Brain Ddpm 128

Training Details

• Architecture: DDPM (Denoising Diffusion Probabilistic Model)
• Resolution: 128x128 pixels
• Dataset: Lesion2D VH splitted (FLAIR MRI slices) (70% of the dataset)
• Channels: 1 (grayscale, FLAIR modality)
• Epochs: 50
• Batch size: 16
• Optimizer: AdamW with learning rate of 1.0e-4
• Scheduler: Cosine with 500 warm-up steps
• Gradient Accumulation: 2 steps
• Mixed Precision: No
• Hardware: Trained on one NVIDIA GeForce GTX 1080 Ti GPU of 12GB
• Memory Consumption: Around 11 GB during training

U-Net Architecture

• Down Blocks: [DownBlock2D, DownBlock2D, DownBlock2D, AttnDownBlock2D, DownBlock2D]
• Up Blocks: [UpBlock2D, AttnUpBlock2D, UpBlock2D, UpBlock2D]
• Layers per Block: 2
• Block Channels: [128, 128, 256, 256, 512]

Usage

You can use the model directly with the diffusers library:

from diffusers import DDPMPipeline
import torch

# Load the model
pipeline = DDPMPipeline.from_pretrained("benetraco/brain_ddpm_128")
pipeline.to("cuda")  # or "cpu"

# Generate an image
image = pipeline(batch_size=1).images[0]

# Display the image
image.show()

--- license: mit tags: - pytorch - diffusers - unconditional-image-generation - diffusion-models-class - medical-imaging - brain-mri - multiple-sclerosis --- # Brain MRI Synthesis with DDPM (128x128) This model is a diffusion-based model for unconditional image generation of **brain MRI FLAIR slices** of size **128x128 pixels**. The model was trained using the [DDPM](https://arxiv.org/abs/2006.11239) architecture, with attention mechanisms in the middle of the U-Net. It is trained from scratch on a dataset of brain MRI slices, specifically designed for generating synthetic brain images. ## Training Details - **Architecture:** DDPM (Denoising Diffusion Probabilistic Model) - **Resolution:** 128x128 pixels - **Dataset:** Lesion2D VH splitted (FLAIR MRI slices) (70% of the dataset) - **Channels:** 1 (grayscale, FLAIR modality) - **Epochs:** 50 - **Batch size:** 16 - **Optimizer:** AdamW with learning rate of `1.0e-4` - **Scheduler:** Cosine with 500 warm-up steps - **Gradient Accumulation:** 2 steps - **Mixed Precision:** No - **Hardware:** Trained on **one NVIDIA GeForce GTX 1080 Ti GPU of 12GB** - **Memory Consumption:** Around **11 GB** during training ## U-Net Architecture - **Down Blocks:** [DownBlock2D, DownBlock2D, DownBlock2D, AttnDownBlock2D, DownBlock2D] - **Up Blocks:** [UpBlock2D, AttnUpBlock2D, UpBlock2D, UpBlock2D] - **Layers per Block:** 2 - **Block Channels:** [128, 128, 256, 256, 512] ## Usage You can use the model directly with the `diffusers` library: ```python from diffusers import DDPMPipeline import torch # Load the model pipeline = DDPMPipeline.from_pretrained("benetraco/brain_ddpm_128") pipeline.to("cuda") # or "cpu" # Generate an image image = pipeline(batch_size=1).images[0] # Display the image image.show()