{"id":31581513,"url":"https://github.com/rasyosef/text-embedding-models-training","last_synced_at":"2025-10-05T21:59:15.035Z","repository":{"id":278031397,"uuid":"934300118","full_name":"rasyosef/text-embedding-models-training","owner":"rasyosef","description":"Notebooks to train and evaluate Amharic Text Embedding Models based on BERT and RoBERTa for Passage Retrieval","archived":false,"fork":false,"pushed_at":"2025-02-17T16:06:01.000Z","size":73,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-09-14T09:02:19.500Z","etag":null,"topics":["bert","embedding-models","embeddings","huggingface","model-training","roberta","sentence-transformers","transformers"],"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/rasyosef.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":"2025-02-17T15:47:32.000Z","updated_at":"2025-03-26T14:18:43.000Z","dependencies_parsed_at":null,"dependency_job_id":"1a01df92-ee88-431c-9795-194e722cbe0b","html_url":"https://github.com/rasyosef/text-embedding-models-training","commit_stats":null,"previous_names":["rasyosef/amharic-text-embedding-models-training","rasyosef/text-embedding-models-training"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/rasyosef/text-embedding-models-training","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rasyosef%2Ftext-embedding-models-training","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rasyosef%2Ftext-embedding-models-training/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rasyosef%2Ftext-embedding-models-training/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rasyosef%2Ftext-embedding-models-training/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/rasyosef","download_url":"https://codeload.github.com/rasyosef/text-embedding-models-training/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rasyosef%2Ftext-embedding-models-training/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":278526223,"owners_count":26001327,"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":"2025-10-05T02:00:06.059Z","response_time":54,"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":["bert","embedding-models","embeddings","huggingface","model-training","roberta","sentence-transformers","transformers"],"created_at":"2025-10-05T21:59:10.846Z","updated_at":"2025-10-05T21:59:15.029Z","avatar_url":"https://github.com/rasyosef.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Training Amharic Text Embedding Models for Passage Retrieval\n\nThis repo contains code for training Amharic Text Embedding models based on 3 Amharic Encoder Base models \n- [RoBERTa Base Amharic](https://huggingface.co/rasyosef/roberta-base-amharic)\n- [RoBERTa Medium Amharic](https://huggingface.co/rasyosef/roberta-medium-amharic)\n- [BERT Medium Amharic](https://huggingface.co/rasyosef/bert-medium-amharic)\n\nWe also Evaluate the highest ranking multilingual embedding models form the [Massive Text Embedding Benchmark (MTEB) Leaderboard](https://huggingface.co/spaces/mteb/leaderboard) using the same test set as our embedding models.\n\n# Results\nOur largest embedding model, RoBERTa-Base-Amharic-Embed beats all of the multilingual embedding models on the MRR@10, NDCG@10 and Recall metrics while having 1/5th of their param count.\n\n|Model | Params | MRR@10 | NDCG@10 | Recall@10 | Recall@50 | Recall@100 |\n|------|--------|--------|---------|-----------|-----------|------------|\n|gte-modernbert-base | 149M | 0.019 | 0.022 | 0.030 | 0.054 | 0.065 |\n|gte-multilingual-base | 305M | 0.649 | 0.684 | 0.794 | 0.876 | 0.904 |\n|multilingual-e5-large-instruct | 560M | 0.713 | 0.747 | 0.853 | 0.924 | 0.946 |\n|snowflake-arctic-embed-l-v2.0 | 568M | 0.719 | 0.755 | 0.868 | 0.941 | 0.957 |\n|BERT-Medium-Amharic-Embed | 40M | 0.657 | 0.696 | 0.817 | 0.916 | 0.945 |\n|RoBERTa-Medium-Amharic-Embed | 42M | 0.707 | 0.744 | 0.861 | 0.941 | 0.963 |\n|RoBERTa-Base-Amharic-Embed | 110M | **0.755** |**0.790** | **0.897** | **0.957** | **0.971** |\n\n# Code\n\nThe training and eval code can be found in the `notebooks` folder.\n\n# Embedding Models\n\nYou can download and use our Amharic Text Embedding models from HuggingFace and they are fully compatable with the Sentence Transformers Library\n - RoBERTa-Base-Amharic-Embed: https://huggingface.co/rasyosef/roberta-amharic-text-embedding-base\n - RoBERTa-Medium-Amharic-Embed: https://huggingface.co/rasyosef/roberta-amharic-text-embedding-medium\n - BERT-Medium-Amharic-Embed: https://huggingface.co/rasyosef/bert-amharic-text-embedding-medium","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frasyosef%2Ftext-embedding-models-training","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frasyosef%2Ftext-embedding-models-training","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frasyosef%2Ftext-embedding-models-training/lists"}