Include more metrics in regression helper methods.
junos
parent
95ab66fd81
commit
8131626c4a
|
|
@ -1,6 +1,6 @@
|
|||
from pathlib import Path
|
||||
from sklearn import linear_model, svm, kernel_ridge, gaussian_process, ensemble
|
||||
from sklearn.model_selection import LeaveOneGroupOut, cross_val_score, cross_validate
|
||||
from sklearn.model_selection import LeaveOneGroupOut, cross_validate, cross_validate
|
||||
from sklearn.metrics import mean_squared_error, r2_score
|
||||
from sklearn.impute import SimpleImputer
|
||||
from sklearn.dummy import DummyRegressor
|
||||
|
|
@ -66,7 +66,7 @@ def construct_full_path(folder: Path, filename_prefix: str, data_type: str) -> P
|
|||
def insert_row(df, row):
|
||||
return pd.concat([df, pd.DataFrame([row], columns=df.columns)], ignore_index=True)
|
||||
|
||||
def prepare_model_input(input_csv):
|
||||
def prepare_regression_model_input(input_csv):
|
||||
|
||||
model_input = pd.read_csv(input_csv)
|
||||
|
||||
|
|
@ -76,10 +76,9 @@ def prepare_model_input(input_csv):
|
|||
data_x, data_y, data_groups = model_input.drop(["target", "pid"], axis=1), model_input["target"], model_input["pid"]
|
||||
|
||||
categorical_feature_colnames = ["gender", "startlanguage", "limesurvey_demand_control_ratio_quartile"]
|
||||
additional_categorical_features = [col for col in data_x.columns if "mostcommonactivity" in col or "homelabel" in col]
|
||||
categorical_feature_colnames += additional_categorical_features
|
||||
#TODO: check whether limesurvey_demand_control_ratio_quartile NaNs could be replaced meaningfully
|
||||
#additional_categorical_features = [col for col in data_x.columns if "mostcommonactivity" in col or "homelabel" in col]
|
||||
#TODO: check if mostcommonactivity is indeed a categorical features after aggregating
|
||||
#categorical_feature_colnames += additional_categorical_features
|
||||
categorical_features = data_x[categorical_feature_colnames].copy()
|
||||
mode_categorical_features = categorical_features.mode().iloc[0]
|
||||
# fillna with mode
|
||||
|
|
@ -96,172 +95,227 @@ def prepare_model_input(input_csv):
|
|||
return train_x, data_y, data_groups
|
||||
|
||||
|
||||
def run_all_models(input_csv):
|
||||
def run_all_regression_models(input_csv):
|
||||
# Prepare data
|
||||
train_x, data_y, data_groups = prepare_model_input(input_csv)
|
||||
data_x, data_y, data_groups = prepare_regression_model_input(input_csv)
|
||||
|
||||
# Prepare cross validation
|
||||
logo = LeaveOneGroupOut()
|
||||
logo.get_n_splits(
|
||||
train_x,
|
||||
data_x,
|
||||
data_y,
|
||||
groups=data_groups,
|
||||
)
|
||||
scores = pd.DataFrame(columns=["method", "median", "max"])
|
||||
metrics = ['r2', 'neg_mean_absolute_error', 'neg_root_mean_squared_error']
|
||||
test_metrics = ["test_" + metric for metric in metrics]
|
||||
scores = pd.DataFrame(columns=["method", "max", "nanmedian"])
|
||||
|
||||
# Validate models
|
||||
lin_reg_rapids = linear_model.LinearRegression()
|
||||
lin_reg_scores = cross_val_score(
|
||||
lin_reg_rapids,
|
||||
X=train_x,
|
||||
dummy_regr = DummyRegressor(strategy="mean")
|
||||
dummy_regr_scores = cross_validate(
|
||||
dummy_regr,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring='r2'
|
||||
scoring=metrics
|
||||
)
|
||||
print("Dummy model:")
|
||||
print("R^2: ", np.nanmedian(dummy_regr_scores['test_r2']))
|
||||
|
||||
scores_df = pd.DataFrame(dummy_regr_scores)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "dummy"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
lin_reg_rapids = linear_model.LinearRegression()
|
||||
lin_reg_scores = cross_validate(
|
||||
lin_reg_rapids,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring=metrics
|
||||
)
|
||||
print("Linear regression:")
|
||||
print(np.median(lin_reg_scores))
|
||||
scores = insert_row(scores, ["Linear regression",np.median(lin_reg_scores),np.max(lin_reg_scores)])
|
||||
print("R^2: ", np.nanmedian(lin_reg_scores['test_r2']))
|
||||
|
||||
scores_df = pd.DataFrame(lin_reg_scores)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "linear_reg"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
ridge_reg = linear_model.Ridge(alpha=.5)
|
||||
ridge_reg_scores = cross_val_score(
|
||||
ridge_reg_scores = cross_validate(
|
||||
ridge_reg,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Ridge regression")
|
||||
print(np.median(ridge_reg_scores))
|
||||
scores = insert_row(scores, ["Ridge regression",np.median(ridge_reg_scores),np.max(ridge_reg_scores)])
|
||||
|
||||
scores_df = pd.DataFrame(ridge_reg_scores)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "ridge_reg"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
|
||||
lasso_reg = linear_model.Lasso(alpha=0.1)
|
||||
lasso_reg_score = cross_val_score(
|
||||
lasso_reg_score = cross_validate(
|
||||
