BacktestDatasetDefinion#

API documentation for tradeexecutor.backtest.preprocessed_backtest.BacktestDatasetDefinion Python class in Trading Strategy framework.

class BacktestDatasetDefinion[source]#

Bases: object

Predefined backtesting dataset

__init__(slug, name, description, chain, time_bucket, start, end, exchanges, always_included_pairs, reserve_token_address, min_tvl=None, min_weekly_volume=None, categories=None, max_fee=None, min_tokensniffer_score=None, filter_duplicates=True, formats=(<ExportFormat.parquet: 'parquet'>, ))#
Parameters:

  • slug (str) –

  • name (str) –

  • description (str) –

  • chain (ChainId) –

  • time_bucket (TimeBucket) –

  • start (<module 'datetime' from '/opt/hostedtoolcache/Python/3.11.13/x64/lib/python3.11/datetime.py'>) –

  • end (<module 'datetime' from '/opt/hostedtoolcache/Python/3.11.13/x64/lib/python3.11/datetime.py'>) –

  • exchanges (set[str]) –

  • always_included_pairs (list[tuple]) –

  • reserve_token_address (str) –

  • min_tvl (float | None) –

  • min_weekly_volume (float | None) –

  • categories (list[str] | None) –

  • max_fee (float | None) –

  • min_tokensniffer_score (int | None) –

  • filter_duplicates (bool) –

  • formats (tuple[tradeexecutor.backtest.preprocessed_backtest.ExportFormat]) –

Return type:

None

Methods

__init__(slug, name, description, chain, ...)

Attributes

categories

filter_duplicates

If we have multiple base/quote matches, try to filter out for the best pair

formats

max_fee

min_tokensniffer_score

min_tvl

Prefilter pairs with this liquidity before calling token sniffer

min_weekly_volume

Filter used in the reporting notebook.

slug

name

description

chain

time_bucket

start

end

exchanges

always_included_pairs

Pair descriptions that are always included, regardless of min_tvl and category filtering

reserve_token_address

The main USDC/USDT token on the chain
always_included_pairs: list[tuple]#

Pair descriptions that are always included, regardless of min_tvl and category filtering

reserve_token_address: str#

The main USDC/USDT token on the chain

We use this to generate equally-weighted index report and as a reserve token in this index.

min_tvl: float | None = None#

Prefilter pairs with this liquidity before calling token sniffer

min_weekly_volume: float | None = None#

Filter used in the reporting notebook.

Note that you still need to do actual volum filtering in the dataset yourself, as volume 0 days are exported.

filter_duplicates: bool = True#

If we have multiple base/quote matches, try to filter out for the best pair

__init__(slug, name, description, chain, time_bucket, start, end, exchanges, always_included_pairs, reserve_token_address, min_tvl=None, min_weekly_volume=None, categories=None, max_fee=None, min_tokensniffer_score=None, filter_duplicates=True, formats=(<ExportFormat.parquet: 'parquet'>, ))#
Parameters:

  • slug (str) –

  • name (str) –

  • description (str) –

  • chain (ChainId) –

  • time_bucket (TimeBucket) –

  • start (<module 'datetime' from '/opt/hostedtoolcache/Python/3.11.13/x64/lib/python3.11/datetime.py'>) –

  • end (<module 'datetime' from '/opt/hostedtoolcache/Python/3.11.13/x64/lib/python3.11/datetime.py'>) –

  • exchanges (set[str]) –

  • always_included_pairs (list[tuple]) –

  • reserve_token_address (str) –

  • min_tvl (float | None) –

  • min_weekly_volume (float | None) –

  • categories (list[str] | None) –

  • max_fee (float | None) –

  • min_tokensniffer_score (int | None) –

  • filter_duplicates (bool) –

  • formats (tuple[tradeexecutor.backtest.preprocessed_backtest.ExportFormat]) –

Return type:

None