{"id":14958879,"url":"https://github.com/atrostan/4107_final_project","last_synced_at":"2025-04-02T08:31:08.657Z","repository":{"id":106668928,"uuid":"162043021","full_name":"atrostan/4107_Final_Project","owner":"atrostan","description":"CNN model for classifying Psytrance Subgenres","archived":false,"fork":false,"pushed_at":"2018-12-17T02:19:48.000Z","size":22333,"stargazers_count":4,"open_issues_count":0,"forks_count":1,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-03-16T16:51:28.237Z","etag":null,"topics":["cnn-keras","genre-classification","keras-tensorflow","mir","tensorflow-experiments"],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","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/atrostan.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}},"created_at":"2018-12-16T21:34:30.000Z","updated_at":"2024-05-22T05:53:44.000Z","dependencies_parsed_at":null,"dependency_job_id":"fe259714-8451-4dc3-a3df-83a477102589","html_url":"https://github.com/atrostan/4107_Final_Project","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/atrostan%2F4107_Final_Project","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atrostan%2F4107_Final_Project/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atrostan%2F4107_Final_Project/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/atrostan%2F4107_Final_Project/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/atrostan","download_url":"https://codeload.github.com/atrostan/4107_Final_Project/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246781993,"owners_count":20832950,"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":["cnn-keras","genre-classification","keras-tensorflow","mir","tensorflow-experiments"],"created_at":"2024-09-24T13:18:26.513Z","updated_at":"2025-04-02T08:31:03.633Z","avatar_url":"https://github.com/atrostan.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 4107_Final_Project\n\nThanks to:\n\n* https://gitlab.crowdai.org/gg12/WWWMusicalGenreRecognitionChallenge?_ga=2.123714251.1594486358.1545011930-1055586743.1544557330\n* https://hackernoon.com/finding-the-genre-of-a-song-with-deep-learning-da8f59a61194\n* https://github.com/Hvass-Labs/TensorFlow-Tutorials/blob/master/18_TFRecords_Dataset_API.ipynb\n* https://ricardodeazambuja.com/deep_learning/2018/05/04/tuning_tensorflow_docker/\n* https://medium.com/@moritzkrger/speeding-up-keras-with-tfrecord-datasets-5464f9836c36\n* https://github.com/markjay4k/Fashion-MNIST-with-Keras/blob/master/pt3%20-%20FMINST%20Embeddings.ipynb\n* https://github.com/keras-team/keras/blob/master/examples/conv_filter_visualization.py\n\n# Requirements:\n\nSeen in [`requirements.txt`](./py/requirements.txt)\n\n|Library|Version|\n|-------|-------|\n|pandas|0.23.4|\n|matplotlib|3.0.2|\n|numpy|1.15.4|\n|tensorflow|1.12.0|\n|scipy|1.1.0|\n|librosa|0.6.2|\n|ipython|7.2.0|\n|keras|2.2.4|\n|scikit_learn|0.20.1|\n\n# Download Tracks\n\nEither download your own music from [http://www.ektoplazm.com/](http://www.ektoplazm.com/), \nor \ndownload music tracks from [`data_urls.txt`](./data_urls.txt)\n\n# Convert Tracks to `TFRecords`\n\nDetailed in [`read_mp3s.ipynb`](./read_mp3s.ipynb)\n\n# Train model \n\nDetailed in [`train_model.ipynb`](./train_model.ipynb).  \nYou can skip this step, and just extract `best_model.tar.gz`: a Keras model\nI've trained.   \nLoad this model using [load_model(best_psy_cnn.h5)](https://stackoverflow.com/a/43263973/8822734)\n\n# Test model \n\nDetailed in [`test_model.ipynb`](./test_model.ipynb).  \nYou can skip this step, and just extract `best_model.tar.gz`: a Keras model\nI've trained.   \n\n# Predict using the model\n\nDetailed in [`predict_model.ipynb`](./predict_model.ipynb).   \nThe model was trained on 5 genres:\n* [Downtempo](http://www.ektoplazm.com/style/downtempo)\n* [Dark](http://www.ektoplazm.com/style/darkpsy)\n* [Techno](http://www.ektoplazm.com/style/techno)\n* [Goa](http://www.ektoplazm.com/style/goa)\n* [Fullon](http://www.ektoplazm.com/style/full-on)\n\nDownload your own psytrance tracks (that can be characterized as one of the\nabove genres) and see how the model does!\n\nTo do this, make sure to compute the [mel-spectrogram](https://librosa.github.io/librosa/generated/librosa.feature.melspectrogram.html)\nof your track, and feed the model with slices of your spectrogram.\n\n# Visualize Convolutional Filters\n\nDetailed in [`visualize_filters.ipynb`](./visualize_filters.ipynb) and \nexamples in [`./tex/diagrams/stitched_filters`](./tex/diagrams/)\n\n# Report\n\nThe report outlining the development of this project is [here](./tex/report.pdf)\n\nProduction of graphs for the report are outlined in [`plot_results.ipynb`](./plot_results.ipynb)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fatrostan%2F4107_final_project","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fatrostan%2F4107_final_project","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fatrostan%2F4107_final_project/lists"}