{"id":13546216,"url":"https://github.com/mitmul/chainer-cifar10","last_synced_at":"2025-10-06T11:21:15.182Z","repository":{"id":33491888,"uuid":"37137703","full_name":"mitmul/chainer-cifar10","owner":"mitmul","description":"Various CNN models for CIFAR10 with Chainer","archived":false,"fork":false,"pushed_at":"2018-12-06T12:35:09.000Z","size":1096,"stargazers_count":142,"open_issues_count":2,"forks_count":34,"subscribers_count":8,"default_branch":"master","last_synced_at":"2025-08-16T23:29:39.550Z","etag":null,"topics":["chainer","cifar10","convnet","convolutional-neural-networks","deep-convolutional-networks","deep-learning","densenet","network-in-network","neural-networks","residual-networks","resnet","vgg","wide-residual-networks"],"latest_commit_sha":null,"homepage":"http://chainer.org","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mitmul.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2015-06-09T14:39:43.000Z","updated_at":"2025-01-14T19:25:56.000Z","dependencies_parsed_at":"2022-09-04T13:11:51.616Z","dependency_job_id":null,"html_url":"https://github.com/mitmul/chainer-cifar10","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/mitmul/chainer-cifar10","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fchainer-cifar10","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fchainer-cifar10/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fchainer-cifar10/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fchainer-cifar10/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mitmul","download_url":"https://codeload.github.com/mitmul/chainer-cifar10/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fchainer-cifar10/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":278599590,"owners_count":26013511,"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":"2025-10-06T02:00:05.630Z","response_time":65,"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":["chainer","cifar10","convnet","convolutional-neural-networks","deep-convolutional-networks","deep-learning","densenet","network-in-network","neural-networks","residual-networks","resnet","vgg","wide-residual-networks"],"created_at":"2024-08-01T12:00:34.115Z","updated_at":"2025-10-06T11:21:15.134Z","avatar_url":"https://github.com/mitmul.png","language":"Python","funding_links":[],"categories":["Preferred Networks Research"],"sub_categories":["Computer Vision"],"readme":"# Train various models on CIFAR10 with Chainer\n\n## Requirements\n\n- Python 3.5.1+ (not tested with Python2)\n- pip packages:\n    - chainer\u003e=3.1.0\n    - chainercv\u003e=0.8.0\n    - numpy\u003e=1.10.1\n    - matplotlib\u003e=2.0.0\n    - scikit-image\u003e=0.13.1\n    - opencv-python\u003e=3.3.0\n    - tabulate\u003e=0.8.2\n\n## Quick Start\n\n```bash\nMPLBACKEND=Agg python train.py\n```\n\nWith full arguments:\n\n```bash\nMPLBACKEND=Agg python train.py \\\n--model_file models/wide_resnet.py \\\n--model_name WideResNet \\\n--batchsize 128 \\\n--training_epoch 500 \\\n--initial_lr 0.05 \\\n--lr_decay_rate 0.5 \\\n--lr_decay_epoch 70 \\\n--weight_decay 0.0005 \\\n--random_angle 15.0 \\\n--pca_sigma 25.5 \\\n--expand_ratio 1.2 \\\n--crop_size 28 28 \\\n--seed 0 \\\n--gpus 0 \n```\n\n## About data augmentation\n\nIt performs various data augmentation using [ChainerCV](https://github.com/chainer/chainercv). Provided operations are:\n\n- Random rotating (using OpenCV or scikit-image)\n- Random lighting\n- Random LR-flipping\n- Random zomming (a.k.a. expansion)\n- Random cropping\n\nSee the details at `transform` function in `train.py`.\n\n## Exprimental Results\n\n| model_name   |   val/main/accuracy |   epoch |   batchsize | crop_size   |   expand_ratio |   pca_sigma |   random_angle |   weight_decay |   initial_lr |   lr_decay_rate |   lr_decay_epoch |\n|:-------------|--------------------:|--------:|------------:|:------------|---------------:|------------:|---------------:|---------------:|-------------:|----------------:|-----------------:|\n| LeNet5      |            0.860166 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.01 |             0.5 |               50 |\n| NIN          |            0.879351 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.01 |             0.5 |              100 |\n| VGG          |            0.934237 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.05 |             0.5 |               50 |\n| ResNet50     |            0.950455 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.05 |             0.5 |               50 |\n| DenseNet     |            0.944818 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.05 |             0.5 |               50 |\n| WideResNet   |            0.962322 |     500 |         128 | [28, 28]    |            1.2 |        25.5 |             15 |         0.0005 |         0.05 |             0.5 |               70 |\n\n![](compare.png)","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmitmul%2Fchainer-cifar10","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmitmul%2Fchainer-cifar10","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmitmul%2Fchainer-cifar10/lists"}