File size: 3,336 Bytes
b77b6a6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 |
import pandas as pd
import pickle
from web3_utils import DATA_DIR, TMP_DIR
from staking import check_list_addresses
def get_trader_type(address: str, service_map: dict) -> str:
# check if it is part of any service id on the map
keys = service_map.keys()
last_key = max(keys)
for key, value in service_map.items():
if value["safe_address"].lower() == address.lower():
# found a service
return "Olas"
return "non_Olas"
def compute_active_traders_dataset():
"""Function to prepare the active traders dataset"""
with open(DATA_DIR / "service_map.pkl", "rb") as f:
service_map = pickle.load(f)
# read tools info
tools_df = pd.read_parquet(TMP_DIR / "tools.parquet")
# rename the request_month_year_week
tools_df.rename(
columns={"request_month_year_week": "month_year_week"}, inplace=True
)
tool_traders = tools_df.trader_address.unique()
mapping = check_list_addresses(tool_traders)
# add trader type to tools_df
tools_df["trader_type"] = tools_df.trader_address.apply(lambda x: mapping[x])
tools_df = tools_df[
["month_year_week", "market_creator", "trader_type", "trader_address"]
]
tools_df.drop_duplicates(inplace=True)
# read trades info
all_trades = pd.read_parquet(DATA_DIR / "all_trades_profitability.parquet")
# read unknown info
unknown_traders = pd.read_parquet(DATA_DIR / "unknown_traders.parquet")
unknown_traders["creation_timestamp"] = pd.to_datetime(
unknown_traders["creation_timestamp"]
)
unknown_traders["creation_timestamp"] = unknown_traders[
"creation_timestamp"
].dt.tz_convert("UTC")
unknown_traders = unknown_traders.sort_values(
by="creation_timestamp", ascending=True
)
unknown_traders["month_year_week"] = (
unknown_traders["creation_timestamp"]
.dt.to_period("W")
.dt.start_time.dt.strftime("%b-%d-%Y")
)
unknown_traders["trader_type"] = "unknown"
unknown_traders = unknown_traders[
["month_year_week", "trader_type", "market_creator", "trader_address"]
]
unknown_traders.drop_duplicates(inplace=True)
all_trades["creation_timestamp"] = pd.to_datetime(all_trades["creation_timestamp"])
all_trades["creation_timestamp"] = all_trades["creation_timestamp"].dt.tz_convert(
"UTC"
)
all_trades = all_trades.sort_values(by="creation_timestamp", ascending=True)
all_trades["month_year_week"] = (
all_trades["creation_timestamp"]
.dt.to_period("W")
.dt.start_time.dt.strftime("%b-%d-%Y")
)
all_trades["trader_type"] = all_trades["staking"].apply(
lambda x: "non_Olas" if x == "non_Olas" else "Olas"
)
all_trades = all_trades[
["month_year_week", "market_creator", "trader_type", "trader_address"]
]
all_trades.drop_duplicates(inplace=True)
filtered_traders_data = pd.concat([all_trades, tools_df], axis=0)
filtered_traders_data.drop_duplicates(inplace=True)
if len(unknown_traders) > 0:
# merge
filtered_traders_data = pd.concat(
[filtered_traders_data, unknown_traders], axis=0
)
filtered_traders_data.to_parquet(TMP_DIR / "active_traders.parquet")
if __name__ == "__main__":
compute_active_traders_dataset()
|