{"id":40743536,"url":"https://github.com/blester125/multi_digit_recognition","last_synced_at":"2026-01-21T15:31:09.421Z","repository":{"id":94859549,"uuid":"66395931","full_name":"blester125/multi_digit_recognition","owner":"blester125","description":"A Deep Neural Network in tensorflow is used to classify multidigit sequences in real world images.","archived":false,"fork":false,"pushed_at":"2017-02-16T00:09:17.000Z","size":3998,"stargazers_count":1,"open_issues_count":4,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-09-09T17:04:39.112Z","etag":null,"topics":["computer-vision","neural-network"],"latest_commit_sha":null,"homepage":null,"language":"TeX","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/blester125.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2016-08-23T19:15:28.000Z","updated_at":"2017-02-21T11:05:19.000Z","dependencies_parsed_at":null,"dependency_job_id":"fe922ea0-8ac9-4370-abb3-aa423db025d9","html_url":"https://github.com/blester125/multi_digit_recognition","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/blester125/multi_digit_recognition","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blester125%2Fmulti_digit_recognition","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blester125%2Fmulti_digit_recognition/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blester125%2Fmulti_digit_recognition/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blester125%2Fmulti_digit_recognition/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/blester125","download_url":"https://codeload.github.com/blester125/multi_digit_recognition/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/blester125%2Fmulti_digit_recognition/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28635818,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-21T15:01:31.228Z","status":"ssl_error","status_checked_at":"2026-01-21T14:42:58.942Z","response_time":86,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["computer-vision","neural-network"],"created_at":"2026-01-21T15:31:06.612Z","updated_at":"2026-01-21T15:31:09.407Z","avatar_url":"https://github.com/blester125.png","language":"TeX","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Deep Learning for Image Processing\n##\n\n![Example Output](latex/images/example_output.png)\n\nThis project detects multidigit sequences in natural scenes.\nThe dataset for this project can be found at \nhttp://ufldl.stanford.edu/housenumbers/\n\nThis project has been updated from when I created it as the Machine \nLearning Capstone project for Udacity. I have added Batch normalization \nand a deeper network as well as creating 5 data points from each picture.\nThe write-up included is out of date. \n\nThe model gets 98 precent training accuracy, 94 precent validation accuracy, and 95 precent test accurcay after 100 epochs. The model included Batch Norm, Dropout, and Weight Decay for regularization. Using so much regularization and still having so much difference between the training and validation error means that the training distribuation is very different than the validation distribution. To solve this more data is most likely needed.\n\nThe following libraries are required:\n\n * Tensorflow 0.9.0\n * numpy 1.10.4\n * scipy 0.17.0\n * matplotlib 1.5.1\n * PIL 1.1.7\n * opencv 3.1.0\n\nThe program is run with `python Camera.py`\nTo download and process the data run the first option \"Process The Datasets\"\nThis will download, extract, and process the images. It will save the data \ninto the files \"train_dataset.npy\", \"train_labels.npy\", \"valid_dataset.npy\",\n\"valid_labels.npy\", test_dataset.npy\", and \"test_labels.py\".\nThe next step it to train the model using the second option \"Train the\n model\". This will train the model and save it into a file called \nmodel.ckpt. Analytics from the datasets and the training can be found \nusing the third option \"Display Analytics\". The fouth option \"Example Use\" \ndisplays some example images along with the labels and predictions.\nThe fifth option \"Use the model\" allows users to use the camera or to \nload an image from disk that will be feed to the network.\n\nThe trained model is included (along with the file \nthat includes the Analytics and a test folder to allow for seeing some \nimages in the example use) so users can run the Example use case or the \ncamera without training the model. Before training the model the data \nwill have to be fetched which should run automatically when running Camera.py\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fblester125%2Fmulti_digit_recognition","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fblester125%2Fmulti_digit_recognition","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fblester125%2Fmulti_digit_recognition/lists"}