{"id":30771913,"url":"https://github.com/mr-faizan-asim/deepcore-ai","last_synced_at":"2026-05-19T14:36:58.281Z","repository":{"id":210537650,"uuid":"726541506","full_name":"Mr-Faizan-Asim/DeepCore-AI","owner":"Mr-Faizan-Asim","description":"A collection of machine learning \u0026 deep learning projects and notebooks — regression, classification, explainable AI, and NLP fine‑tuning experiments.","archived":false,"fork":false,"pushed_at":"2025-08-26T23:26:47.000Z","size":7133,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-08-27T01:38:12.695Z","etag":null,"topics":["ai","collab","deep-neural-networks","gnn","kaggle","machine-learning","ml","research","rnn","xai"],"latest_commit_sha":null,"homepage":"https://www.kaggle.com/muhammadfaizanasim","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/Mr-Faizan-Asim.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}},"created_at":"2023-12-02T17:23:44.000Z","updated_at":"2025-08-26T23:28:05.000Z","dependencies_parsed_at":"2025-02-23T19:27:55.507Z","dependency_job_id":"93bf21f1-f79c-4710-967d-9f736ec7e1b5","html_url":"https://github.com/Mr-Faizan-Asim/DeepCore-AI","commit_stats":null,"previous_names":["mr-faizan-asim/ml-","mr-faizan-asim/ml-dl-solutions","mr-faizan-asim/deepcore-ai"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Mr-Faizan-Asim/DeepCore-AI","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mr-Faizan-Asim%2FDeepCore-AI","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mr-Faizan-Asim%2FDeepCore-AI/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mr-Faizan-Asim%2FDeepCore-AI/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mr-Faizan-Asim%2FDeepCore-AI/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Mr-Faizan-Asim","download_url":"https://codeload.github.com/Mr-Faizan-Asim/DeepCore-AI/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mr-Faizan-Asim%2FDeepCore-AI/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273690697,"owners_count":25150643,"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":"2025-09-04T02:00:08.968Z","response_time":61,"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":["ai","collab","deep-neural-networks","gnn","kaggle","machine-learning","ml","research","rnn","xai"],"created_at":"2025-09-05T00:06:54.480Z","updated_at":"2026-05-19T14:36:58.235Z","avatar_url":"https://github.com/Mr-Faizan-Asim.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DeepCore-AI\n\n\u003e A collection of machine learning \u0026 deep learning projects and notebooks — regression, classification, explainable AI, and NLP fine‑tuning experiments.\n\n**Author:** Muhammad Faizan Asim ([mr.faizan.asim@gmail.com](mailto:mr.faizan.asim@gmail.com))\n**Kaggle:** [Muhammad Faizan Asim](https://www.kaggle.com/muhammadfaizanasim)\n\n---\n\n## Overview\n\nThis repository (DeepCore-AI) contains a set of data science and deep learning projects collected as reproducible Jupyter notebooks and small helper scripts. Projects range from classical regression and PCA analyses to explainable AI (XAI) notebooks and a Hugging Face BERT fine‑tuning workflow for text classification.\n\nUse this README to quickly understand each component, how to run the notebooks locally/Colab/Kaggle, and how to reproduce training and evaluation steps.\n\n---\n\n## Repo structure (high level)\n\n```\nDeepCore-AI/\n├─ Linear Regression Project of Price Predication/   # linear regression project(s)\n├─ MultiVariablePredication/                         # multivariable regression\n├─ DiaXAi_MODEL_+_XAI.ipynb                           # model + explainability notebook\n├─ Email.py                                          # small utility / email helper\n├─ Principal component analysis (PCA).py              # PCA script\n├─ data-preprocessing-feature-engineering.ipynb      # preprocessing \u0026 feature engineering\n├─ dna-change-prediction-using-decisiontree*.ipynb    # DNA change / decision tree experiments\n├─ dna-classification-and-disease-risk-prediction.ipynb\n├─ fine-tuning-bert-with-hugging-face-for-text-cl...ipynb  # BERT fine-tune notebook (text classification)\n├─ tuberculosis-classification-from-chest-x-ray.ipynb\n└─ README.md\n```\n\n\u003e Filenames in the repo are descriptive — open the notebooks to see dataset links, hyperparameters and results.\n\n---\n\n## Quickstart — recommended workflow\n\n1. Clone this repository:\n\n```bash\ngit clone https://github.com/Mr-Faizan-Asim/DeepCore-AI.git\ncd DeepCore-AI\n```\n\n2. Create an isolated Python environment (recommended):\n\n```bash\n# using venv\npython -m venv .venv\nsource .venv/bin/activate  # Linux / macOS\n.\\.venv\\Scripts\\activate   # Windows (PowerShell: .\\.venv\\Scripts\\Activate.ps1)\n\n# or using conda\nconda create -n deepai python=3.10 -y\nconda activate deepai\n```\n\n3. Install dependencies. Create a `requirements.txt` file (if not present) with key packages below and then:\n\n```bash\npip install -r requirements.txt\n```\n\nSuggested minimum `requirements.txt`:\n\n```\nnumpy\npandas\nscikit-learn\nmatplotlib\nseaborn\njupyterlab\nnotebook\nxgboost\nlightgbm\ntensorflow\ntorch\ntorchvision\ntransformers\ndatasets\naccelerate\nevaluate\nopencv-python\nlime\nshap\nscikit-image\n```\n\n\u003e Note: GPU-accelerated packages (TensorFlow/PyTorch with CUDA) may require additional platform-specific installation.\n\n---\n\n## Running notebooks (choices)\n\n### Option A — Run locally\n\n1. Activate the environment.