Skip to content

Quick Start

Get Agentomics-ML running in under 5 minutes using pre-built Docker images.

Prerequisites

Steps

1. Clone the Repository

git clone https://github.com/BioGeMT/Agentomics-ML.git
cd Agentomics-ML

2. Create a .env File and Set a Key

Docker mode requires a .env file in the repo root.

cp .env.example .env
# Edit .env and set at least one provider key:
# OPENROUTER_API_KEY, OPENAI_API_KEY, or ANTHROPIC_API_KEY

3. Run the Agent

./run.sh

4. Follow the Interactive Prompts

The agent will prompt you to:

  1. Select a model - Choose from available LLMs
  2. Select a dataset - Use your own or download examples
  3. Configure iterations - How many optimization cycles to run

The validation metric defaults to AUROC for classification and MAE for regression. To choose one explicitly, pass --val-metric; see ./run.sh --list-metrics.

Using Your Own Dataset

Place your data in datasets/<your_dataset_name>/:

datasets/my_dataset/
├── train/
│   ├── input/          # Required: model input files
│   └── labels.csv      # Required: id,label
├── validation/         # Optional
│   ├── input/
│   └── labels.csv
└── dataset_description.md

(Optional) Put hidden test data under the matching test_datasets/ folder:

test_datasets/my_dataset/
└── test/
    ├── input/
    └── labels.csv

See Preparing Datasets for details.

Example Datasets

Download example dataset to try:

./scripts/download_example_dataset.sh

List other available examples with:

./scripts/download_example_dataset.sh --list

What Happens Next

The agent will:

  1. Prepare your dataset
  2. Run iterative ML development cycles
  3. Save the best model to outputs/<agent_id>/

Results include trained models, inference scripts, markdown reports in outputs/<agent_id>/reports/markdown/, and PDF reports in outputs/<agent_id>/reports/pdf/.

Next Steps