{"id":31774088,"url":"https://github.com/levmn/iot-deep-learning-cp","last_synced_at":"2026-06-23T04:32:17.309Z","repository":{"id":318271292,"uuid":"1070561626","full_name":"levmn/iot-deep-learning-cp","owner":"levmn","description":null,"archived":false,"fork":false,"pushed_at":"2025-10-06T14:47:44.000Z","size":33,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-14T07:23:41.078Z","etag":null,"topics":[],"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/levmn.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":"2025-10-06T05:53:09.000Z","updated_at":"2025-10-06T14:47:48.000Z","dependencies_parsed_at":"2025-10-06T08:53:14.007Z","dependency_job_id":null,"html_url":"https://github.com/levmn/iot-deep-learning-cp","commit_stats":null,"previous_names":["levmn/iot-deep-learning-cp"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/levmn/iot-deep-learning-cp","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/levmn%2Fiot-deep-learning-cp","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/levmn%2Fiot-deep-learning-cp/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/levmn%2Fiot-deep-learning-cp/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/levmn%2Fiot-deep-learning-cp/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/levmn","download_url":"https://codeload.github.com/levmn/iot-deep-learning-cp/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/levmn%2Fiot-deep-learning-cp/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34675970,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-23T02:00:07.161Z","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":[],"created_at":"2025-10-10T04:47:53.380Z","updated_at":"2026-06-23T04:32:17.285Z","avatar_url":"https://github.com/levmn.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"## Treinamento de Redes Neurais (Keras) com Dados Tabulares\n\nEste repositório contém as implementações dos exercícios de Classificação Multiclasse (Wine — UCI) e Regressão (California Housing — scikit-learn), com comparação a modelos do scikit-learn.\n\n## Integrantes\n- [RM558948] [Allan Brito Moreira](https://github.com/Allanbm100)\n- [RM558868] [Caio Liang](https://github.com/caioliang)\n- [RM98276] [Levi Magni](https://github.com/levmn)\n\n### Arquivos\n- `iot-nns-training.ipynb` - notebook pronto para executar no Google Colab;\n- `iot-nns-training.py` - script python equivalente para execução local.\n\n### Como executar no Google Colab (recomendado)\n1. Acesse `https://colab.research.google.com/`;\n2. Abra o arquivo `iot-nns-training.ipynb`;\n3. Vá em Runtime/Executar tudo (ou Runtime/Run all) e aguarde o término.\n\nObservação: o Colab já possui as dependências (TensorFlow/Keras, scikit-learn, NumPy, Pandas) pré-instaladas.\n\n### Como executar localmente via `.py`\nPré-requisitos:\n- python 3.10+ e `pip` instalados.\n\nInstale as dependências:\n\n```bash\npython -m pip install --upgrade pip\npip install numpy pandas scikit-learn tensorflow keras\n```\n\nExecute o script:\n\n```bash\npython iot-nns-training.py\n```\n\nNotas:\n- Em Apple Silicon, se preferir, use `tensorflow-macos`. Caso tenha dificuldades com TensorFlow local, rode no Colab;\n- O script baixa os datasets automaticamente e imprime as métricas ao final de cada experimento.\n\n### Configurações dos modelos\n- **Classificação (Wine/UCI)**:\n  - Keras: 2 camadas ocultas (32 neurônios, reLU) + saída Softmax (3 classes)\n  - Perda: `categorical_crossentropy`, Otimizador: `Adam`\n  - Comparação: `RandomForestClassifier`\n- **Regressão (California Housing)**:\n  - Keras: 3 camadas ocultas (64, 32, 16 neurônios, reLU) + saída Linear (1 neurônio)\n  - Perda: `mse`, Otimizador: `Adam`\n  - Comparação: `LinearRegression`\n\n### Resultados obtidos\nOs resultados abaixo foram obtidos com `random_state=42` e normalização dos atributos.\n\n#### Exercício 1 — Classificação (Wine)\n- Rede neural (Keras):\n  - Acurácia de teste = `1.0000`\n- RandomForestClassifier:\n  - Acurácia de teste = `1.0000`\n- Observação: ambos os modelos atingiram 100% de acurácia neste conjunto de treino/teste.\n\n#### Exercício 2 — Regressão (California Housing)\n- Rede Neural (Keras): `MSE 0.2629` | `RMSE 0.5127` | `MAE 0.3345` | `R² 0.7994`\n- LinearRegression: `MSE 0.5559` | `RMSE 0.7456` | `MAE 0.5955` | `R² 0.5758`\n- Observação: a rede neural superou a regressão linear em todas as métricas.\n\n### Datasets utilizados\n- [Wine (UCI)](https://archive.ics.uci.edu/dataset/109/wine)\n- [California Housing (scikit-learn)](https://scikit-learn.org/stable/datasets/real_world.html#california-housing-dataset)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flevmn%2Fiot-deep-learning-cp","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flevmn%2Fiot-deep-learning-cp","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flevmn%2Fiot-deep-learning-cp/lists"}