├── README.md
├── Residential-Building-Data-Set.xlsx
├── final_model.py
├── paper_final_experiments.py
├── requirements.txt
├── sonar.all-data.csv
└── synth_data_experiments.py


/README.md:
--------------------------------------------------------------------------------
 1 | # Binary Mask Optimization Algorithm Repository
 2 | 
 3 | This repository provides the code for our paper: Binary Feature Mask Optimization for Feature Selection
 4 | The repository includes the used datasets.
 5 | 
 6 | ## Installation
 7 | 
 8 | 1. Clone the repository:
 9 | git clone https://github.com/mefe06/feature-selection.git
10 | 
11 | 2. Navigate to the repository directory:
12 | cd feature-selection
13 | 
14 | 3. Install the required packages:
15 | pip install -r requirements.txt
16 | 
17 | ## Results
18 | 
19 | 1. Run "python3 paper_final_experiments.py" for real life dataset results or "python3 synth_data_experiments.py" for synthetic data results.
20 | 
21 | ## Contributing
22 | 
23 | Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
24 | 
25 | ## License
26 | 
27 | This project is licensed under the MIT License.
28 | 


--------------------------------------------------------------------------------
/Residential-Building-Data-Set.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mefe06/feature-selection/8fe3bd92b99423ad0d0656806d87d9838433b91e/Residential-Building-Data-Set.xlsx


--------------------------------------------------------------------------------
/final_model.py:
--------------------------------------------------------------------------------
  1 | import lightgbm as lgb
  2 | import matplotlib.pyplot as plt
  3 | import numpy as np
  4 | from scipy.optimize import minimize
  5 | from sklearn.model_selection import RandomizedSearchCV
  6 | from sklearn.metrics import roc_auc_score, accuracy_score, f1_score, recall_score, precision_score, mean_squared_error,mean_absolute_error, log_loss
  7 | from sklearn.neural_network import MLPRegressor, MLPClassifier
  8 | from sklearn.feature_selection import RFE
  9 | from sklearn.ensemble import GradientBoostingClassifier, RandomForestClassifier
 10 | from sklearn.model_selection import cross_val_score, KFold,StratifiedKFold
 11 | from sklearn.neighbors import KNeighborsClassifier
 12 | from sklearn.metrics import make_scorer, roc_auc_score
 13 | import random
 14 | from sklearn.svm import SVR, SVC
 15 | from sklearn.feature_selection import mutual_info_classif, mutual_info_regression
 16 | import pandas as pd
 17 | import time
 18 | 
 19 | class LGBM_w_Feature_Selector():
 20 | 
 21 |     def __init__(self, layer_sizes = None, boosting_type=None, num_leaves=None, max_depth=None, learning_rate=0.01, subsample=None, colsample_bytree=None, reg_alpha=None,
 22 |                  reg_lambda=None, model="lgbm",problem_type="Classifier", objective="binary", param_grid=None,n_estimators=1000, random_state=42, target_name="Y",
 23 |                  early_stopping_rounds=100, X_train=None, X_test=None, X_val_1=None, X_val_2=None, y_val_1=None, y_val_2=None, y_train=None, show_loss_plot = False,
 24 |                  y_test=None, iterations=1000, slack=0.01):
 25 | 
 26 |         self.problem_type = problem_type
 27 |         if model=="lgbm":
 28 |             self.model_type = "lgbm"
 29 |             if problem_type == "Classifier":
 30 |                 self.model = lgb.LGBMClassifier(max_depth=max_depth,verbose= -1)
 31 |                 self.initial_model=lgb.LGBMClassifier(random_state=42, verbose= -1)
 32 |                 self.criterion = log_loss
 33 |                 self.cv_scoring = "neg_log_loss"
 34 | 
 35 |             else:
 36 | 
 37 |                 self.model = lgb.LGBMRegressor(max_depth=max_depth,verbose= -1)
 38 |                 self.initial_model=lgb.LGBMRegressor(random_state=42,verbose= -1)
 39 | 
 40 |                 self.criterion = mean_squared_error
 41 |                 self.cv_scoring = "neg_mean_squared_error"
 42 |         else:
 43 |             self.model_type = "mlp"
 44 | 
 45 |             if problem_type == "Classifier":
 46 |                 self.model = MLPClassifier(hidden_layer_sizes= layer_sizes, activation='relu', random_state=42)
 47 |                 self.initial_model = MLPClassifier(random_state=42)
 48 |                 self.criterion = log_loss
 49 |                 self.cv_scoring = "neg_log_loss"
 50 | 
 51 |             else:
 52 |             #self.model = MLPRegressor(hidden_layer_sizes=layer_sizes, warm_start=False, max_iter=500)
 53 |                 self.model = MLPRegressor(hidden_layer_sizes= layer_sizes, activation='relu', random_state=42)
 54 |                 self.initial_model = MLPRegressor(random_state=42)
 55 |                 self.criterion = mean_squared_error
 56 |                 self.cv_scoring = "neg_mean_squared_error"
 57 |         self.params = param_grid
 58 |         self.val_losses = []
 59 |         self.selected_feature_nbs_gbmo=[]
 60 |         self.best_ft_nbs = []
 61 |         self.early_stopping_rounds = early_stopping_rounds
 62 |         self.initial_model = self.model
 63 |         self.iterations = iterations
 64 |         self.X_train = X_train
 65 |         self.X_test = X_test
 66 |         self.X_val_1 = X_val_1
 67 |         self.X_val_2 = X_val_2
 68 |         self.X_full_train = np.vstack((X_train,X_val_1))
 69 |         self.slack = slack
 70 |         self.y_train = y_train
 71 |         self.y_val_1 = y_val_1
 72 |         self.y_val_2 = y_val_2
 73 |         self.y_test = y_test
 74 |         self.y_full_train = np.vstack((y_train,y_val_1))
 75 |         self.plot_loss = show_loss_plot
 76 |         self.num_of_features = self.X_train.shape[1]
 77 |         self.mask = np.ones(self.num_of_features)
 78 |         self.target_name=target_name
 79 |         #if problem_type == "Classifier":
 80 |         #    self.criterion = log_loss
 81 |         #else:
 82 |         #    self.criterion = mean_squared_error
 83 | 
 84 |     def shuffle_column(self,arr, i):
 85 |         # Copy the array to avoid modifying the original array
 86 |         arr_copy = np.copy(arr)
 87 |         
 88 |         # Get the ith column
 89 |         column = arr_copy[:, i]
 90 |         
 91 |         # Shuffle the values in the column
 92 |         np.random.shuffle(column)
 93 |         
 94 |         # Update the ith column in the copied array
 95 |         arr_copy[:, i] = column
 96 |         
 97 |         return arr_copy 
 98 |     def set_column_to_zero(self, input_array, i):
 99 |         if i < 0 or i >= input_array.shape[1]:
100 |             raise ValueError("Invalid column index")
101 | 
102 |         result_array = input_array.copy()
103 |         result_array[:, i] = 0
104 |         return result_array
105 |     # def general_feature_importance_calc(self):
106 |     #     self.model.fit(self.X_train, self.y_train)
107 |     #     inital_error = mean_squared_error(self.model.predict(self.X_val), self.y_val)
108 | 
109 |     #     importance_dict = {}
110 |     #     for feat in range(self.X_train.shape[1]):
111 |     #         importance_dict[feat] = mean_squared_error(self.model.predict(self.shuffle_column(self.X_val, feat)), self.y_val) - inital_error
112 | 
113 |     #     sorted_items = sorted(importance_dict.items(), key=lambda x: x[1],reverse=True)
114 | 
115 |     #     sorted_keys = [item[0] for item in sorted_items]
116 |     #     sorted_values = [item[1] for item in sorted_items]        
117 |     #     return sorted_keys, sorted_values
118 | 
119 | 
120 | 
121 |     def search(self,mask,x_val, y_val, shuffle = False):
