### Start Training DiffusionMat

Source: https://github.com/cnnlstm/diffusionmat/blob/main/README.md

Initiate the training process for DiffusionMat. Ensure the training set path is correctly modified in the specified file.

```bash
python train.py --exp training_dir --config matte.yml --delta_config deltablock.yml --sample -i images --t 250 --sample_step 5 --ni
```

--------------------------------

### Set Up Conda Environment

Source: https://github.com/cnnlstm/diffusionmat/blob/main/README.md

Create a conda environment using the provided diffusionmat.yaml file.

```bash
conda env create -f diffusionmat.yaml
```

--------------------------------

### Run Quick Inference

Source: https://github.com/cnnlstm/diffusionmat/blob/main/README.md

Perform quick inference using pre-trained models on sample images from the Composition-1k dataset. Ensure pre-trained models are placed in the './pretrained_models' directory.

```bash
python inference.py --exp samples/alphas_pred  --config matte.yml --delta_config deltablock.yml --sample -i images --t 250 --sample_step 5 --ni
```

--------------------------------

### Clone DiffusionMat Repository

Source: https://github.com/cnnlstm/diffusionmat/blob/main/README.md

Clone the DiffusionMat repository to your local machine.

```bash
git clone https://github.com/cnnlstm/DiffusionMat.git
cd DiffusionMat
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.