tiny-team-example: A Python repository from dadinn

Tiny Team

$ uvicorn app.main:app --host 0.0.0.0 --port 8080 --reload
$ curl -X POST "http://localhost:7777/item" -H "Content-Type: application/json" -d '"Hello, Redis!"'
# {"status": "success"}
$ curl "http://localhost:7777"
# {"status": "Hello, Redis!"}

Install and Run

Make sure to install Docker before installing Docker Compose. Follow the instructions at https://docs.docker.com/engine/install/ubuntu/ then come back and continue below.

$ curl -SL https://github.com/docker/compose/releases/download/v2.20.2/docker-compose-linux-x86_64 -o /usr/local/bin/docker-compose
$ sudo chmod +x /usr/local/bin/docker-compose
$ curl -fsSL https://raw.githubusercontent.com/tilt-dev/tilt/master/scripts/install.sh | bash
$ tilt up --host 0.0.0.0 --port 10350
# or use normal docker compose
$ docker-compose up -d

Services

$ open http://[HOST_IP_ADDRESS]:8002 # RedisInsight
$ open http://[HOST_IP_ADDRESS]:9080 # Kafka Dashboard

Workbench

Visual Studio Code has a Jupyter extension which allows you to write code directly on a python file and by adding #%% transform it in an interactive cell. Try it out :)

It is recommended to install Miniconda in order to manage the different environments and packages. Below the installation instructions.

$ wget -P ~/ https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
$ chmod +x ~/Miniconda3-latest-Linux-x86_64.sh
$ ~/Miniconda3-latest-Linux-x86_64.sh -b
$ export PATH=~/miniconda3/bin:$PATH
$ conda init
# IMPORTANT: close and start a new session
$ conda config --set auto_activate_base false

Then you will have access to a conda enviorment. Try few commands to make sure everything works as expected

$ conda env list
$ conda create --name tiny python=3.8 -y
$ conda activate tiny
$ conda install -y pytorch torchvision cpuonly -c pytorch

Python Package Index (PIP) can be used inside a conda enviorment.

$ pip install fastapi
$ pip install "uvicorn[standard]"
$ pip install redis
$ pip install psycopg2-binary
$ pip install confluent-kafka
$ pip install httpx
$ pip install wandb
# once done installing, save the packages
# in a "requirements.txt" file
$ pip freeze > requirements.txt
# next time they can be installed directly
$ pip install -r requirements.txt

Scripts

$ ./scripts/setup_exporter.sh
$ conda activate tiny
$ set -a; source .env; set +a; python scripts/create_pg_schema.py
$ set -a; source .env; set +a; python scripts/load_items_into_pg.py
$ set -a; source .env; set +a; python scripts/load_users_into_pg.py
$ set -a; source .env; set +a; python scripts/load_items_into_redis.py
$ set -a; source .env; set +a; python scripts/load_users_into_redis.py

Dashboards