122 |         min_loss = np.inf
123 |         least_useful_feature_ind = None
124 |         ret_mask = mask.copy()
125 |         losses = []
126 |         for i in range(len(self.mask)):
127 |             temp_mask = mask.copy()
128 |             if mask[i] != 0:
129 |                 if shuffle:
130 |                     x = self.shuffle_column(x_val, i)
131 |                 else:
132 |                     x = self.set_column_to_zero(x_val, i)
133 |                 # try: 
134 |                 #     temp_loss = self.criterion(self.best_model_opt_1.predict(x*temp_mask), y_val)
135 |                 # except: 
136 |                 if self.problem_type=="Classifier":
137 |                     temp_loss = self.criterion( y_val, self.model.predict_proba(x*temp_mask)[:,1])
138 |                 else: 
139 |                     temp_loss = self.criterion(self.model.predict(x*temp_mask), y_val)
140 | 
141 |                 losses.append(temp_loss)
142 |                 if temp_loss < min_loss:
143 |                     min_loss = temp_loss
144 |                     least_useful_feature_ind = i
145 |         
146 |         ret_mask[least_useful_feature_ind] = 0
147 |         return min_loss, ret_mask, losses
148 |         
149 |     def main_search_1(self, run_cv = False, p =0.1, loss_tolerance = 2, lamda=15):
150 |         ## get best hyperparameters on val
151 |         # if run_cv:
152 |         #     random_search = RandomizedSearchCV(estimator=self.model, param_distributions=self.params, n_iter=20, cv=5,
153 |         #                                         random_state=42, verbose=0,scoring=self.cv_scoring)
154 |         #     # Fit the model
155 |         #     random_search.fit(self.X_train, self.y_train)
156 | 
157 |         #     # Update the best model and best parameters
158 |         #     best_params_opt_1 = random_search.best_params_
159 |         #     self.best_model_opt_1 = random_search.best_estimator_
160 |         #     print("Best parameters for iteration {} are: {}".format(iter, best_params_opt_1))
161 |         #     self.model = self.best_model_opt_1
162 | 
163 |         self.model.fit(self.X_train, self.y_train)
164 |         cur_loss = 20
165 |         feat_nb = self.X_val_1.shape[1]
166 |         mask = np.ones(feat_nb)
167 |         #loss_tolerance = 2 #int(feat_nb/tol_param)
168 |         self.selected_feature_nbs_gbmo=[]
169 |         self.val_losses = []
170 | 
171 |         update_tolerance = 3
172 |         prev_loss = np.inf
173 |         loss_patience, update_patience = 0, 0
174 |         cur_loss, cur_mask, losses = self.search(mask, self.X_val_1, self.y_val_1)
175 |         self.val_losses.append(cur_loss)
176 |         prev_mask = np.zeros(feat_nb)
177 |         while (loss_patience < loss_tolerance) and (np.sum(mask)>2): #and (cur_loss<=self.val_losses[0]): #and (np.sum(mask)>2): #&(np.sum(mask)>feat_nb/lamda):
178 |             self.selected_feature_nbs_gbmo.append(np.sum(mask))
179 |             prev_loss = cur_loss
180 |             prev_mask = cur_mask
181 |             cur_loss, cur_mask, losses = self.search(mask, self.X_val_1, self.y_val_1)
182 |             self.val_losses.append(cur_loss)
183 | 
184 |             if (cur_loss<prev_loss*(1+self.slack)): 
185 |                 loss_patience = 0
186 |                 #if random.random()<p:
187 |                 mask = cur_mask            #print(mask)
188 |             else:
189 |                 loss_patience+=1
190 |             # if np.array_equal(cur_mask, prev_mask):
191 |             #     print("same feature selected")
192 |         
193 |         ### optimal mask
194 |         #print(mask)
195 |         return mask
196 |     def main_search_2(self, f_number,  run_cv = False,shuffle=False):
197 |             ## get best hyperparameters on val
198 |             # if run_cv:
199 |             #     random_search = RandomizedSearchCV(estimator=self.model, param_distributions=self.params, n_iter=20, cv=5,
200 |             #                                         random_state=42, verbose=0,scoring=self.cv_scoring)
201 |             #     # Fit the model
202 |             #     random_search.fit(self.X_train, self.y_train)
203 |     
204 |             #     # Update the best model and best parameters
205 |             #     best_params_opt_1 = random_search.best_params_
206 |             #     self.best_model_opt_1 = random_search.best_estimator_
207 |             #     print("Best parameters for iteration {} are: {}".format(iter, best_params_opt_1))
208 |             #     self.model = self.best_model_opt_1
209 |     
210 |             self.model.fit(self.X_train, self.y_train)
211 |             feat_nb = self.X_val_1.shape[1]
212 |             mask = np.ones(feat_nb)
213 |             self.selected_feature_nbs_flbmo=[]
214 |             while (np.sum(mask) > f_number):
215 |                 self.selected_feature_nbs_flbmo.append(np.sum(mask))
216 |                 
217 |                 cur_loss, cur_mask, losses = self.search(mask, self.X_val_1, self.y_val_1, shuffle = shuffle)
218 |                 mask = cur_mask            #print(mask)
219 | 
220 |             ### optimal mask
221 |             #print(mask)
222 |             return mask
223 | 
224 | 
225 |     # def CV_test(self, X,y):
226 |     #     '''
227 |     #     StratifiedKFold Cross-Validation with KNeighborsClassifier
228 |     #     '''
229 |     #     score = []
230 |     #     if self.problem_type=="Classifier":
231 |     #         cv = StratifiedKFold(5, random_state=42, shuffle = True)
232 |     #     else:
233 |     #         cv = KFold(5, random_state=42, shuffle = True)
234 |     #     #reg = MLPRegressor()
235 |     #     if self.problem_type == "Classifier":
236 |     #         #reg = SVC()
237 |     #         score.append(cross_val_score(self.model, X, y, cv=cv, scoring=make_scorer(roc_auc_score)).mean())
238 |     #     else: 
239 |     #         #reg = SVR()
240 |     #         score.append(cross_val_score(self.model, X, y, cv=cv, scoring=make_scorer(mean_squared_error)).mean())
241 |     # return np.mean(score)
242 | 
243 | 
244 |     def create_mask(self, selected_indices, n):
245 |         result_array = [0] * n  # Initialize the array with zeros
246 |         
247 |         for index in selected_indices:
248 |             if 0 <= index < n:
249 |                 result_array[index] = 1
250 |         
251 |         return result_array
252 | 
253 |     # def compare_fi(self, X,y,d, debug=False):
254 |     #     '''
255 |     #     Compare feature selection methods
256 |     #     '''
257 |     #     nn_lst = []
258 |     #     gb_list = []
259 |     #     acc_nn, acc_gg = [], []
260 |         
261 |     #     for f_nn, f_gb in zip(d['nn'], d['gb']):
262 |     #         nn_lst.append(f_nn)
263 |     #         gb_list.append(f_gb)
264 |     #         if debug:
265 |     #             print(nn_lst)
266 |     #             print(gb_list)
267 |     #         acc_nn.append(self.CV_test(X[:,nn_lst], y).mean())
268 |     #         acc_gg.append(self.CV_test(X[:,gb_list], y).mean())
269 |     #     return acc_nn, acc_gg
270 | 
271 |     # def compare_fi_f_nb(self, X,y,d, debug=False):
272 |     #     '''
273 |     #     Compare feature selection methods
274 |     #     '''
275 |     #     nn_lst = []
276 |     #     gb_list = []
277 |     #     acc_nn, acc_gg = [], []
278 |         
279 |     #     for f_nn, f_gb in zip(d['nn'], d['gb']):
280 |     #         nn_lst.append(f_nn)
281 |     #         gb_list.append(f_gb)
282 |     #         if debug:
