Skip to content

Add parallel model training (n_jobs) to LazyClassifier/LazyRegressor #35

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

Open
vile319 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add parallel model training (n_jobs) to LazyClassifier/LazyRegressor #35

Show file tree
Hide file tree
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
164 changes: 83 additions & 81 deletions src/protify/probes/lazy_predict.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import xgboost
import lightgbm
from tqdm import tqdm
from joblib import Parallel, delayed
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler, OneHotEncoder, OrdinalEncoder
Expand Down Expand Up @@ -248,6 +249,7 @@ def __init__(
predictions=False,
random_state=None,
classifiers="all",
n_jobs=1,
):
self.verbose = verbose
self.ignore_warnings = ignore_warnings
Expand All @@ -256,6 +258,7 @@ def __init__(
self.models = {}
self.random_state = random_state or get_global_seed()
self.classifiers = classifiers
self.n_jobs = n_jobs

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 @@ -323,9 +326,8 @@ def fit(self, X_train, X_test, y_train, y_test):
print_message(exception)
print_message("Invalid Classifier(s)")

pbar = tqdm(self.classifiers)
for name, model in pbar:
pbar.set_description(f"Training {name}")
def _fit_classifier(name, model):
"""Train a single classifier and return results."""
start = time.time()
try:
if "random_state" in model().get_params().keys():
Expand All @@ -341,63 +343,57 @@ def fit(self, X_train, X_test, y_train, y_test):
)

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)
b_accuracy = balanced_accuracy_score(y_test, y_pred)
f1 = f1_score(y_test, y_pred, average="weighted")
try:
roc_auc = roc_auc_score(y_test, y_pred)
except Exception as exception:
except:
roc_auc = None
if self.ignore_warnings is False:
print_message("ROC AUC couldn't be calculated for " + name)
print_message(exception)
names.append(name)
Accuracy.append(accuracy)
B_Accuracy.append(b_accuracy)
ROC_AUC.append(roc_auc)
F1.append(f1)
TIME.append(time.time() - start)
fit_time = time.time() - start
custom = self.custom_metric(y_test, y_pred) if self.custom_metric else None
return {"name": name, "pipe": pipe, "y_pred": y_pred, "accuracy": accuracy,
"b_accuracy": b_accuracy, "roc_auc": roc_auc, "f1": f1,
"time": fit_time, "custom": custom, "failed": False}
except Exception as e:
return {"name": name, "failed": True, "error": e}

# Parallel or sequential execution
if self.n_jobs != 1:
results = Parallel(n_jobs=self.n_jobs, prefer="threads")(
delayed(_fit_classifier)(name, model) for name, model in self.classifiers
)
else:
results = [_fit_classifier(name, model) for name, model in tqdm(self.classifiers, desc="Training classifiers")]

# Collect results
for r in results:
if r["failed"]:
if self.ignore_warnings is False:
print_message(f'\n{r["name"]} model failed to execute')
print_message(r["error"])
else:
self.models[r["name"]] = r["pipe"]
names.append(r["name"])
Accuracy.append(r["accuracy"])
B_Accuracy.append(r["b_accuracy"])
ROC_AUC.append(r["roc_auc"])
F1.append(r["f1"])
TIME.append(r["time"])
if self.custom_metric is not None:
custom_metric = self.custom_metric(y_test, y_pred)
CUSTOM_METRIC.append(custom_metric)

CUSTOM_METRIC.append(r["custom"])
if self.verbose > 0:
if self.custom_metric is not None:
print_message(
{
"Model": name,
"Accuracy": accuracy,
"Balanced Accuracy": b_accuracy,
"ROC AUC": roc_auc,
"F1 Score": f1,
"Custom Metric": custom_metric,
"Time taken": time.time() - start,
}
)
print_message({"Model": r["name"], "Accuracy": r["accuracy"],
"Balanced Accuracy": r["b_accuracy"], "ROC AUC": r["roc_auc"],
"F1 Score": r["f1"], "Custom Metric": r["custom"], "Time taken": r["time"]})
else:
print_message(
{
"Model": name,
"Accuracy": accuracy,
"Balanced Accuracy": b_accuracy,
"ROC AUC": roc_auc,
"F1 Score": f1,
"Time taken": time.time() - start,
}
)
print_message({"Model": r["name"], "Accuracy": r["accuracy"],
"Balanced Accuracy": r["b_accuracy"], "ROC AUC": r["roc_auc"],
"F1 Score": r["f1"], "Time taken": r["time"]})
if self.predictions:
predictions[name] = y_pred


except Exception as exception:
if self.ignore_warnings is False:
print_message(f'\n{name} model failed to execute')
print_message(exception)
pbar.update(1)
pbar.close()
predictions[r["name"]] = r["y_pred"]

if self.custom_metric is None:
scores = pd.DataFrame(
Expand Down Expand Up @@ -492,6 +488,7 @@ def __init__(
predictions=False,
random_state=None,
regressors="all",
n_jobs=1,
):
self.verbose = verbose
self.ignore_warnings = ignore_warnings
Expand All @@ -500,6 +497,7 @@ def __init__(
self.models = {}
self.random_state = random_state or get_global_seed()
self.regressors = regressors
self.n_jobs = n_jobs

