Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat(timeout): add timeout option for lazy estimators #403

Open
wants to merge 1 commit into
base: dev
Choose a base branch
from
Open

feat(timeout): add timeout option for lazy estimators #403

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 40 additions & 2 deletions lazypredict/Supervised.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,9 @@
"""
# Author: Shankar Rao Pandala <[email protected]>

import contextlib
import signal

import numpy as np
import pandas as pd
from tqdm import tqdm
Expand Down Expand Up @@ -110,6 +113,21 @@

# Helper function

@contextlib.contextmanager
def timeout_ctx(timeout: int, model_name: str):
if not isinstance(timeout, int) and timeout <= 0:
raise TypeError("timeout parameter must be an integer and greater than 0.")

def _handle_timeout(signum, frame):
raise TimeoutError(f"{model_name!r} timed out")

signal.signal(signal.SIGALRM, _handle_timeout)
signal.alarm(timeout)
try:
yield
finally:
signal.alarm(0)


def get_card_split(df, cols, n=11):
"""
Expand Down Expand Up @@ -154,6 +172,9 @@ class LazyClassifier:
When set to True, the predictions of all the models models are returned as dataframe.
classifiers : list, optional (default="all")
When function is provided, trains the chosen classifier(s).
timeout : int, optional (default=None)
Timeout parameter set for each model fit. If not None and > 0, the model will be fitting for a maximum of
`timout` seconds, before raising a `TimeoutError` and moving onto the next model.

Examples
--------
Expand Down Expand Up @@ -210,6 +231,7 @@ def __init__(
predictions=False,
random_state=42,
classifiers="all",
timeout=None,
):
self.verbose = verbose
self.ignore_warnings = ignore_warnings
Expand All @@ -218,6 +240,7 @@ def __init__(
self.models = {}
self.random_state = random_state
self.classifiers = classifiers
self.timeout = timeout

def fit(self, X_train, X_test, y_train, y_test):
"""Fit Classification algorithms to X_train and y_train, predict and score on X_test, y_test.
Expand Down Expand Up @@ -300,7 +323,12 @@ def fit(self, X_train, X_test, y_train, y_test):
steps=[("preprocessor", preprocessor), ("classifier", model())]
)

pipe.fit(X_train, y_train)
if self.timeout is not None:
with timeout_ctx(self.timeout, model_name=name):
pipe.fit(X_train, y_train)
else:
pipe.fit(X_train, y_train)

self.models[name] = pipe
y_pred = pipe.predict(X_test)
accuracy = accuracy_score(y_test, y_pred, normalize=True)
Expand Down Expand Up @@ -436,6 +464,9 @@ class LazyRegressor:
When set to True, the predictions of all the models models are returned as dataframe.
regressors : list, optional (default="all")
When function is provided, trains the chosen regressor(s).
timeout : int, optional (default=None)
Timeout parameter set for each model fit. If not None and > 0, the model will be fitting for a maximum of
`timout` seconds, before raising a `TimeoutError` and moving onto the next model.

Examples
--------
Expand Down Expand Up @@ -510,6 +541,7 @@ def __init__(
predictions=False,
random_state=42,
regressors="all",
timeout=None,
):
self.verbose = verbose
self.ignore_warnings = ignore_warnings
Expand All @@ -518,6 +550,7 @@ def __init__(
self.models = {}
self.random_state = random_state
self.regressors = regressors
self.timeout = timeout

def fit(self, X_train, X_test, y_train, y_test):
"""Fit Regression algorithms to X_train and y_train, predict and score on X_test, y_test.
Expand Down Expand Up @@ -600,7 +633,12 @@ def fit(self, X_train, X_test, y_train, y_test):
steps=[("preprocessor", preprocessor), ("regressor", model())]
)

pipe.fit(X_train, y_train)
if self.timeout is not None:
with timeout_ctx(self.timeout, model_name=name):
pipe.fit(X_train, y_train)
else:
pipe.fit(X_train, y_train)

self.models[name] = pipe
y_pred = pipe.predict(X_test)

Expand Down