Hey, if someone was wondering, if it's possible to use Jesse in Colab, like I did, it's possible, but rather unpleasant, because every time you start working on it it takes 8 minutes to install Jesse + several minutes to download data.
Even bigger issue is changing strategies, the most effective way would be to write them locally an upload them using sidebar on the left side of the editor or you can run code inside notebook to write files containing strategy rules.
The upside is that it's super easy to install and setup (so if you already have a ML project in Colab you want to backtest you can do this):
!sh -c "$(curl -fsSL https://raw.githubusercontent.com/jesse-ai/stack-installer/master/ubuntu-18.04.sh)"
!jesse make-project projekt
!service postgresql start
!sudo -u postgres psql -c "CREATE DATABASE jesse_db;"
!sudo -u postgres psql -c "CREATE USER jesse_user WITH PASSWORD 'password';"
!sudo -u postgres psql -c "GRANT ALL PRIVILEGES ON DATABASE jesse_db to jesse_user;"
!jesse import-candles Binance BTCUSDT 2019-01-01
!jesse backtest 2019-09-01 2020-05-05
The code above installs Jesse and dependencies, creates and runs db, downloads data and runs an example strategy backtest. Whats left to do is to implement trading strategy, change routes etc. and replace respective files in Colab.
I think Im just gonna install a virtual machine with Linux and run Jesse locally though.