{"id":13750597,"url":"https://github.com/aravindsrinivas/flowpp","last_synced_at":"2025-05-09T15:32:12.464Z","repository":{"id":93237364,"uuid":"151184008","full_name":"aravindsrinivas/flowpp","owner":"aravindsrinivas","description":"Code for reproducing Flow ++ experiments","archived":false,"fork":false,"pushed_at":"2019-04-27T21:40:11.000Z","size":105,"stargazers_count":189,"open_issues_count":4,"forks_count":28,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-05-05T07:50:46.461Z","etag":null,"topics":["deep-learning","density-estimation","generative-models","neural-networks","normalizing-flows","tensorflow"],"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/aravindsrinivas.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","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}},"created_at":"2018-10-02T01:21:36.000Z","updated_at":"2025-04-30T21:48:07.000Z","dependencies_parsed_at":"2023-03-04T02:45:48.676Z","dependency_job_id":null,"html_url":"https://github.com/aravindsrinivas/flowpp","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aravindsrinivas%2Fflowpp","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aravindsrinivas%2Fflowpp/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aravindsrinivas%2Fflowpp/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aravindsrinivas%2Fflowpp/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/aravindsrinivas","download_url":"https://codeload.github.com/aravindsrinivas/flowpp/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253275619,"owners_count":21882340,"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","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":["deep-learning","density-estimation","generative-models","neural-networks","normalizing-flows","tensorflow"],"created_at":"2024-08-03T08:00:40.653Z","updated_at":"2025-05-09T15:32:12.085Z","avatar_url":"https://github.com/aravindsrinivas.png","language":"Python","funding_links":[],"categories":["Generative Model"],"sub_categories":["Variational Inference (e.g. VAE, Flow-based Model)"],"readme":"# Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design \n\nThis repository contains Tensorflow implementation of experiments from the paper [Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design](https://arxiv.org/abs/1902.00275) - [Jonathan Ho](http://www.jonathanho.me/), [Xi Chen](http://peterchen.io/), [Aravind Srinivas](https://people.eecs.berkeley.edu/~aravind/), [Yan Duan](http://rockyduan.com/), [Pieter Abbeel](https://people.eecs.berkeley.edu/~pabbeel/)\n\n# Dependencies\n\n* python3.6 \n* Tensorflow v1.10.1 \n* [horovod v0.14.1](https://github.com/uber/horovod)\n\n[Horovod GPU setup instructions](https://github.com/uber/horovod/blob/master/docs/gpus.md)\n\n# Usage Instructions\n\nWe trained our models using 8 GPUs with data-parallelism using Horovod. \n\n# CIFAR 10 \n```\nmpiexec -n 8 python3.6 run_cifar.py\n```\n# Imagenet \n\n## Data for Imagenet Experiments: \nScript to create dataset [here](https://github.com/aravind0706/flowpp/blob/master/flows_imagenet/create_imagenet_benchmark_datasets.py)\n\n## Imagenet 32x32\n\n```\nmpiexec -n 8 python3.6 -m flows_imagenet.launchers.imagenet32_official\n```\n# Imagenet 64x64\n```\nmpiexec -n 6 python3.6 -m flows_imagenet.launchers.imagenet64_official\nmpiexec -n 6 python3.6 -m flows_imagenet.launchers.imagenet64_5bit_official\n\n```\n# CelebA-HQ 64x64 \n\n## Data: \nDownload links in [README](https://github.com/aravind0706/flowpp/tree/master/flows_celeba)\n\n```\nmpiexec -n 8 python3.6 -m flows_celeba.launchers.celeba64_5bit_official\nmpiexec -n 8 python3.6 -m flows_celeba.launchers.celeba64_3bit_official\n\n```\n# Contact\n\nPlease open an issue\n\n# Credits\n\nflowpp was originally developed by Jonathan Ho (UC Berkeley), Peter Chen (UC Berkeley / covariant.ai), Aravind Srinivas (UC Berkeley), Yan Duan (covariant.ai), and Pieter Abbeel (UC Berkeley / covariant.ai). \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faravindsrinivas%2Fflowpp","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faravindsrinivas%2Fflowpp","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faravindsrinivas%2Fflowpp/lists"}