model_red = catboost.CatBoostClassifier()
model_red_GPU = catboost.CatBoostClassifier()
model_red.load_model("./catboost modele i wyniki/12 depth/CPU RED basic")
model_red_GPU.load_model(
"./catboost modele i wyniki/12 depth/GPU RED basic")
if model_red.tree_count_ > 10 and model_red_GPU.tree_count_ > 10:
os.remove("./catboost modele i wyniki/12 depth/CPU RED basic")
os.remove("./catboost modele i wyniki/12 depth/GPU RED basic")
model_red.shrink(model_red.tree_count_, model_red.tree_count_ - 2)
model_red_GPU.shrink(model_red_GPU.tree_count_,
model_red_GPU.tree_count_ - 2)
model_red.save_model(
"./catboost modele i wyniki/12 depth/CPU RED basic")
model_red_GPU.save_model(
"./catboost modele i wyniki/12 depth/GPU RED basic")
red_data_training, red_data_test, red_quality_training, red_quality_test, white_data_training, white_data_test, white_quality_training, white_quality_test = wines_import.read_data(
False)
red_quality_predicted_CPU = model_red.predict(red_data_test)
red_quality_predicted_GPU = model_red_GPU.predict(red_data_test)
print("\nCatboost uczony CPU wyniki\n")
print(
metrics.classification_report(red_quality_test,
red_quality_predicted_CPU,
zero_division=0))
print(metrics.confusion_matrix(red_quality_test,
red_quality_predicted_CPU))
print("\nCatboost uczony GPU wyniki\n")
print(
metrics.classification_report(red_quality_test,
red_quality_predicted_GPU,
zero_division=0))
#import needed library
from wines_import import read_data
import pandas as pd
import time
#import linear model and train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn import metrics
#Read the data file
data=pd.read_csv('../Data/winequality-white.csv')
<<<<<<< HEAD
red_data_training, red_data_test, red_quality_training, red_quality_test,white_data_training,white_data_test, white_quality_training, white_quality_test = read_data(True)
time_before=time.time()
reg = LogisticRegression(solver='saga', random_state=42, max_iter=2000, multi_class='auto')
=======
red_data_training, red_data_test, red_quality_training, red_quality_test,white_data_training,white_data_test, white_quality_training, white_quality_test = read_data(False)
reg = LogisticRegression()
>>>>>>> 103021ef48f3900bc979295db7647819d0bb5109
reg.fit(white_data_training, white_quality_training)
y_pred = reg.predict(white_data_test)
time_after=time.time()
print("Regression coefficient is ", reg.coef_)
#classification report
print(metrics.classification_report(white_quality_test, y_pred))
Conf_Mat = metrics.confusion_matrix(white_quality_test, y_pred)
print("The confusion matrix is\n", Conf_Mat)
#print("Accuracy is ",metrics.accuracy_score(white_quality_test, y_pred))
print("Czas wykonania: ",round(time_after-time_before,3)," sekund")