283 |     #             print(nn_lst)
284 |     #             print(gb_list)
285 |     #         acc_nn.append(self.CV_test(X[:,nn_lst], y).mean())
286 |     #         acc_gg.append(self.CV_test(X[:,gb_list], y).mean())
287 |     #     return acc_nn, acc_gg
288 | 
289 |     def cv_for_other_fs_methods(self,f_numbers ):
290 |         best_gbm=1000000
291 |         best_mi=1000000
292 |         if self.problem_type=="Classifier":
293 |             mi_importances= mutual_info_classif(self.X_full_train, self.y_full_train)
294 |         else:
295 |             mi_importances= mutual_info_regression(self.X_full_train, self.y_full_train)
296 |         temp = np.hstack((self.X_full_train, self.y_full_train))#pd.concat([self.X_full_train, self.y_full_train], axis=1)
297 |         corr = np.corrcoef(temp, rowvar=False)
298 |         target_corr = corr[:-1, -1]
299 |         for f_number in f_numbers:
300 | 
301 |             mi_fi = mi_importances.argsort()[-f_number:][::-1]
302 |             gb_fi = np.argsort(np.abs(target_corr))[-f_number:]
303 |             mi_score = self.val_with_mask( np.expand_dims(self.create_mask(mi_fi, self.mask_len),axis=0))
304 |             gb_score = self.val_with_mask( np.expand_dims(self.create_mask(gb_fi, self.mask_len),axis=0))
305 |             if mi_score<best_mi:
306 |                 best_mi = mi_score
307 |                 print("f number:{}".format(str(f_number)))
308 |                 print(mi_fi)
309 |                 mi = self.val_with_mask( np.expand_dims(self.create_mask(mi_fi, self.mask_len),axis=0))
310 | 
311 |             if gb_score<best_gbm:
312 |                 best_gbm = gb_score            
313 |                 print("f number:{}".format(str(f_number)))
314 |                 print(gb_fi)
315 |                 gb = self.test_with_mask( np.expand_dims(self.create_mask(gb_fi, self.mask_len),axis=0))
316 |         return gb, mi
317 |     def cv_for_rfe(self,f_numbers):
318 |         best_rfe = 1000000
319 |         t0=time.process_time()
320 |         for f_number in f_numbers:
321 |             self.model.fit(self.X_full_train, self.y_full_train)
322 |             rfe = RFE(self.model, n_features_to_select=f_number) 
323 |             rfe.fit(self.X_full_train, self.y_full_train)
324 |             rfe_mask = 1*rfe.support_
325 |             rfe_score = self.val_with_mask( np.expand_dims(rfe_mask,axis=0))
326 |             if rfe_score<best_rfe:
327 |                 best_rfe= rfe_score   
328 |                 print("f number:{}".format(str(f_number)))
329 |                 print(rfe_mask)
330 |                 rfe = self.test_with_mask( np.expand_dims(rfe_mask ,axis=0))
331 |         t1=time.process_time()
332 |         print("rfe time:{}".format(str(t1-t0)))    
333 |         return best_rfe 
334 |     def cv_gbmo(self,loss_ts, slacks):
335 |         best_gbmo=1000000
336 |         t0=time.process_time()
337 |         for loss_t in loss_ts:
338 |             for slack in slacks:
339 |                 self.slack=slack
340 |                 cancelout_weights_importance = self.main_search_1(loss_tolerance = loss_t, run_cv=True)
341 |                 nn_fi = [i for i, x in enumerate(cancelout_weights_importance) if x == 1]
342 |                 mo_score = self.val_with_mask( np.expand_dims(self.create_mask(nn_fi, self.mask_len),axis=0))
343 |                 if mo_score<best_gbmo:
344 |                     self.best_ft_nbs=self.selected_feature_nbs_gbmo
345 |                     self.best_ft_val_losses = self.val_losses[1:]
346 |                     best_gbmo= mo_score           
347 |                     print("Loss t:{}, slack: {}, nb of selected feat:{}".format(str(loss_t), str(slack), str(len(nn_fi))))
348 |                     gbmo = self.test_with_mask( np.expand_dims(self.create_mask(nn_fi, self.mask_len),axis=0))
349 |         t1=time.process_time()
350 |         print(self.best_ft_nbs)
351 | 
352 |         print("gbmo time:{}".format(str(t1-t0)))   
353 |         return gbmo 
354 |     
355 |     def cv_flbmo(self, f_numbers):
356 |         best_flbmo = 1000000
357 |         t0=time.process_time()
358 |         for f_number in f_numbers:
359 |             cancelout_weights_importance = self.main_search_2(f_number = f_number, run_cv=True)
360 |             nn_fi = [i for i, x in enumerate(cancelout_weights_importance) if x == 1]
361 |             mo_score = self.val_with_mask( np.expand_dims(self.create_mask(nn_fi, self.mask_len),axis=0))
362 |             if mo_score<best_flbmo:
363 |                 self.flbmo_selected_ft_nbs=self.selected_feature_nbs_flbmo
364 |                 best_flbmo= mo_score           
365 |                 print("nb of selected feat:{}".format(str(f_number)))
366 |                 print(nn_fi)
367 |                 flbmo = self.test_with_mask( np.expand_dims(self.create_mask(nn_fi, self.mask_len),axis=0))
368 |         t1=time.process_time()
369 |         print("flbmo time:{}".format(str(t1-t0)))   
370 |         return flbmo 
371 |     def test_on_synth_data(self,  f_numbers, loss_ts, slacks,seed=42):
372 |         np.random.seed(seed)
373 |         random_search = RandomizedSearchCV(estimator=self.model, param_distributions=self.params, n_iter=20, cv=5,
374 |                                             random_state=42, verbose=0,scoring=self.cv_scoring)
375 |         # Fit the model
376 |         random_search.fit(self.X_train, self.y_train)
377 |         self.mask_len = self.X_test.shape[1]        
378 | 
379 |         # Update the best model and best parameters
380 |         best_params_opt_1 = random_search.best_params_
381 |         print("Best parameters for iteration {} are: {}".format(iter, best_params_opt_1))
382 |         self.model = random_search.best_estimator_
383 |         #nn_fi = [i for i, x in enumerate(cancelout_weights_importance) if x == 1]
384 |         mask_len = self.X_test.shape[1]        
385 |         gbmo_score = self.cv_gbmo(loss_ts, slacks)
386 |         flbmo_score = self.cv_flbmo(f_numbers)
387 |         return gbmo_score, flbmo_score
388 | 
389 | 
390 |     def feature_extraction(self, f_number, method="number of features", seed =88, include_RFE=False, loss_tolerance = 3, shuffle=False, run_CV=False, f_numbers=[10,20, 30, 40], loss_ts=[1,2,3,4], slacks=[0, 0.05, 0.1, 0.15, 0.2]):
391 |         
392 |         feature_weights = {}
393 |         # Gradient Boosting FS
394 |         np.random.seed(seed)
395 |         # if self.problem_type=="Classifier":
396 |         #     forest=lgb.LGBMClassifier(max_depth = 3, n_estimators=10,random_state=0,importance_type='gain', verbose=-1)
397 |         #     mi_importances= mutual_info_classif(self.X_train, self.y_train)
398 |         # else:
399 |         #     forest = lgb.LGBMRegressor(max_depth = 3, n_estimators=10,random_state=0,importance_type='gain', verbose=-1)
400 |         #     mi_importances= mutual_info_regression(self.X_train, self.y_train)
401 |         # forest.fit(self.X_train, self.y_train.ravel())
402 |         # gb_importances = forest.feature_importances_
403 |         self.mask_len = self.X_test.shape[1]        
404 | 
405 |         # if method == "number of features":
406 |         #     cancelout_weights_importance = self.main_search_2(f_number= f_number, shuffle=shuffle, run_cv=run_CV)
407 |         # elif method == "convergence": 
408 |         #     cancelout_weights_importance = self.main_search_1(loss_tolerance = loss_tolerance, run_cv=run_CV)
409 |         # else:
410 |         #     raise ValueError("choose a valid feature selection method.")
