SimonBlanke /
Hyperactive
| Total Complexity | 6 |
| Total Lines | 42 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
| 1 | # Author: Simon Blanke |
||
| 2 | # Email: [email protected] |
||
| 3 | # License: MIT License |
||
| 4 | |||
| 5 | import os |
||
| 6 | import numpy as np |
||
| 7 | |||
| 8 | from sklearn.externals import joblib |
||
| 9 | |||
| 10 | |||
| 11 | def find_best_hyperpara(features, scores): |
||
| 12 | N_best_features = 1 |
||
| 13 | |||
| 14 | scores = np.array(scores) |
||
| 15 | index_best_scores = list(scores.argsort()[-N_best_features:][::-1]) |
||
| 16 | |||
| 17 | best_score = scores[index_best_scores][0] |
||
| 18 | best_features = features.iloc[index_best_scores] |
||
| 19 | |||
| 20 | return best_features, best_score |
||
| 21 | |||
| 22 | |||
| 23 | class Predictor: |
||
| 24 | def __init__(self): |
||
| 25 | pass |
||
| 26 | |||
| 27 | def search(self, X_test): |
||
| 28 | best_para, best_score = self._predict(X_test) |
||
| 29 | return best_para, best_score |
||
| 30 | |||
| 31 | def load_model(self, path): |
||
| 32 | if os.path.isfile(path): |
||
| 33 | self.meta_reg = joblib.load(path) |
||
| 34 | else: |
||
| 35 | print("File at path", path, "not found") |
||
| 36 | |||
| 37 | def _predict(self, X_test): |
||
| 38 | score_pred = self.meta_reg.predict(X_test) |
||
| 39 | best_features, best_score = find_best_hyperpara(X_test, score_pred) |
||
| 40 | |||
| 41 | return best_features, best_score |
||
| 42 |
