{"id":26375512,"url":"https://github.com/oaarnikoivu/emotion-classifier","last_synced_at":"2025-03-17T02:17:38.319Z","repository":{"id":37661846,"uuid":"231822149","full_name":"oaarnikoivu/emotion-classifier","owner":"oaarnikoivu","description":"An attention-based BiLSTM for emotion classification. ","archived":false,"fork":false,"pushed_at":"2022-12-08T09:53:24.000Z","size":20719,"stargazers_count":5,"open_issues_count":7,"forks_count":2,"subscribers_count":2,"default_branch":"master","last_synced_at":"2023-03-05T14:13:39.591Z","etag":null,"topics":["bert","deep-learning","embeddings","emotion","emotion-detection","flask","multi-label-classification","python","pytorch","react","text-classification","transformers","typescript"],"latest_commit_sha":null,"homepage":"","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/oaarnikoivu.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":"2020-01-04T20:11:15.000Z","updated_at":"2023-03-05T14:13:39.591Z","dependencies_parsed_at":"2023-01-25T09:30:31.822Z","dependency_job_id":null,"html_url":"https://github.com/oaarnikoivu/emotion-classifier","commit_stats":null,"previous_names":[],"tags_count":null,"template":null,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oaarnikoivu%2Femotion-classifier","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oaarnikoivu%2Femotion-classifier/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oaarnikoivu%2Femotion-classifier/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oaarnikoivu%2Femotion-classifier/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/oaarnikoivu","download_url":"https://codeload.github.com/oaarnikoivu/emotion-classifier/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243960623,"owners_count":20375106,"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":["bert","deep-learning","embeddings","emotion","emotion-detection","flask","multi-label-classification","python","pytorch","react","text-classification","transformers","typescript"],"created_at":"2025-03-17T02:17:37.731Z","updated_at":"2025-03-17T02:17:38.309Z","avatar_url":"https://github.com/oaarnikoivu.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# An attention-based BiLSTM for emotion classification. \n\nThis model was trained to detect multiple emotions from text. The model achieves comparable results to the state-of-the-art by using an Attention BiLSTM with contextualized embeddings produced by a frozen BERT transformer model.\n\n## Overview\n\n- [/client](https://github.com/oaarnikoivu/ainoa/tree/master/client) contains all client-side code related to the UI of the application.\n- [/server](https://github.com/oaarnikoivu/ainoa/tree/master/server) contains the server-side code related to the application.\n- [/notebooks](https://github.com/oaarnikoivu/ainoa/tree/master/notebooks) contains all the iPython notebooks to carry out the methods \u0026 experimental setup.\n\n## Technologies\n\nThe model architecture is constructed using PyTorch. The [HuggingFace Transformers](https://github.com/huggingface/transformers) library is made use of in order to apply and retrive the embeddings from a pre-trained BERT model. [Scikit Learn](https://scikit-learn.org/) is also made use of to assess model performance using the Jaccard index, and micro-averaged and macro-averaged F1 scores, as well as for the implemenation of the baseline machine learning algorithms. The models were trained with Google Colaboratory notebooks using an NVIDIA Tesla K80 GPU.\n\n## Dataset\n\nThe dataset used is the [SemEval Task 1: Affect in Tweets](https://www.aclweb.org/anthology/S18-1001/) emotion classification dataset where given a tweet, the task is to classify the text as having no emotion or as one, or more, emotions for eight of the [Plutchik](https://www.6seconds.org/2020/08/11/plutchik-wheel-emotions/) categories plus optimism, pessimism, and love. The dataset consists of 6838 training examples, 886 validation examples and 3259 testing examples.\n\n[Report](https://github.com/oaarnikoivu/ainoa/blob/master/1502639%20AARNIKOIVU%20Oliver%20-%20Thesis.pdf)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foaarnikoivu%2Femotion-classifier","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Foaarnikoivu%2Femotion-classifier","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foaarnikoivu%2Femotion-classifier/lists"}