def startjob(
csv_file_path=r'C:\Users\57855\Desktop\2%test.csv', #训练文件的路径
model_file=r'C:\Users\57855\Desktop\2%.yaml', #模型配置文件路径
test_file=r'C:\Users\57855\Desktop\2%test_data.csv'): #结果输出路径
#Lugwig教程上的代码
with open(model_file, encoding='utf-8', mode='r') as file:
model_definition = yaml.load(file.read())
print(model_definition)
ludwig_model = LudwigModel(model_definition)
train_stats = ludwig_model.train(csv_file_path,
logging_level=logging_DEBUG)
print(train_stats)
predictions = ludwig_model.predict(test_file,
logging_level=logging_DEBUG)
print(predictions)
ludwig_model.close()
'name': 'oldpeak',
'type': 'numerical',
'encoder': 'rnn'
}, {
'name': 'slope',
'type': 'category',
'encoder': 'rnn'
}, {
'name': 'ca',
'type': 'category',
'encoder': 'rnn'
}, {
'name': 'thal',
'type': 'category',
'encoder': 'rnn'
}],
'output_features': [{
'name': 'target',
'type': 'binary'
}],
'training': {
'epochs': 10
}
}
model = LudwigModel(model_definition)
train_stats = model.train(data)
# obtain predictions
predictions = model.predict(data)
model.close()
},
{
'name': 'pe',
'type': 'category'
},
{
'name': 'ane',
'type': 'category'
},
],
'output_features': [{
'name': 'classification',
'type': 'binary'
}]
}
print('creating model')
model = LudwigModel(model_definition)
print('training model')
train_stats = model.train(data_df=train_df)
#Run predictions
predictions = model.predict(data_df=X_test)
predictions["classification_predictions"] = np.where(
predictions.iloc[:, 0] == True, 1, 0)
print(accuracy_score(Y_test, predictions.iloc[:, 0]))
print(confusion_matrix(Y_test, predictions.iloc[:, 0]))
print(classification_report(Y_test, predictions.iloc[:, 0]))
model.close()