Created
July 2, 2019 08:58
-
-
Save worldmind/04974de7539ed6dac0bf3816b9522ac7 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import numpy as np | |
| import pandas as pd | |
| from sklearn.model_selection import GridSearchCV, train_test_split | |
| from sklearn.metrics import roc_auc_score, precision_score, recall_score, f1_score, accuracy_score | |
| import torch | |
| from torch import nn, tensor, optim | |
| from skorch import NeuralNetClassifier | |
| def get_data(): | |
| raw_data = pd.read_csv('train.csv') | |
| X = raw_data.drop(columns=['PassengerId', 'Survived', 'Name', 'Ticket', 'Cabin']) | |
| X = pd.get_dummies(X) | |
| X.fillna({'Age': X.Age.median()}, inplace=True) | |
| y = raw_data.Survived | |
| X = X.to_numpy(dtype=np.float32) | |
| y = y.to_numpy(dtype=np.int64) | |
| X_train, X_test, y_train, y_test = train_test_split(X, y) | |
| return X_train, X_test, y_train, y_test | |
| class MyNN(nn.Module): | |
| def __init__(self, num_units=10): | |
| super().__init__() | |
| self.net = nn.Sequential( | |
| nn.Linear(10, num_units), | |
| nn.ReLU(), | |
| nn.Dropout(0.5), | |
| nn.Linear(num_units, 10), | |
| nn.ReLU(), | |
| nn.Linear(10, 2), | |
| nn.Softmax() | |
| ) | |
| def forward(self, X, **kwargs): | |
| return self.net(X) | |
| X_train, X_test, y_train, y_test = get_data() | |
| net = NeuralNetClassifier( | |
| MyNN, | |
| max_epochs=12, | |
| lr=0.1, | |
| ) | |
| params = { | |
| 'lr': [0.05, 0.1], | |
| 'module__num_units': [10, 20], | |
| } | |
| selector = GridSearchCV(net, params, cv=5) | |
| selector.fit(X_train, y_train) | |
| best_model = selector.best_estimator_ | |
| y_pred = best_model.predict(X_test) | |
| print(selector.best_params_) | |
| print("score: {}, accuracy_score: {}, roc_auc_score: {}, precision: {}, recall: {}, f1: {}".format( | |
| best_model.score(X_test, y_test), | |
| accuracy_score(y_test, y_pred), | |
| roc_auc_score(y_test, y_pred), | |
| precision_score(y_test, y_pred), | |
| recall_score(y_test, y_pred), | |
| f1_score(y_test, y_pred) | |
| )) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment