Skip to content

Liquidity provision strategy with reinforcement learning - #28

Open
sssay10 wants to merge 33 commits into
Logarithm-Labs:devfrom
sssay10:lp_strategy_with_rl
Open

Liquidity provision strategy with reinforcement learning#28
sssay10 wants to merge 33 commits into
Logarithm-Labs:devfrom
sssay10:lp_strategy_with_rl

Conversation

@sssay10

@sssay10 sssay10 commented May 12, 2025

Copy link
Copy Markdown

Code Changelog

  • Added a query to the graph to get hourly aggregations for uniswap v3 data.
  • New fields have been added to the global state and internal state of UniswapV3LPEntity for convenient calculation of impermanent loss, earned fees for the needs of training RL based strategy.
  • Updated the logic for calculating the trading fee in UniswapV3LPEntity. Now the tax is taken only for the transfer of token_1 to token_0. ​​This is less than it was before.
  • Added gymnasium like environment for training RL-agent
  • Added class for calculating various statistics useful for training RL-agent.
  • Added class implementing strategy based on RL.
  • Minor fixes for tests and linters have been made.

Checklist:

  • PR to dev branch
  • Label of PR
  • Review is requested up
  • tests created & runned

0xBoringWozniak and others added 30 commits July 9, 2024 00:03
* Fix Logarithm-Labs#2
* Fix Logarithm-Labs#4
* Fix Logarithm-Labs#5
* Fix Logarithm-Labs#8
* Rename loaders tests due to conflict
* Add test for calculate_repay
* Fix Logarithm-Labs#10
* Fix Lido APR
* hyperliquid prices and FR loader
* hyperliquid prices and FR loader
* Hyperliquid loader and tests update
* Fix .idea artifacts
* Update Hyperliquid Entities
* minor fixes Hyperliquid Entities
* tau strategy initial
* Add tau strategy examples
* Tau Strategy: Add minutes data
* Fix timestamps to UTC
* Create an example pipeline with AI Agent trader.

* isort formatting

* Linters refactoring

* Update fractal/core/entities/single_spot_exchange.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update examples/agent_trader/agent_strategy.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update fractal/core/launcher.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- set UTC datetime only
- upgrade tests structure
- update imports
@0xBoringWozniak
0xBoringWozniak self-requested a review May 13, 2025 17:11
@0xBoringWozniak 0xBoringWozniak added research Notebook / study / strategy write-up under examples/ — research artifact, not framework code. new strategy labels May 13, 2025
@0xBoringWozniak 0xBoringWozniak added area: strategy fractal/strategies/ — BasisTrading, HyperliquidBasis, TauReset. and removed new strategy labels May 1, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area: strategy fractal/strategies/ — BasisTrading, HyperliquidBasis, TauReset. research Notebook / study / strategy write-up under examples/ — research artifact, not framework code.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants