{"id":25180353,"url":"https://github.com/nihalrt/ltsm-cloud-predict","last_synced_at":"2026-04-05T21:32:19.689Z","repository":{"id":276020931,"uuid":"927949256","full_name":"Nihalrt/ltsm-cloud-predict","owner":"Nihalrt","description":"AI-powered time-series forecasting system that predicts CPU, Memory, Network, and Disk utilization in cloud environments using LTSM neural networks.","archived":false,"fork":false,"pushed_at":"2025-03-03T20:14:02.000Z","size":312,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-04T04:46:39.080Z","etag":null,"topics":["keras","ltsm","matplotlib","pandas","python","tensorflow"],"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/Nihalrt.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":"2025-02-05T20:05:19.000Z","updated_at":"2025-03-03T20:14:20.000Z","dependencies_parsed_at":null,"dependency_job_id":"47b7be33-066f-40f2-b29f-d54be3d5ed1d","html_url":"https://github.com/Nihalrt/ltsm-cloud-predict","commit_stats":null,"previous_names":["nihalrt/ltsm-cloud-predict"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Nihalrt%2Fltsm-cloud-predict","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Nihalrt%2Fltsm-cloud-predict/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Nihalrt%2Fltsm-cloud-predict/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Nihalrt%2Fltsm-cloud-predict/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Nihalrt","download_url":"https://codeload.github.com/Nihalrt/ltsm-cloud-predict/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247123100,"owners_count":20887260,"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":["keras","ltsm","matplotlib","pandas","python","tensorflow"],"created_at":"2025-02-09T16:17:44.245Z","updated_at":"2025-12-30T23:08:57.819Z","avatar_url":"https://github.com/Nihalrt.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Cloud Utilization Prediction using LSTMs\n\n## Overview\nThis project predicts **CPU, Memory, Network, and Disk utilization** in a cloud environment using **LTSM neural networks**. By leveraging AI-powered time-series forecasting, this system provides insights into cloud resource usage trends, helping cloud providers optimize performance, reduce costs, and prevent downtime.\n\n## Features\n- **Multi-Resource Prediction**: Forecasts CPU, Memory, Network, and Disk usage.\n- **Anomaly Detection**: Identifies unexpected spikes in resource utilization.\n- **Hyperparameter Tuning**: Optimized LSTM model for improved accuracy.\n- **Real-Time API Deployment**: Uses FastAPI to serve predictions.\n- **Enhanced Visualizations**: Provides clear insights with time-series graphs.\n- **Model Saving \u0026 Deployment**: Enables real-world integration and inference.\n\n## Dataset\nThe dataset consists of simulated cloud workload metrics, including:\n- **CPU Usage** (Percentage)\n- **Memory Usage** (Percentage)\n- **Network Utilization** (MB/s)\n- **Disk Utilization** (MB/s)\n\nIn real-world scenarios, data can be sourced from **AWS CloudWatch, Google Cloud Metrics, or Azure Monitor**.\n\n## Installation \u0026 Setup\n### 1. Clone the Repository\n```sh\ngit clone https://github.com/your-username/Cloud-Resource-Predictor.git\ncd Cloud-Resource-Predictor\n```\n\n### 2. Install Dependencies\n```sh\npip install -r requirements.txt\n```\n\n### 3. Run Jupyter Notebook\n```sh\njupyter notebook\n```\nOpen `Cloud_Resource_Utilization_Prediction_With_Graphs.ipynb` and execute the cells.\n\n## Model Architecture\nThe LSTM model consists of:\n- **LSTM Layers**: Captures time-series dependencies.\n- **Dropout Layers**: Prevents overfitting.\n- **Dense Layers**: Outputs multi-resource predictions.\n\n## Usage\n### Training the Model\nRun the training script in the notebook:\n```python\nhistory = model.fit(X_train, y_train, epochs=50, batch_size=32, validation_data=(X_test, y_test))\n```\n\n### Evaluating the Model\n```python\npredictions = model.predict(X_test)\nmae = mean_absolute_error(y_test, predictions)\nrmse = math.sqrt(mean_squared_error(y_test, predictions))\n```\n\n### Saving and Loading the Model\n```python\nmodel.save('cloud_utilization_prediction.h5')\n```\nTo load the model for inference:\n```python\nfrom tensorflow.keras.models import load_model\nmodel = load_model('cloud_utilization_prediction.h5')\n```\n\n## Visualization\nTo compare actual vs. predicted CPU usage:\n```python\nplt.plot(y_test[:, 0], label='Actual CPU Usage', color='blue')\nplt.plot(predictions[:, 0], label='Predicted CPU Usage', linestyle='dashed', color='orange')\nplt.xlabel('Time')\nplt.ylabel('CPU Usage')\nplt.title('Actual vs. Predicted CPU Utilization')\nplt.legend()\nplt.show()\n```\n\n## Future Enhancements\n- **Integrate with Real Cloud Data**: AWS CloudWatch, Google Cloud Metrics.\n- **Deploy Model as an API**: Provide real-time cloud usage forecasts.\n- **Extend to Additional Metrics**: Include latency, IOPS, and other factors.\n- **Improve Anomaly Detection**: Implement advanced outlier detection methods.\n\n## License\nThis project is open-source under the **MIT License**.\n\n## Contributors\n- **Your Name** – [GitHub Profile](https://github.com/your-username)\n\n## Acknowledgments\n- TensorFlow/Keras for deep learning support.\n- Open-source datasets for simulation.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnihalrt%2Fltsm-cloud-predict","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnihalrt%2Fltsm-cloud-predict","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnihalrt%2Fltsm-cloud-predict/lists"}