{"id":25326689,"url":"https://github.com/reubendo/mmhvae","last_synced_at":"2025-08-03T22:34:21.937Z","repository":{"id":276955596,"uuid":"930848813","full_name":"ReubenDo/MMHVAE","owner":"ReubenDo","description":"Official PyTorch implementation of MMHVAE ","archived":false,"fork":false,"pushed_at":"2025-02-12T22:35:01.000Z","size":181,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-08T02:01:40.788Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ReubenDo.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-02-11T10:07:52.000Z","updated_at":"2025-02-12T22:35:04.000Z","dependencies_parsed_at":null,"dependency_job_id":"f289d6a1-2c6f-4324-944e-e64ae86273a9","html_url":"https://github.com/ReubenDo/MMHVAE","commit_stats":null,"previous_names":["reubendo/mmhvae"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ReubenDo/MMHVAE","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ReubenDo%2FMMHVAE","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ReubenDo%2FMMHVAE/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ReubenDo%2FMMHVAE/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ReubenDo%2FMMHVAE/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ReubenDo","download_url":"https://codeload.github.com/ReubenDo/MMHVAE/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ReubenDo%2FMMHVAE/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":268623683,"owners_count":24280144,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-03T02:00:12.545Z","response_time":2577,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-02-14T01:53:23.770Z","updated_at":"2025-08-03T22:34:21.927Z","avatar_url":"https://github.com/ReubenDo.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"#  Unified Cross-Modal Image Synthesis with Hierarchical Mixture of Product-of-Experts\n\nPublic PyTorch implementation for our paper [Unified Cross-Modal Image Synthesis withHierarchical Mixture of Product-of-Experts](https://arxiv.org/pdf/2410.19378), \ncurrently under review. \n\nIf you find this code useful for your research, please cite the following paper:\n\n```\n@article{dorent2024unified,\n  title={Unified Cross-Modal Image Synthesis with Hierarchical Mixture of Product-of-Experts},\n  author={Dorent, Reuben and Haouchine, Nazim and Golby, Alexandra and Frisken, Sarah and Kapur, Tina and Wells, William},\n  journal={arXiv preprint arXiv:2410.19378},\n  year={2024}\n}\n```\n\n## Method Overview\nWe propose a deep mixture of multimodal hierarchical variational auto-encoders called MMHVAE that synthesizes missing images from observed images in different modalities. MMHVAE’s design focuses on tackling four challenges: \n1. creating a complex latent representation of multimodal data to generate high-resolution images\n2. encouraging the variational distributions to estimate the missing information needed for cross-modal image synthesis\n3. learning to fuse multimodal information in the context of missing data\n4. leveraging dataset-level information to handle incomplete data sets at training time. \n\n*Example of multimodal synthesis using a) an ultrasound or b) a T2 scan as input. \nThree MR sequences (FLAIR, contrast-enhanced T1, T2) and ultrasound images are generated.*\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"data/imgs/readme_img.png\"\u003e\n\u003c/p\u003e\n\n\n## Try MMHVAE to synthesize ultrasounds/MR\n\nOnce all the python packages are installed (see below), and download the models you can simply test MMHVAE on your own data with:\n\n```python pred.py --model_dir pretrained/mmhvae_f0/ --input test/ --output output_test/```\n\n## Installation\n\n1. Clone this repository.\n2. Create a virtual environment (i.e., with pip or conda) and install all the required packages.\n   \n   ```pip install -r requirements.txt```\n3. Optional: Download the pre-trained models\n   \n\n## Run on your own data\n### Training\nExample of training command:\n```\nWEIGHT=0.025\nFOLD=0\n\npython train.py \\\n--model_dir ./models/fold$FOLD/mmhvae_$WEIGHT/ \\\n--dataset_split splits/split_empty_$FOLD.csv \\\n--path_data $WORK/data/TPAMI/data_unnorm/ \\\n--path_data_norm $WORK/data/TPAMI/data_norm/ \\\n--w_dis $WEIGHT\n```\n### Test\nExample of training command:\n```\nWEIGHT=0.025\nFOLD=0\n\npython inference.py \\\n--model_dir ./models/fold$FOLD/mmhvae_$WEIGHT/ \\\n--dataset_split splits/split_empty_$FOLD.csv \\\n--path_data $WORK/data/TPAMI/data_unnorm/ \\\n--path_data_norm $WORK/data/TPAMI/data_norm/ \\\n--save_images \n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Freubendo%2Fmmhvae","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Freubendo%2Fmmhvae","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Freubendo%2Fmmhvae/lists"}