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 @@ -567,9 +565,10 @@ def fit(self, X_train, X_test, y_train, y_test):
print_message(exception)
print_message("Invalid Regressor(s)")

pbar = tqdm(self.regressors)
for name, model in pbar:
pbar.set_description(f"Training {name}")
n_test, n_features = X_test.shape[0], X_test.shape[1]

def _fit_regressor(name, model):
"""Train a single regressor and return results."""
start = time.time()
try:
if "random_state" in model().get_params().keys():
Expand All @@ -585,48 +584,51 @@ def fit(self, X_train, X_test, y_train, y_test):
)

pipe.fit(X_train, y_train)
self.models[name] = pipe
y_pred = pipe.predict(X_test)

r_squared = r2_score(y_test, y_pred)
adj_rsquared = adjusted_rsquared(
r_squared, X_test.shape[0], X_test.shape[1]
)
adj_rsquared = adjusted_rsquared(r_squared, n_test, n_features)
rmse = np.sqrt(mean_squared_error(y_test, y_pred))

names.append(name)
R2.append(r_squared)
ADJR2.append(adj_rsquared)
RMSE.append(rmse)
TIME.append(time.time() - start)
fit_time = time.time() - start
custom = self.custom_metric(y_test, y_pred) if self.custom_metric else None
return {"name": name, "pipe": pipe, "y_pred": y_pred, "r2": r_squared,
"adj_r2": adj_rsquared, "rmse": rmse, "time": fit_time,
"custom": custom, "failed": False}
except Exception as e:
return {"name": name, "failed": True, "error": e}

# Parallel or sequential execution
if self.n_jobs != 1:
results = Parallel(n_jobs=self.n_jobs, prefer="threads")(
delayed(_fit_regressor)(name, model) for name, model in self.regressors
)
else:
results = [_fit_regressor(name, model) for name, model in tqdm(self.regressors, desc="Training regressors")]

# Collect results
for r in results:
if r["failed"]:
if self.ignore_warnings is False:
print_message(f'\n{r["name"]} model failed to execute')
print_message(r["error"])
else:
self.models[r["name"]] = r["pipe"]
names.append(r["name"])
R2.append(r["r2"])
ADJR2.append(r["adj_r2"])
RMSE.append(r["rmse"])
TIME.append(r["time"])
if self.custom_metric:
custom_metric = self.custom_metric(y_test, y_pred)
CUSTOM_METRIC.append(custom_metric)

CUSTOM_METRIC.append(r["custom"])
if self.verbose > 0:
scores_verbose = {
"Model": name,
"R-Squared": r_squared,
"Adjusted R-Squared": adj_rsquared,
"RMSE": rmse,
"Time taken": time.time() - start,
}

scores_verbose = {"Model": r["name"], "R-Squared": r["r2"],
"Adjusted R-Squared": r["adj_r2"], "RMSE": r["rmse"], "Time taken": r["time"]}
if self.custom_metric:
scores_verbose[self.custom_metric.__name__] = custom_metric

scores_verbose[self.custom_metric.__name__] = r["custom"]
print_message(scores_verbose)
if self.predictions:
predictions[name] = y_pred

except Exception as exception:
if self.ignore_warnings is False:
print_message(f'\n{name} model failed to execute')
print_message(exception)

pbar.update(1)
pbar.close()
predictions[r["name"]] = r["y_pred"]

scores = {
"Model": names,
Expand Down
48 changes: 48 additions & 0 deletions src/protify/testing_suite/test_lazy_predict_sped_up.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,48 @@
# Example usage:
# python -m src.protify.testing_suite.test_lazy_predict_sped_up

import time
import numpy as np
from src.protify.probes.lazy_predict import LazyClassifier, LazyRegressor

# Larger synthetic data for more reliable benchmarking
X = np.random.randn(1000, 256) # 1000 samples, 256 features
y_cls = np.random.randint(0, 2, 1000)
y_reg = np.random.randn(1000)

NUM_RUNS = 3
times = []

for run in range(NUM_RUNS):
print(f"\n=== Run {run + 1}/{NUM_RUNS} ===")
run_start = time.time()

# Classifier (n_jobs=-1 uses all cores)
clf = LazyClassifier(classifiers="all", verbose=0, n_jobs=-1)
clf_scores = clf.fit(X[:800], X[800:], y_cls[:800], y_cls[800:])

# Regressor (n_jobs=-1 uses all cores)
reg = LazyRegressor(regressors="all", verbose=0, n_jobs=-1)
reg_scores = reg.fit(X[:800], X[800:], y_reg[:800], y_reg[800:])

run_time = time.time() - run_start
times.append(run_time)
print(f"Run {run + 1} time: {run_time:.2f}s")

print(f"\n=== Results ===")
print(f"Times: {[f'{t:.2f}s' for t in times]}")
print(f"Average: {np.mean(times):.2f}s")
print(f"Std: {np.std(times):.2f}s")

# BASELINE:
# Times: ['26.06s', '26.31s', '26.94s']
# Average: 26.43s
# Std: 0.37s
# PARALLELIZED VERSION:
# === Results ===
# Times: ['23.54s', '11.55s', '8.83s']
# Average: 14.64s
# Std: 6.39s
# Times: ['14.22s', '14.22s', '10.64s']
# Average: 13.03s
# Std: 1.69s