411 |         #mi_fi = mi_importances.argsort()[-f_number:][::-1]
412 | 
413 |         if method == "number of features":
414 |             feature_weights['nn'] = cancelout_weights_importance.argsort()[::-1]
415 |             feature_weights['gb'] = gb_importances.argsort()[::-1]
416 |             nn_fi = cancelout_weights_importance.argsort()[-f_number:][::-1]
417 |             gb_fi = gb_importances.argsort()[-f_number:][::-1]
418 |             mask_len = self.X_test.shape[1]        
419 |             all_score = self.test_with_mask( np.expand_dims(np.ones(mask_len),axis=0))
420 |             gb_score = self.test_with_mask( np.expand_dims(self.create_mask(gb_fi, mask_len),axis=0))
421 |             mi_score = self.test_with_mask( np.expand_dims(self.create_mask(mi_fi, mask_len),axis=0))
422 |             mo_score = self.test_with_mask( np.expand_dims(self.create_mask(nn_fi, mask_len),axis=0))
423 | 
424 |             if include_RFE:#&(self.model_type=="lgbm"):
425 |                 rfe = RFE(self.model, n_features_to_select=f_number) 
426 |                 rfe.fit(self.X_train, self.y_train)
427 |                 rfe_mask = 1*rfe.support_
428 |                 rfe_score = self.test_with_mask( np.expand_dims(rfe_mask,axis=0))
429 |                 return feature_weights,all_score, gb_score, mo_score, rfe_score, mi_score
430 |             
431 |             return feature_weights,all_score, gb_score, mo_score, mi_score
432 |         
433 |         else:
434 |             random_search = RandomizedSearchCV(estimator=self.model, param_distributions=self.params, n_iter=20, cv=5,
435 |                                                 random_state=42, verbose=0,scoring=self.cv_scoring)
436 |             # Fit the model
437 |             random_search.fit(self.X_train, self.y_train)
438 | 
439 |             # Update the best model and best parameters
440 |             best_params_opt_1 = random_search.best_params_
441 |             print("Best parameters for iteration {} are: {}".format(iter, best_params_opt_1))
442 |             self.model = random_search.best_estimator_
443 |             #nn_fi = [i for i, x in enumerate(cancelout_weights_importance) if x == 1]
444 |             mask_len = self.X_test.shape[1]        
445 |             gbmo_score = self.cv_gbmo(loss_ts, slacks)
446 |             flbmo_score = self.cv_flbmo(f_numbers)
447 |             cc_score, mi_score = self.cv_for_other_fs_methods(f_numbers)
448 |             all_score = self.test_with_mask( np.expand_dims(np.ones(mask_len),axis=0))
449 |             if include_RFE: #(self.model_type=="lgbm"):
450 |                 rfe_score=self.cv_for_rfe(f_numbers)
451 |                 return feature_weights,all_score, flbmo_score, gbmo_score, cc_score, rfe_score, mi_score
452 | 
453 |             return feature_weights,all_score, flbmo_score, gbmo_score, cc_score, mi_score
454 | 
455 | 
456 |     def val_with_mask(self,mask):
457 |         zero_columns = np.where(mask == 0)
458 |         masked_train_x = np.delete(self.X_train, zero_columns, axis=1)
459 |         masked_test_x = np.delete(self.X_val_2, zero_columns, axis=1)
460 |         #print(np.shape(masked_train_x))
461 |         self.model.fit(masked_train_x, self.y_train)    
462 |         if self.problem_type == "Classifier":
463 |             #loss = roc_auc_score(self.y_test ,self.model.predict_proba(masked_test_x)[:,1])
464 |             loss = log_loss(self.y_val_2 ,self.model.predict_proba(masked_test_x)[:,1])
465 |         else:  
466 |             loss = mean_squared_error(self.y_val_2  ,self.model.predict(masked_test_x))
467 |         return loss
468 | 
469 |     def test_with_mask(self,mask):
470 |         zero_columns = np.where(mask == 0)
471 |         masked_train_x = np.delete(self.X_train, zero_columns, axis=1)
472 |         masked_test_x = np.delete(self.X_test, zero_columns, axis=1)
473 |         random_search = RandomizedSearchCV(estimator=self.model, param_distributions=self.params, n_iter=20, cv=5,
474 |                                             random_state=42, verbose=0,scoring=self.cv_scoring)
475 |         # Fit the model
476 |         random_search.fit(masked_train_x, self.y_train)
477 |         #self.model.fit(masked_train_x, self.y_train)    
478 |         if self.problem_type == "Classifier":
479 |             #loss = roc_auc_score(self.y_test ,self.model.predict_proba(masked_test_x)[:,1])
480 |             loss = log_loss(self.y_test ,random_search.best_estimator_.predict_proba(masked_test_x)[:,1])
481 |         else:  
482 |             loss = mean_squared_error(self.y_test  ,random_search.best_estimator_.predict(masked_test_x))
483 |         
484 |         return loss
485 | 
486 | 
487 | 


--------------------------------------------------------------------------------
/paper_final_experiments.py:
--------------------------------------------------------------------------------
  1 | 
  2 | import os
  3 | from final_model import LGBM_w_Feature_Selector 
  4 | import logging
  5 | import warnings
  6 | from sklearn.exceptions import DataConversionWarning,ConvergenceWarning
  7 | from sklearn.preprocessing import MinMaxScaler, StandardScaler
  8 | import pandas as pd
  9 | import re
 10 | import matplotlib.pyplot as plt
 11 | 
 12 | import numpy as np
 13 | warnings.filterwarnings("ignore", category=DataConversionWarning)
 14 | warnings.filterwarnings("ignore", category=ConvergenceWarning)
 15 | logging.basicConfig(filename='final_exp.log', filemode='w', level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')
 16 | 
 17 | lgbm_param_grid={
 18 | 'num_leaves': [7, 15],
 19 | 'learning_rate': [0.01, 0.025, 0.05],
 20 | 'n_estimators': [10, 20],
 21 | 'subsample': [0.6, 0.8],
 22 | 'colsample_bytree': [0.6, 0.8],
 23 | # 'reg_alpha': [0.0, 0.1, 0.5],
 24 | # 'reg_lambda': [0.0, 0.1, 0.5],
 25 | 'min_child_samples': [5, 10],
 26 | }
 27 | # lgbm_param_grid={
 28 | # 'num_leaves': [63, 127],
 29 | # 'learning_rate': [0.01, 0.025, 0.05],
 30 | # 'n_estimators': [50, 100],
 31 | # 'subsample': [0.6, 0.8],
 32 | # 'colsample_bytree': [0.6, 0.8],
 33 | # # 'reg_alpha': [0.0, 0.1, 0.5],
 34 | # # 'reg_lambda': [0.0, 0.1, 0.5],
 35 | # #'min_child_samples': [5, 10],
 36 | # }
 37 | mlp_param_grid ={
 38 |     "hidden_layer_sizes": [(20),(40),(10),(20,10)],#[(20),(40),(80),(40,10),(80,20),(80,40),(80,20,10),(80,40,20),(80,40,10)], 
 39 |     "activation":['relu','logistic'], 
 40 |     "solver":['adam',], 
 41 |     "alpha":[0.0001, 0.001, 0.01], 
 42 |     "learning_rate_init":[0.001, 0.01], 
 43 |     "warm_start":[False],  
 44 |     "momentum":[0.9], 
 45 | } 
 46 | # mlp_param_grid ={
 47 | #     "hidden_layer_sizes": [(80),(120),(80,40),(120,40)],#[(20),(40),(80),(40,10),(80,20),(80,40),(80,20,10),(80,40,20),(80,40,10)], 
 48 | #     "activation":['relu','logistic'], 
 49 | #     "solver":['adam',], 
