Yearn DATA Analytics

Research Reports

We share the research outputs from the YFI Data Analysis team in the research directory.

Vault-level Risk

API endpoint: https://d3971bp2359cnv.cloudfront.net/api

This project aims to provide additional quantitative metrics on top of the existing Risk Framework, to help discover informative risk measures and collect the data necessary to build the metrics. Furthermore, the project also aims to provide aggregated views for Yearn's Vaults and associated DeFi protocols. See the project README for more details.

~~Usage examples for this project can be seen in examples/risk.ipynb.~~

TODO: need to rewrite examples for the new structure

Prerequisites

This project uses Poetry for dependency management. Please refer to the documentation for installing Poetry.

Run the following command to install the dependencies:

poetry install

Then run the following command to install the git pre-commit hooks:

pre-commit install

You will also need to set up the Web3 provider endpoints and chain explorers in the environment file .env. The necessary variables and some of their default values are shown in .env.example:

# Mainnet
ETH_PROVIDER=
ETHERSCAN_TOKEN=

# Fantom
FTM_PROVIDER=https://rpc.ftm.tools/
FTMSCAN_TOKEN=

# Arbitrum
ARB_PROVIDER=https://arb1.arbitrum.io/rpc
ARBISCAN_TOKEN=

# Database
DATABASE_URI=sqlite:////data/api_cache.sqlite3

Furthermore, you need to have Docker and Docker Compose that can support compose file format of version 2.0 or higher, see the documentation for the list of compatible versions and installation guides.

Usage Examples

Running the API

To run the API locally at port 80:

docker-compose up --build

Auto-generated docs can be seen at localhost/docs

DevContainer

If you're using VSCode, you can use DevContainer to setup all the required dependencies (almost) automatically. But you first need to set up ./.env file though. If you have filled out ./.env, open cloned dir in VSC and just press Reopen in Container button on the right bottom side.

Or use command palette to search Remote-Containers: Reopen in Container. Then VSC will do the rest. pytest in the integrated terminal after the build process. You'll pass all tests if your DevContainer has no problem and .env is set properly.