lasso_reg,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Lasso regression")
|
||||
print(np.median(lasso_reg_score))
|
||||
scores = insert_row(scores, ["Lasso regression",np.median(lasso_reg_score),np.max(lasso_reg_score)])
|
||||
|
||||
scores_df = pd.DataFrame(lasso_reg_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "lasso_reg"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
bayesian_ridge_reg = linear_model.BayesianRidge()
|
||||
bayesian_ridge_reg_score = cross_val_score(
|
||||
bayesian_ridge_reg_score = cross_validate(
|
||||
bayesian_ridge_reg,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Bayesian Ridge")
|
||||
print(np.median(bayesian_ridge_reg_score))
|
||||
scores = insert_row(scores, ["Bayesian Ridge",np.median(bayesian_ridge_reg_score),np.max(bayesian_ridge_reg_score)])
|
||||
|
||||
scores_df = pd.DataFrame(bayesian_ridge_reg_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "bayesian_ridge"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
ransac_reg = linear_model.RANSACRegressor()
|
||||
ransac_reg_score = cross_val_score(
|
||||
ransac_reg_score = cross_validate(
|
||||
ransac_reg,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("RANSAC (outlier robust regression)")
|
||||
print(np.median(ransac_reg_score))
|
||||
scores = insert_row(scores, ["RANSAC",np.median(ransac_reg_score),np.max(ransac_reg_score)])
|
||||
|
||||
scores_df = pd.DataFrame(ransac_reg_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "RANSAC"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
svr = svm.SVR()
|
||||
svr_score = cross_val_score(
|
||||
svr_score = cross_validate(
|
||||
svr,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Support vector regression")
|
||||
print(np.median(svr_score))
|
||||
scores = insert_row(scores, ["Support vector regression",np.median(svr_score),np.max(svr_score)])
|
||||
|
||||
scores_df = pd.DataFrame(svr_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "SVR"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
kridge = kernel_ridge.KernelRidge()
|
||||
kridge_score = cross_val_score(
|
||||
kridge_score = cross_validate(
|
||||
kridge,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Kernel Ridge regression")
|
||||
print(np.median(kridge_score))
|
||||
scores = insert_row(scores, ["Kernel Ridge regression",np.median(kridge_score),np.max(kridge_score)])
|
||||
|
||||
scores_df = pd.DataFrame(kridge_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "kernel_ridge"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
gpr = gaussian_process.GaussianProcessRegressor()
|
||||
gpr_score = cross_val_score(
|
||||
gpr_score = cross_validate(
|
||||
gpr,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Gaussian Process Regression")
|
||||
print(np.median(gpr_score))
|
||||
scores = insert_row(scores, ["Gaussian Process Regression",np.median(gpr_score),np.max(gpr_score)])
|
||||
|
||||
scores_df = pd.DataFrame(gpr_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "gaussian_proc"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
rfr = ensemble.RandomForestRegressor(max_features=0.3, n_jobs=-1)
|
||||
rfr_score = cross_val_score(
|
||||
rfr_score = cross_validate(
|
||||
rfr,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("Random Forest Regression")
|
||||
print(np.median(rfr_score))
|
||||
scores = insert_row(scores, ["Random Forest Regression",np.median(rfr_score),np.max(rfr_score)])
|
||||
|
||||
scores_df = pd.DataFrame(rfr_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "random_forest"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
xgb = XGBRegressor()
|
||||
xgb_score = cross_val_score(
|
||||
xgb_score = cross_validate(
|
||||
xgb,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("XGBoost Regressor")
|
||||
print(np.median(xgb_score))
|
||||
scores = insert_row(scores, ["XGBoost Regressor",np.median(xgb_score),np.max(xgb_score)])
|
||||
|
||||
scores_df = pd.DataFrame(xgb_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "XGBoost"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
ada = ensemble.AdaBoostRegressor()
|
||||
ada_score = cross_val_score(
|
||||
ada_score = cross_validate(
|
||||
ada,
|
||||
X=train_x,
|
||||
X=data_x,
|
||||
y=data_y,
|
||||
groups=data_groups,
|
||||
cv=logo,
|
||||
n_jobs=-1,
|
||||
scoring="r2"
|
||||
scoring=metrics
|
||||
)
|
||||
print("ADA Boost Regressor")
|
||||
print(np.median(ada_score))
|
||||
scores = insert_row(scores, ["ADA Boost Regressor",np.median(ada_score),np.max(ada_score)])
|
||||
|
||||
scores_df = pd.DataFrame(ada_score)[test_metrics]
|
||||
scores_df = scores_df.agg(['max', np.nanmedian]).transpose()
|
||||
scores_df["method"] = "ADA_boost"
|
||||
scores = pd.concat([scores, scores_df])
|
||||
|
||||
return scores
|
||||
|
|
|
|||
Loading…
Reference in New Issue