 50 | #     "alpha":[0.0001, 0.001, 0.01], 
 51 | #     "learning_rate_init":[0.001, 0.01], 
 52 | #     "warm_start":[False],  
 53 | #     "momentum":[0.9], 
 54 | # } 
 55 | 
 56 | # #Voice data
 57 | # all_scores = []
 58 | # mo_scores = []
 59 | # flbmo_scores = []
 60 | # lgbm_scores=[]
 61 | # mi_scores=[]
 62 | # rfe_scores=[]
 63 | 
 64 | # data = pd.read_excel('LSVT_voice_rehabilitation.xlsx', sheet_name=0)
 65 | # target = pd.read_excel('LSVT_voice_rehabilitation.xlsx', sheet_name=1)
 66 | # data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
 67 | # seed=51
 68 | # nb = len(data)
 69 | # val_1_split = 0.3
 70 | # val_2_split = 0.1
 71 | # test_split = 0.15
 72 | # data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
 73 | # X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
 74 | # y_train, y_val_1, y_val_2, y_test = target.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], target.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], target.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], target.iloc[int((1-(test_split))*nb):]
 75 | # scaler = MinMaxScaler()
 76 | # normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns).values
 77 | # normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns).values
 78 | # normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns).values
 79 | # normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns).values
 80 | # network = LGBM_w_Feature_Selector(model="lgbm",problem_type="Classsifier",param_grid=lgbm_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
 81 | #                                         X_val_1=normalized_X_val_1, y_val_1=y_val_1.values,X_val_2=normalized_X_val_2, y_val_2=y_val_2.values, y_train=y_train.values,y_test=y_test.values,iterations=10) 
 82 | 
 83 | # _,all_score,flbmo_score, mo_score, gbm_score, rfe_score, mi_score = network.feature_extraction(5, seed=42 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=True, f_numbers=[50, 60, 75, 150], loss_ts=[1], slacks=[0.00025, 0.001, 0.01, 0.05])
 84 | # all_scores.append(all_score)
 85 | # flbmo_scores.append(flbmo_score)
 86 | # mo_scores.append(mo_score)
 87 | # lgbm_scores.append(gbm_score)
 88 | # mi_scores.append(mi_score)
 89 | # rfe_scores.append(rfe_score)
 90 | # logging.info("For Voice Data lgbm base, all feature Log loss: {}, flbmo: {} GBMO {}, CC {}, MI{}, RFE: {}".format( np.mean(all_scores), np.mean(flbmo_scores), np.mean(mo_scores), np.mean(lgbm_scores), np.mean(mi_scores),np.mean(rfe_scores)))
 91 | # plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
 92 | 
 93 | # # Adding title and labels (optional)
 94 | # plt.title("GBMO Iterations Features Selected")
 95 | # plt.xlabel("Iterations")
 96 | # plt.ylabel("Number of Selected Features")
 97 | # plt.ylim([0, 310])
 98 | 
 99 | # # Save the figure
100 | # plt.savefig("voice_lgbm_gbmo_plot.png")  # You can change the filename and extension as needed
101 | # plt.figure()
102 | # plt.scatter(range(len(network.val_losses)), network.val_losses, s=10)
103 | 
104 | # ##Adding title and labels (optional)
105 | # plt.title("Validation Losses with respect to Iterations")
106 | # plt.xlabel("Iterations")
107 | # plt.ylabel("Validation Loss")
108 | 
109 | # ##Save the figure
110 | # plt.savefig("voice_lgbm_gbmo_plot_losses.png") 
111 | # # Display the plot
112 | # plt.show()
113 | # all_scores = []
114 | # flbmo_scores = []
115 | # mo_scores = []
116 | # lgbm_scores=[]
117 | # mi_scores=[]
118 | # data = pd.read_excel('LSVT_voice_rehabilitation.xlsx', sheet_name=0)
119 | # target = pd.read_excel('LSVT_voice_rehabilitation.xlsx', sheet_name=1)
120 | # data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
121 | # seed = 57
122 | # nb = len(data)
123 | # val_1_split = 0.3
124 | # val_2_split = 0.1
125 | # test_split = 0.15
126 | # data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
127 | # X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
128 | # y_train, y_val_1, y_val_2, y_test = target.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], target.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], target.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], target.iloc[int((1-(test_split))*nb):]
129 | # scaler = MinMaxScaler()
130 | # normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns).values
131 | # normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns).values
132 | # normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns).values
133 | # normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns).values
134 | # network = LGBM_w_Feature_Selector(model="mlp",problem_type="Classifier",param_grid=mlp_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
135 | #                                         X_val_1=normalized_X_val_1, y_val_1=y_val_1.values,X_val_2=normalized_X_val_2, y_val_2=y_val_2.values, y_train=y_train.values,y_test=y_test.values,iterations=10) 
136 | # _,all_score, flbmo_score, mo_score, gbm_score, mi_score = network.feature_extraction(5, seed=0 , method="convergence", loss_tolerance=1,run_CV=True, f_numbers=[50, 60, 75, 150], loss_ts=[1], slacks=[0.00025, 0.001, 0.01, 0.05])
137 | # all_scores.append(all_score)
138 | # flbmo_scores.append(flbmo_score)
139 | # mo_scores.append(mo_score)
140 | # lgbm_scores.append(gbm_score)
141 | # mi_scores.append(mi_score)
142 | # logging.info("For Voice Data, MLP base, all feature Log loss: {}, flbmo:{} GBMO: {}, CC:{}, MI:{}".format( np.mean(all_scores), np.mean(flbmo_scores), np.mean(mo_scores), np.mean(lgbm_scores), np.mean(mi_scores)))
143 | # plt.figure()
144 | # plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
145 | 
146 | # # Adding title and labels (optional)
147 | # plt.title("GBMO Iterations Features Selected")
148 | # plt.xlabel("Iterations")
149 | # plt.ylabel("Number of Selected Features")
150 | # plt.ylim([0, 310])
151 | 
152 | # # Save the figure
153 | # plt.savefig("voice_mlp_gbmo_plot.png")  # You can change the filename and extension as needed
154 | # plt.figure()
155 | # plt.scatter(range(len(network.val_losses)), network.val_losses, s=10)
156 | 
157 | # ##Adding title and labels (optional)
158 | # plt.title("Validation Losses with respect to Iterations")
159 | # plt.xlabel("Iterations")
160 | # plt.ylabel("Validation Loss")
161 | 
162 | # ##Save the figure
163 | # plt.savefig("voice_mlp_gbmo_plot_losses.png") 
164 | ## Sonar Data
165 | seed= 13 
166 | all_scores = []
167 | mo_scores = []
168 | flbmo_scores = []
169 | 
170 | lgbm_scores=[]
171 | mi_scores=[]
172 | rfe_scores=[]
173 | data = pd.read_csv("sonar.all-data.csv")
174 | data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
175 | data["Y"] = data['Label'].map({'R':1, 'M':0})
176 | data=data.drop(['Label'], axis=1)
177 | nb = len(data)
178 | val_1_split = 0.3
179 | val_2_split = 0.1
180 | test_split = 0.15
181 | data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
182 | X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
183 | y_train, y_val_1, y_val_2, y_test = X_train[["Y"]], X_val_1[["Y"]], X_val_2[["Y"]], X_test[["Y"]]
184 | scaler = MinMaxScaler()
185 | normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns)
186 | normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns)
187 | normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns)
188 | normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns)
189 | normalized_X_train, normalized_X_val_1, normalized_X_val_2,  normalized_X_test = normalized_X_train.drop(columns=['Y']).values, normalized_X_val_1.drop(columns=['Y']).values, normalized_X_val_2.drop(columns=['Y']).values, normalized_X_test.drop(columns=['Y']).values
190 | network = LGBM_w_Feature_Selector(model="lgbm",problem_type="Classifier",param_grid=lgbm_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
191 |                                         X_val_1=normalized_X_val_1, y_val_1=y_val_1.values,X_val_2=normalized_X_val_2, y_val_2=y_val_2.values, y_train=y_train.values,y_test=y_test.values, iterations=10) 
192 | _,all_score, flbmo_score, mo_score, gbm_score, rfe_score, mi_score = network.feature_extraction(5, seed=0 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=True, f_numbers=[10, 12, 15, 30], loss_ts=[1], slacks=[0.00025, 0.001, 0.01, 0.05]) #slacks=[0.005, 0.01])
193 | all_scores.append(all_score)
194 | flbmo_scores.append(flbmo_score)
195 | 
196 | mo_scores.append(mo_score)
197 | lgbm_scores.append(gbm_score)
198 | mi_scores.append(mi_score)
199 | rfe_scores.append(rfe_score)
200 | logging.info("For sonar Data, lgbm base, all feature log loss : {}, FLBMO : {}, GBMO : {},CC : {} MI :{}, RFE : {}".format( all_score,flbmo_score,mo_score,gbm_score, mi_score, rfe_score))
201 | plt.figure()
202 | 
203 | plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
204 | 
205 | # Adding title and labels (optional)
206 | plt.title("GBMO Iterations Features Selected")
207 | plt.xlabel("Iterations")
208 | plt.ylabel("Number of Selected Features")
209 | plt.ylim([0, 70])
210 | 
211 | # Save the figure
212 | plt.savefig("sonar_lgbm_gbmo_plot.png")  # You can change the filename and extension as needed
213 | plt.figure()
214 | plt.scatter(range(len(network.val_losses[1:])), network.val_losses[1:], s=10)
215 | 
216 | ##Adding title and labels (optional)
217 | plt.title("Validation Losses with respect to Iterations")
218 | plt.xlabel("Iterations")
219 | plt.ylabel("Validation Loss")
220 | 
221 | ##Save the figure
222 | plt.savefig("sonar_lgbm_gbmo_plot_losses.png") 
223 | 
224 | seed = 12
225 | all_scores = []
226 | mo_scores = []
227 | lgbm_scores=[]
228 | flbmo_scores=[]
229 | mi_scores=[]
230 | rfe_scores=[]
231 | data = pd.read_csv("sonar.all-data.csv")
232 | data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
233 | data["Y"] = data['Label'].map({'R':1, 'M':0})
234 | data=data.drop(['Label'], axis=1)
235 | #data = data.drop(['ID'], axis=1)
236 | #print(data.shape)
237 | nb = len(data)
238 | val_1_split = 0.3
239 | val_2_split = 0.1
240 | test_split = 0.15
241 | #69 nice
242 | data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
243 | X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
244 | y_train, y_val_1, y_val_2, y_test = X_train[["Y"]], X_val_1[["Y"]], X_val_2[["Y"]], X_test[["Y"]]
245 | scaler = MinMaxScaler()
246 | normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns)
247 | normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns)
248 | normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns)
249 | normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns)
250 | normalized_y_train = pd.DataFrame(scaler.fit_transform(y_train), columns=y_train.columns).values
251 | normalized_y_val_1 = pd.DataFrame(scaler.transform(y_val_1), columns=y_val_1.columns).values
252 | normalized_y_val_2 = pd.DataFrame(scaler.transform(y_val_2), columns=y_val_2.columns).values
253 | normalized_y_test = pd.DataFrame(scaler.transform(y_test), columns=y_test.columns).values
254 | normalized_X_train, normalized_X_val_1, normalized_X_val_2,  normalized_X_test = normalized_X_train.drop(columns=['Y']).values, normalized_X_val_1.drop(columns=['Y']).values, normalized_X_val_2.drop(columns=['Y']).values, normalized_X_test.drop(columns=['Y']).values
255 | network = LGBM_w_Feature_Selector(model="mlp",problem_type="classifier",param_grid=mlp_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
256 |                                         X_val_1=normalized_X_val_1, y_val_1=y_val_1.values,X_val_2=normalized_X_val_2, y_val_2=y_val_2.values, y_train=y_train.values,y_test=y_test.values, iterations=10) 
257 | _,all_score, flbmo_score,mo_score, gbm_score, mi_score = network.feature_extraction(5, seed=0 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=False,f_numbers=[10, 12, 15, 30], loss_ts=[1], slacks=[0.00025, 0.001, 0.01, 0.05]) #slacks=[0.005, 0.01])
258 | all_scores.append(all_score)
259 | flbmo_scores.append(flbmo_score)
260 | mo_scores.append(mo_score)
261 | lgbm_scores.append(gbm_score)
262 | mi_scores.append(mi_score)
263 | logging.info("For Sonar Data MLP base, all feature log loss: {}, FLBMO: {},  GBMO : {}, CC :{}, MI :{}".format(np.mean(all_scores), flbmo_score, np.mean(mo_scores), np.mean(lgbm_scores), np.mean(mi_scores)))
264 | plt.figure()
265 | 
266 | plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
267 | 
268 | # Adding title and labels (optional)
269 | plt.title("GBMO Iterations Features Selected")
270 | plt.xlabel("Iterations")
271 | plt.ylabel("Number of Selected Features")
272 | plt.ylim([0, 70])
273 | 
274 | # Save the figure
275 | plt.savefig("sonar_mlp_gbmo_plot.png")  # You can change the filename and extension as needed
276 | plt.figure()
277 | plt.scatter(range(len(network.val_losses[1:])), network.val_losses[1:], s=10)
278 | 
279 | ##Adding title and labels (optional)
280 | plt.title("Validation Losses with respect to Iterations")
281 | plt.xlabel("Iterations")
282 | plt.ylabel("Validation Loss")
283 | 
284 | ##Save the figure
285 | plt.savefig("sonar_mlp_gbmo_plot_losses.png") 
286 | 
287 | ## Residential Data
288 | all_scores = []
289 | mo_scores = []
290 | flbmo_scores = []
291 | lgbm_scores=[]
292 | mi_scores=[]
293 | rfe_scores=[]
294 | seed=29
295 | data = pd.read_excel('Residential-Building-Data-Set.xlsx', sheet_name=0)
296 | data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
297 | new_header = data.iloc[0] #grab the first row for the header
298 | data = data[1:] #take the data less the header row
299 | data.columns = new_header #
300 | data = data.drop(['V-1', 'V-10'], axis=1)
301 | print(data.shape)
302 | nb = len(data)
303 | val_1_split = 0.3
304 | val_2_split = 0.1
305 | test_split = 0.15
306 | data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
307 | X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
308 | y_train, y_val_1, y_val_2, y_test = X_train[["V-9"]], X_val_1[["V-9"]], X_val_2[["V-9"]], X_test[["V-9"]]
309 | scaler = MinMaxScaler()
310 | normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns)
311 | normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns)
312 | normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns)
313 | normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns)
314 | normalized_y_train = pd.DataFrame(scaler.fit_transform(y_train), columns=y_train.columns).values
315 | normalized_y_val_1 = pd.DataFrame(scaler.transform(y_val_1), columns=y_val_1.columns).values
316 | normalized_y_val_2 = pd.DataFrame(scaler.transform(y_val_2), columns=y_val_2.columns).values
317 | normalized_y_test = pd.DataFrame(scaler.transform(y_test), columns=y_test.columns).values
318 | normalized_X_train, normalized_X_val_1, normalized_X_val_2,  normalized_X_test = normalized_X_train.drop(columns=['V-9']).values, normalized_X_val_1.drop(columns=['V-9']).values, normalized_X_val_2.drop(columns=['V-9']).values, normalized_X_test.drop(columns=['V-9']).values
319 | network = LGBM_w_Feature_Selector(model="lgbm",problem_type="regression",param_grid=lgbm_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
320 |                                         X_val_1=normalized_X_val_1, y_val_1=normalized_y_val_1,X_val_2=normalized_X_val_2, y_val_2=normalized_y_val_2, y_train=normalized_y_train,y_test=normalized_y_test,iterations=10) 
321 | 
322 | _,all_score, flbmo_score,mo_score, gbm_score, rfe_score, mi_score = network.feature_extraction(5, seed=42 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=True, f_numbers=[20, 24, 30, 60], loss_ts=[1], slacks=[0.00025, 0.001,0.01, 0.05])
323 | all_scores.append(all_score)
324 | flbmo_scores.append(flbmo_score)
325 | mo_scores.append(mo_score)
326 | lgbm_scores.append(gbm_score)
327 | mi_scores.append(mi_score)
328 | rfe_scores.append(rfe_score)
329 | logging.info("For residential Data, seed: {} lgbm base, all feature MSE: {}, FLBMO MSE: {}, GBMO MSE: {}, GBM MSE:{}, MI MSE:{}, RFE MSE: {}".format(seed, np.mean(all_scores), np.mean(flbmo_scores), np.mean(mo_scores), np.mean(lgbm_scores), np.mean(mi_scores),np.mean(rfe_scores)))
330 | plt.figure()
331 | 
332 | plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
333 | 
334 | ###Adding title and labels (optional)
335 | plt.title("GBMO Iterations Features Selected")
336 | plt.xlabel("Iterations")
337 | plt.ylabel("Number of Selected Features")
338 | plt.ylim([0, 120])
339 | 
340 | ###Save the figure
341 | plt.savefig("res_lgbm_gbmo_plot.png")  # You can change the filename and extension as needed
342 | all_scores = []
343 | mo_scores = []
344 | lgbm_scores=[]
345 | mi_scores=[]
346 | 
347 | plt.figure()
348 | plt.scatter(range(len(network.val_losses[1:])), network.val_losses[1:], s=10)
349 | 
350 | ##Adding title and labels (optional)
351 | plt.title("Validation Losses with respect to Iterations")
352 | plt.xlabel("Iterations")
353 | plt.ylabel("Validation Loss")
354 | 
355 | ##Save the figure
356 | plt.savefig("res_lgbm_gbmo_plot_losses.png") 
357 | 
358 | seed= 58
359 | data = pd.read_excel('Residential-Building-Data-Set.xlsx', sheet_name=0)
360 | data = data.rename(columns = lambda x:re.sub('[^A-Za-z0-9_]+', '', x))
361 | new_header = data.iloc[0] #grab the first row for the header
362 | data = data[1:] #take the data less the header row
363 | data.columns = new_header #
364 | data = data.drop(['V-1', 'V-10'], axis=1)
365 | #print(data.shape)
366 | nb = len(data)
367 | val_1_split = 0.3
368 | val_2_split = 0.1
369 | test_split = 0.15
370 | data = data.sample(frac=1, random_state=seed).reset_index(drop=True)
371 | X_train, X_val_1, X_val_2, X_test = data.iloc[:int((1-(val_1_split+val_2_split+test_split))*nb)], data.iloc[int((1-(val_1_split+val_2_split+test_split))*nb):int((1-(val_2_split+test_split))*nb)], data.iloc[int((1-(val_2_split+test_split))*nb):int((1-(test_split))*nb)], data.iloc[int((1-(test_split))*nb):]
372 | y_train, y_val_1, y_val_2, y_test = X_train[["V-9"]], X_val_1[["V-9"]], X_val_2[["V-9"]], X_test[["V-9"]]
373 | scaler = MinMaxScaler()
374 | normalized_X_train = pd.DataFrame(scaler.fit_transform(X_train), columns=X_train.columns)
375 | normalized_X_val_1 = pd.DataFrame(scaler.transform(X_val_1), columns=X_val_1.columns)
376 | normalized_X_val_2 = pd.DataFrame(scaler.transform(X_val_2), columns=X_val_2.columns)
377 | normalized_X_test = pd.DataFrame(scaler.transform(X_test), columns=X_test.columns)
378 | normalized_y_train = pd.DataFrame(scaler.fit_transform(y_train), columns=y_train.columns).values
379 | normalized_y_val_1 = pd.DataFrame(scaler.transform(y_val_1), columns=y_val_1.columns).values
380 | normalized_y_val_2 = pd.DataFrame(scaler.transform(y_val_2), columns=y_val_2.columns).values
381 | normalized_y_test = pd.DataFrame(scaler.transform(y_test), columns=y_test.columns).values
382 | normalized_X_train, normalized_X_val_1, normalized_X_val_2,  normalized_X_test = normalized_X_train.drop(columns=['V-9']).values, normalized_X_val_1.drop(columns=['V-9']).values, normalized_X_val_2.drop(columns=['V-9']).values, normalized_X_test.drop(columns=['V-9']).values
383 | network = LGBM_w_Feature_Selector(model="mlp",problem_type="regression",param_grid=mlp_param_grid,X_train=normalized_X_train, X_test=normalized_X_test, slack=0.0, 
384 |                                         X_val_1=normalized_X_val_1, y_val_1=normalized_y_val_1,X_val_2=normalized_X_val_2, y_val_2=normalized_y_val_2, y_train=normalized_y_train,y_test=normalized_y_test,iterations=10) 
385 | ## 13 fena değil
386 | # 53te hesp çok iyi
387 | # 17 iyi
388 | _,all_score, flbmo_score, mo_score, gbm_score, mi_score = network.feature_extraction(5, seed=42 , method="convergence", loss_tolerance=1,run_CV=True, f_numbers=[20, 24, 30, 60], loss_ts=[1], slacks=[0.00025, 0.001, 0.01, 0.05])
389 | all_scores.append(all_score)
390 | mo_scores.append(mo_score)
391 | flbmo_scores.append(flbmo_score)
392 | lgbm_scores.append(gbm_score)
393 | mi_scores.append(mi_score)
394 | logging.info("For residential Data, mlp base, seed: {}, all feature MSE: {}, FLBMO MSE: {}, GBMO MSE: {}, GBM MSE:{}, MI MSE:{}".format(seed, np.mean(all_scores), np.mean(flbmo_scores),np.mean(mo_scores), np.mean(lgbm_scores), np.mean(mi_scores)))
395 | plt.figure()
396 | 
397 | plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
398 | 
399 | # Adding title and labels (optional)
400 | plt.title("GBMO Iterations Features Selected")
401 | plt.xlabel("Iterations")
402 | plt.ylabel("Number of Selected Features")
403 | plt.ylim([0, 120])
404 | 
405 | # Save the figure
406 | plt.savefig("res_mlp_gbmo_plot.png")  # You can change the filename and extension as needed
407 | plt.figure()
408 | plt.scatter(range(len(network.val_losses[1:])), network.val_losses[1:], s=10)
