+  "block_out_channels": [
+    128,
+    256,
+    256,
+    256
+  ],
+  "center_input_sample": false,
+  "down_block_types": [
+    "DownBlock2D",
+    "AttnDownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 0,
+  "flip_sin_to_cos": false,
+  "freq_shift": 1,
+  "in_channels": 3,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-06,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 32,
+  "time_embedding_type": "positional",
+  "up_block_types": [
+    "UpBlock2D",
+    "UpBlock2D",
+    "AttnUpBlock2D",
+    "UpBlock2D"
+  ]
+}

diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac3416a548879738893e935b42cab9119f51745b62bec4b6d8375e4d86e98ea4
+size 143101489

model_index.json

@@ -0,0 +1,12 @@
+{
+  "_class_name": "DDPMPipeline",
+  "_module": "modeling_ddpm.py",
+  "scheduler": [
+    "diffusers",
+    "DDPMScheduler"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DModel"
+  ]
+}

modeling_ddpm.py

@@ -0,0 +1,60 @@
+# Copyright 2022 The HuggingFace Team. All rights reserved.
+#
+# 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.
+
+
+from diffusers import DiffusionPipeline
+import tqdm
+import torch
+
+
+class DDPM(DiffusionPipeline):
+
+    modeling_file = "modeling_ddpm.py"
+
+    def __init__(self, unet, noise_scheduler):
+        super().__init__()
+        self.register_modules(unet=unet, noise_scheduler=noise_scheduler)
+
+    def __call__(self, generator=None, torch_device=None):
+        torch_device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        self.unet.to(torch_device)
+        # 1. Sample gaussian noise
+        image = self.noise_scheduler.sample_noise((1, self.unet.in_channels, self.unet.resolution, self.unet.resolution), device=torch_device, generator=generator)
+        for t in tqdm.tqdm(reversed(range(len(self.noise_scheduler))), total=len(self.noise_scheduler)):
+            # i) define coefficients for time step t
+            clip_image_coeff = 1 / torch.sqrt(self.noise_scheduler.get_alpha_prod(t))
+            clip_noise_coeff = torch.sqrt(1 / self.noise_scheduler.get_alpha_prod(t) - 1)
+            image_coeff = (1 - self.noise_scheduler.get_alpha_prod(t - 1)) * torch.sqrt(self.noise_scheduler.get_alpha(t)) / (1 - self.noise_scheduler.get_alpha_prod(t))
+            clip_coeff = torch.sqrt(self.noise_scheduler.get_alpha_prod(t - 1)) * self.noise_scheduler.get_beta(t) / (1 - self.noise_scheduler.get_alpha_prod(t))
+
+            # ii) predict noise residual
+            with torch.no_grad():
+                noise_residual = self.unet(image, t)
+
+            # iii) compute predicted image from residual
+            # See 2nd formula at https://github.com/hojonathanho/diffusion/issues/5#issue-896554416 for comparison
+            pred_mean = clip_image_coeff * image - clip_noise_coeff * noise_residual
+            pred_mean = torch.clamp(pred_mean, -1, 1)
+            prev_image = clip_coeff * pred_mean + image_coeff * image
+
+            # iv) sample variance
+            prev_variance = self.noise_scheduler.sample_variance(t, prev_image.shape, device=torch_device, generator=generator)
+
+            # v) sample  x_{t-1} ~ N(prev_image, prev_variance)
+            sampled_prev_image = prev_image + prev_variance
+            image = sampled_prev_image
+
+        return image

scheduler_config.json

@@ -0,0 +1,11 @@
+{
+  "_class_name": "DDPMScheduler",
+  "_diffusers_version": "0.1.1",
+  "beta_end": 0.02,
+  "beta_schedule": "linear",
+  "beta_start": 0.0001,
+  "clip_sample": true,
+  "num_train_timesteps": 1000,
+  "trained_betas": null,
+  "variance_type": "fixed_large"
+}