Manual Installation on Linux

requirements: python (and conda)

1. Configure Environment Variable

Set the OpenAI API key:

export OPENAI_API_KEY=enter_your_chatgpt_api_key_here

2. Install and Run Python Environment

Execute the following script to install and run the Python environment:

./install.sh

3. Install and Run Ollama on Linux

Follow the installation instructions for Ollama from https://ollama.com/.

To install a model, follow the instructions provided on the website.

If necessary, update the environment variable for Ollama:

export OLLAMA_API_BASE_URL=http://localhost:11434

4. Launch the Streamlit App

Activate the evocell environment and launch the app:

conda activate evocell
cd evocell/app
streamlit run main.py

Manual Installation on Windows

0. Install python, conda

1. Configure Environment Variable

To set the OpenAI API key, open a Command Prompt and run:

set OPENAI_API_KEY=enter_your_chatgpt_api_key_here

2. Install and Run Python Environment

Run the install.bat file to install dependencies:

install.bat

3. Install and Run Ollama on Windows

Follow the instructions for installing Ollama from https://ollama.com/.

After installation, update the environment variable if necessary:

set OLLAMA_API_BASE_URL=http://localhost:11434

4. Launch the Streamlit App

Activate the Python environment and run the Streamlit app:

conda activate evocell
cd evocell/app
streamlit run main.py

Manual Installation on macOS

0. Install python, conda

1. Configure Environment Variable

Set the OpenAI API key in the terminal:

export OPENAI_API_KEY=enter_your_chatgpt_api_key_here

2. Install and Run Python Environment

Run the install.sh script to install dependencies:

./install.sh

3. Install and Run Ollama on macOS

Follow the installation instructions for Ollama from https://ollama.com/.

After installation, if necessary, update the environment variable:

export OLLAMA_API_BASE_URL=http://localhost:11434

4. Launch the Streamlit App

Activate the Python environment and run the app:

conda activate evocell
cd evocell/app
streamlit run main.py

Docker Installation

0. Install Docker if not already done

1. Update API Key

In the docker-compose.yml file:

• Change the OPENAI_API_KEY if you are using OpenAI ChatGPT.

2. Access Local Hard Drive

For accessing your local hard drive for data, modify the volumes section:

• From:

  volumes: 
    device: /home/mamat/accplatform/article1-sanofi/data

• To:

  volumes:
    device: [location on your hard drive]

3. Build and Run Docker

docker compose build
docker compose up