3_body_problem_bot

How to deploy this bot using Docker:

1. Clone this repository:

   git clone https://github.com/robolamp/3_body_problem_bot.git
   

2. Enter the directry with cloned repo:

   cd 3_body_problem_bot
   

3. Edit Dockerflie: replace [TOKEN] and [BOT_NAME] with your bot token and your channel name

   RUN (crontab -l ; echo "0 */12 * * * python3 /home/3_body_problem_bot/generate_3_body_simulation.py -V -T [TOKEN] -N [BOT_NAME] --fps 15 --min-score 24 --duration 150 >> /var/log/cron.log") | crontab
   

4. Build docker image using command:

   docker build -t [your image name] -f ./Dockerfile .
   

5. Using built image, run the container in background:

   docker run -d [your image name]:latest
   

6. Check that the container is running and find id of your container:

   docker container ls
   

7. Check that crontable into the container is set up as you expected:

   docker exec -ti [your container id] bash -c "cat /var/log/cron.log"
   

8. Check log when it'll be time to tun the script

   docker exec -ti [your container id] bash -c "crontab -l"
   

How to run this script locally:

1. Install/check general requirements:

   ffmpeg python3
   

2. Install/check python3 requirements with pip3:

   pip3 install numpy scipy matplotlib python-telegram-bot
   

3. Now you can run this script!

   With the following command you can run the simulation with default params:

   ./generate_3_body_simulation.py
   

   Without providing arguments -T and -N it will run without Telegram bot functionality.

   To modify the simulation, you can use the following arguments:

   --dt DT for simulation step;

   --fps FPS for frames per second in the video;

   --n-bodies N_BODIES number of bodies (tested with up to 60);

   --duration DURATION simulation duration;

   --video-duration video duration (if simulation duration is bigger, the visualization will be accelerated);

   --min-score MIN_SCORE Minimal "interest-ness" score;

   --max-score MAX_SCORE Maximal "interest" score;

   With bigger difference between MIN_SCORE and MAX_SCORE, the script will run fewer simulations before achieving appropriate one.

   Simulations with high "interest-ness" score are looking too chaotic, and simulations with "interest-ness" score are boring (almost nothing happens).

   --max-field-size MAX_FIELD_SIZE Maximal field size.