\n2. Start JupyterLab:\n\n```bash\njupyter lab\n```\n\n3. Open and run the desired notebook (e.g. `fine-tuning-bert-with-hugging-face-for-text-cl...ipynb`).\n\n### Option B — Run on Google Colab\n\n1. Upload the notebook to Colab (Open → Upload notebook) or open directly from GitHub using Colab’s GitHub integration.\n2. If using large datasets, prefer mounting Google Drive or loading from Kaggle.\n\n### Option C — Run on Kaggle Notebooks\n\nMany notebooks were authored for Kaggle — you can open them directly on Kaggle and run with Kaggle datasets and free GPU/TPU.\n\n---\n\n## Notebook / file descriptions\n\nA short description for each key file:\n\n* `Linear Regression Project of Price Predication/` — Project folder for housing a classical price prediction pipeline using linear regression and model evaluation.\n* `MultiVariablePredication/` — Multivariate regression experiments using multiple features and feature selection.\n* `data-preprocessing-feature-engineering.ipynb` — Utilities and patterns to clean raw datasets, impute missing values, encode categorical variables and create features.\n* `DiaXAi_MODEL_+_XAI.ipynb` — Model training combined with explainability (SHAP, LIME) visualizations demonstrating how predictions are influenced by features.\n* `Principal component analysis (PCA).py` — Script showing dimensionality reduction examples and how to visualize principal components for high-dimensional data.\n* `dna-change-prediction-using-decisiontree...ipynb` and `dna-classification-and-disease-risk-prediction.ipynb` — Bioinformatics classification experiments (DNA change prediction, disease risk classifiers) — include data preprocessing, model training and evaluation.\n* `fine-tuning-bert-with-hugging-face-for-text-cl...ipynb` — Hugging Face Transformers fine-tuning notebook (BERT) for text classification. Includes tokenization, dataset preparation, Trainer/TrainingArguments or `accelerate` setup, and evaluation metrics.\n* `tuberculosis-classification-from-chest-x-ray.ipynb` — CNN-based chest X‑ray image classification for TB detection (data loading, augmentation, training loop and metrics).\n* `Email.py` — Small helper: likely used to send automated emails (double-check credentials and never commit secrets). If it uses SMTP credentials, move them to environment variables.\n\n---\n\n## Reproducibility tips and large files\n\n* **Datasets \u0026 large model files:** Do not commit large datasets or model checkpoints directly to Git. Use one of:\n\n  * Provide direct download links in each notebook (Kaggle dataset links, Google Drive public links).\n  * Use Git LFS for large binary files.\n  * Document where to place files locally (e.g. `data/` folder) and how to name them.\n\n* **Secrets \u0026 credentials:** Never push API keys, tokens or email credentials. Use environment variables or a `.env` file added to `.gitignore`.\n\n---\n\n## BERT fine-tuning (short actionable guide)\n\nIf you want to reproduce the BERT fine-tuning notebook quickly:\n\n1. Ensure these packages are installed: `transformers`, `datasets`, `accelerate`, `evaluate`.\n2. Prepare a CSV/JSON dataset with `text` and `label` columns.\n3. Minimal example commands inside a notebook:\n\n```python\nfrom datasets import load_dataset\nfrom transformers import AutoTokenizer, AutoModelForSequenceClassification, TrainingArguments, Trainer\n\nmodel_name = 'bert-base-uncased'\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=NUM_LABELS)\n\ndataset = load_dataset('csv', data_files={'train':'train.csv','validation':'val.csv'})\n\ndef preprocess(ex):\n    return tokenizer(ex['text'], truncation=True, padding='max_length')\ndataset = dataset.map(preprocess, batched=True)\n\nargs = TrainingArguments(\n    output_dir='./outputs',\n    evaluation_strategy='epoch',\n    save_strategy='epoch',\n    per_device_train_batch_size=16,\n    per_device_eval_batch_size=32,\n    num_train_epochs=3,\n    logging_steps=50,\n)\n\ntrainer = Trainer(\n    model=model,\n    args=args,\n    train_dataset=dataset['train'],\n    eval_dataset=dataset['validation'],\n    tokenizer=tokenizer,\n)\n\ntrainer.train()\n```\n\n\u003e For larger scale runs use `accelerate` and a GPU runtime.\n\n---\n\n## Good to know / best practices\n\n* Add a `requirements.txt` and an optional `environment.yml` (conda) to help others reproduce your environment.\n* Add a `LICENSE` file (MIT is a common permissive choice).\n* Add a `CONTRIBUTING.md` if you want external contributors.\n* Add small sample datasets (or scripts to download them) so users can run notebooks quickly.\n\n---\n\n## Contributing\n\n1. Fork the repo\n2. Create a feature branch: `git checkout -b feature/your-feature`\n3. Commit changes and open a Pull Request\n\nPlease include reproducible steps and small datasets or pointers to data when submitting PRs that touch notebooks.\n\n---\n\n## License\n\nThis repository does not include an explicit license file yet. If you are happy to make it open source, consider adding `LICENSE` with the MIT license.\n\n---\n\n## Contact\n\nMuhammad Faizan Asim — [mr.faizan.asim@gmail.com](mailto:mr.faizan.asim@gmail.com)\nGitHub: [Mr-Faizan-Asim](https://github.com/Mr-Faizan-Asim)\nKaggle: [Muhammad Faizan Asim](https://www.kaggle.com/muhammadfaizanasim)\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmr-faizan-asim%2Fdeepcore-ai","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmr-faizan-asim%2Fdeepcore-ai","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmr-faizan-asim%2Fdeepcore-ai/lists"}