Spaces:
Sleeping
Sleeping
File size: 3,370 Bytes
2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b c219cd7 2dff47b |
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 |
import pandas as pd
import numpy as np
from tqdm import tqdm
import lightgbm as lgb
from sklearn.model_selection import TimeSeriesSplit
from intraCols import model_cols
def walk_forward_validation(df, target_column, num_periods, mode='full'):
df = df[model_cols + [target_column]]
df[target_column] = df[target_column].astype(bool)
tscv = TimeSeriesSplit(n_splits=len(df)-1, max_train_size=None, test_size=num_periods) # num_splits is the number of splits you want
if mode == 'full':
overall_results = []
# Iterate over the rows in the DataFrame, one step at a time
# Split the time series data using TimeSeriesSplit
for train_index, test_index in tqdm(tscv.split(df), total=tscv.n_splits):
# Extract the training and testing data for the current split
X_train = df.drop(target_column, axis=1).iloc[train_index]
y_train = df[target_column].iloc[train_index]
X_test = df.drop(target_column, axis=1).iloc[test_index]
y_test = df[target_column].iloc[test_index]
y_train = y_train.astype(bool)
model = lgb.LGBMClassifier(n_estimators=10, random_state=42, verbosity=-1)
model.fit(X_train, y_train)
# Make a prediction on the test data
predictions = model.predict_proba(X_test)[:,-1]
# Create a DataFrame to store the true and predicted values
result_df = pd.DataFrame({'IsTrue': y_test, 'Predicted': predictions}, index=y_test.index)
overall_results.append(result_df)
df_results = pd.concat(overall_results)
# Calibrate Probabilities
def get_quantiles(df, col_name, q):
return df.groupby(pd.cut(df[col_name], q))['IsTrue'].mean()
greenprobas = []
pvals = []
for i, pct in tqdm(enumerate(df_results['Predicted']), desc='Calibrating Probas',total=len(df_results)):
try:
df_q = get_quantiles(df_results.iloc[:i], 'Predicted', 10)
for q in df_q.index:
if q.left <= pct <= q.right:
p = df_q[q]
calib_scores = np.abs(df_results['Predicted'].iloc[:i] - 0.5)
score = abs(df_results['Predicted'].iloc[i] - 0.5)
pv = np.mean(calib_scores >= score)
except:
p = None
pv = None
greenprobas.append(p)
pvals.append(pv)
df_results['CalibPredicted'] = greenprobas
df_results['Pvalue'] = pvals
elif mode == 'single':
X_train = df.drop(target_column, axis=1).iloc[:-1]
y_train = df[target_column].iloc[:-1]
X_test = df.drop(target_column, axis=1).iloc[-1]
y_test = df[target_column].iloc[-1]
y_train = y_train.astype(bool)
model = lgb.LGBMClassifier(n_estimators=10, random_state=42, verbosity=-1)
model.fit(X_train, y_train)
predictions = model.predict_proba(X_test.values.reshape(1, -1))[:,-1]
df_results = pd.DataFrame({'IsTrue': y_test, 'Predicted': predictions}, index=[df.index[-1]])
return df_results, model
def seq_predict_proba(df, trained_clf_model):
clf_pred_proba = trained_clf_model.predict_proba(df[model_cols])[:,-1]
return clf_pred_proba |