409 | 
410 | ##Adding title and labels (optional)
411 | plt.title("Validation Losses with respect to Iterations")
412 | plt.xlabel("Iterations")
413 | plt.ylabel("Validation Loss")
414 | 
415 | ##Save the figure
416 | plt.savefig("res_mlp_gbmo_plot_losses.png") 
417 | 


--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
  1 | absl-py==1.4.0
  2 | appnope==0.1.3
  3 | asttokens==2.2.1
  4 | astunparse==1.6.3
  5 | backcall==0.2.0
  6 | cachetools==5.3.1
  7 | certifi==2023.5.7
  8 | charset-normalizer==3.1.0
  9 | comm==0.1.3
 10 | contourpy==1.2.0
 11 | cycler==0.12.1
 12 | debugpy==1.6.7
 13 | decorator==5.1.1
 14 | distlib==0.3.7
 15 | et-xmlfile==1.1.0
 16 | executing==1.2.0
 17 | filelock==3.13.1
 18 | flatbuffers==23.5.26
 19 | fonttools==4.47.2
 20 | fsspec==2023.12.2
 21 | gast==0.4.0
 22 | google-auth==2.19.0
 23 | google-auth-oauthlib==1.0.0
 24 | google-pasta==0.2.0
 25 | graphviz==0.20.1
 26 | grpcio==1.54.2
 27 | h5py==3.8.0
 28 | idna==3.4
 29 | importlib-metadata==6.6.0
 30 | importlib-resources==6.1.1
 31 | ipykernel==6.23.1
 32 | ipython==8.13.2
 33 | jedi==0.18.2
 34 | Jinja2==3.1.3
 35 | joblib==1.2.0
 36 | jupyter_client==8.2.0
 37 | jupyter_core==5.3.0
 38 | keras==2.13.1rc0
 39 | kiwisolver==1.4.4
 40 | libclang==16.0.0
 41 | lightgbm==3.3.5
 42 | Markdown==3.4.3
 43 | MarkupSafe==2.1.5
 44 | matplotlib==3.8.2
 45 | matplotlib-inline==0.1.6
 46 | mpmath==1.3.0
 47 | nest-asyncio==1.5.6
 48 | networkx==3.2.1
 49 | numpy==1.26.3
 50 | oauthlib==3.2.2
 51 | openpyxl==3.1.2
 52 | opt-einsum==3.3.0
 53 | packaging==23.1
 54 | pandas==2.2.0
 55 | parso==0.8.3
 56 | patsy==0.5.6
 57 | pexpect==4.8.0
 58 | pickleshare==0.7.5
 59 | pillow==10.2.0
 60 | platformdirs==4.1.0
 61 | plotly==5.18.0
 62 | prompt-toolkit==3.0.38
 63 | protobuf==4.23.2
 64 | psutil==5.9.5
 65 | ptyprocess==0.7.0
 66 | pure-eval==0.2.2
 67 | pyasn1==0.5.0
 68 | pyasn1-modules==0.3.0
 69 | Pygments==2.15.1
 70 | pyparsing==3.1.1
 71 | python-dateutil==2.8.2
 72 | pytz==2023.4
 73 | pyzmq==25.1.0
 74 | requests==2.31.0
 75 | requests-oauthlib==1.3.1
 76 | rsa==4.9
 77 | scikit-learn==1.4.0
 78 | scipy==1.10.1
 79 | stack-data==0.6.2
 80 | statsmodels==0.14.1
 81 | sympy==1.12
 82 | tenacity==8.2.3
 83 | tensorboard==2.13.0
 84 | tensorboard-data-server==0.7.0
 85 | tensorflow==2.13.0rc0
 86 | tensorflow-addons==0.20.0
 87 | tensorflow-estimator==2.13.0rc0
 88 | tensorflow-macos==2.13.0rc0
 89 | termcolor==2.3.0
 90 | threadpoolctl==3.2.0
 91 | torch==2.2.0
 92 | tornado==6.3.2
 93 | traitlets==5.9.0
 94 | typeguard==2.13.3
 95 | typing_extensions==4.10.0
 96 | tzdata==2023.4
 97 | urllib3==1.26.16
 98 | virtualenv==20.25.0
 99 | wcwidth==0.2.6
100 | Werkzeug==2.3.4
101 | wrapt==1.14.1
102 | xgboost==2.0.3
103 | zipp==3.15.0
104 | 


--------------------------------------------------------------------------------
/sonar.all-data.csv:
--------------------------------------------------------------------------------
  1 | Freq_1,Freq_2,Freq_3,Freq_4,Freq_5,Freq_6,Freq_7,Freq_8,Freq_9,Freq_10,Freq_11,Freq_12,Freq_13,Freq_14,Freq_15,Freq_16,Freq_17,Freq_18,Freq_19,Freq_20,Freq_21,Freq_22,Freq_23,Freq_24,Freq_25,Freq_26,Freq_27,Freq_28,Freq_29,Freq_30,Freq_31,Freq_32,Freq_33,Freq_34,Freq_35,Freq_36,Freq_37,Freq_38,Freq_39,Freq_40,Freq_41,Freq_42,Freq_43,Freq_44,Freq_45,Freq_46,Freq_47,Freq_48,Freq_49,Freq_50,Freq_51,Freq_52,Freq_53,Freq_54,Freq_55,Freq_56,Freq_57,Freq_58,Freq_59,Freq_60,Label
  2 | 0.02,0.0371,0.0428,0.0207,0.0954,0.0986,0.1539,0.1601,0.3109,0.2111,0.1609,0.1582,0.2238,0.0645,0.066,0.2273,0.31,0.2999,0.5078,0.4797,0.5783,0.5071,0.4328,0.555,0.6711,0.6415,0.7104,0.808,0.6791,0.3857,0.1307,0.2604,0.5121,0.7547,0.8537,0.8507,0.6692,0.6097,0.4943,0.2744,0.051,0.2834,0.2825,0.4256,0.2641,0.1386,0.1051,0.1343,0.0383,0.0324,0.0232,0.0027,0.0065,0.0159,0.0072,0.0167,0.018,0.0084,0.009,0.0032,R
  3 | 0.0453,0.0523,0.0843,0.0689,0.1183,0.2583,0.2156,0.3481,0.3337,0.2872,0.4918,0.6552,0.6919,0.7797,0.7464,0.9444,1.0,0.8874,0.8024,0.7818,0.5212,0.4052,0.3957,0.3914,0.325,0.32,0.3271,0.2767,0.4423,0.2028,0.3788,0.2947,0.1984,0.2341,0.1306,0.4182,0.3835,0.1057,0.184,0.197,0.1674,0.0583,0.1401,0.1628,0.0621,0.0203,0.053,0.0742,0.0409,0.0061,0.0125,0.0084,0.0089,0.0048,0.0094,0.0191,0.014,0.0049,0.0052,0.0044,R
  4 | 0.0262,0.0582,0.1099,0.1083,0.0974,0.228,0.2431,0.3771,0.5598,0.6194,0.6333,0.706,0.5544,0.532,0.6479,0.6931,0.6759,0.7551,0.8929,0.8619,0.7974,0.6737,0.4293,0.3648,0.5331,0.2413,0.507,0.8533,0.6036,0.8514,0.8512,0.5045,0.1862,0.2709,0.4232,0.3043,0.6116,0.6756,0.5375,0.4719,0.4647,0.2587,0.2129,0.2222,0.2111,0.0176,0.1348,0.0744,0.013,0.0106,0.0033,0.0232,0.0166,0.0095,0.018,0.0244,0.0316,0.0164,0.0095,0.0078,R
  5 | 0.01,0.0171,0.0623,0.0205,0.0205,0.0368,0.1098,0.1276,0.0598,0.1264,0.0881,0.1992,0.0184,0.2261,0.1729,0.2131,0.0693,0.2281,0.406,0.3973,0.2741,0.369,0.5556,0.4846,0.314,0.5334,0.5256,0.252,0.209,0.3559,0.626,0.734,0.612,0.3497,0.3953,0.3012,0.5408,0.8814,0.9857,0.9167,0.6121,0.5006,0.321,0.3202,0.4295,0.3654,0.2655,0.1576,0.0681,0.0294,0.0241,0.0121,0.0036,0.015,0.0085,0.0073,0.005,0.0044,0.004,0.0117,R
  6 | 0.0762,0.0666,0.0481,0.0394,0.059,0.0649,0.1209,0.2467,0.3564,0.4459,0.4152,0.3952,0.4256,0.4135,0.4528,0.5326,0.7306,0.6193,0.2032,0.4636,0.4148,0.4292,0.573,0.5399,0.3161,0.2285,0.6995,1.0,0.7262,0.4724,0.5103,0.5459,0.2881,0.0981,0.1951,0.4181,0.4604,0.3217,0.2828,0.243,0.1979,0.2444,0.1847,0.0841,0.0692,0.0528,0.0357,0.0085,0.023,0.0046,0.0156,0.0031,0.0054,0.0105,0.011,0.0015,0.0072,0.0048,0.0107,0.0094,R
  7 | 0.0286,0.0453,0.0277,0.0174,0.0384,0.099,0.1201,0.1833,0.2105,0.3039,0.2988,0.425,0.6343,0.8198,1.0,0.9988,0.9508,0.9025,0.7234,0.5122,0.2074,0.3985,0.589,0.2872,0.2043,0.5782,0.5389,0.375,0.3411,0.5067,0.558,0.4778,0.3299,0.2198,0.1407,0.2856,0.3807,0.4158,0.4054,0.3296,0.2707,0.265,0.0723,0.1238,0.1192,0.1089,0.0623,0.0494,0.0264,0.0081,0.0104,0.0045,0.0014,0.0038,0.0013,0.0089,0.0057,0.0027,0.0051,0.0062,R
  8 | 0.0317,0.0956,0.1321,0.1408,0.1674,0.171,0.0731,0.1401,0.2083,0.3513,0.1786,0.0658,0.0513,0.3752,0.5419,0.544,0.515,0.4262,0.2024,0.4233,0.7723,0.9735,0.939,0.5559,0.5268,0.6826,0.5713,0.5429,0.2177,0.2149,0.5811,0.6323,0.2965,0.1873,0.2969,0.5163,0.6153,0.4283,0.5479,0.6133,0.5017,0.2377,0.1957,0.1749,0.1304,0.0597,0.1124,0.1047,0.0507,0.0159,0.0195,0.0201,0.0248,0.0131,0.007,0.0138,0.0092,0.0143,0.0036,0.0103,R
  9 | 0.0519,0.0548,0.0842,0.0319,0.1158,0.0922,0.1027,0.0613,0.1465,0.2838,0.2802,0.3086,0.2657,0.3801,0.5626,0.4376,0.2617,0.1199,0.6676,0.9402,0.7832,0.5352,0.6809,0.9174,0.7613,0.822,0.8872,0.6091,0.2967,0.1103,0.1318,0.0624,0.099,0.4006,0.3666,0.105,0.1915,0.393,0.4288,0.2546,0.1151,0.2196,0.1879,0.1437,0.2146,0.236,0.1125,0.0254,0.0285,0.0178,0.0052,0.0081,0.012,0.0045,0.0121,0.0097,0.0085,0.0047,0.0048,0.0053,R
 10 | 0.0223,0.0375,0.0484,0.0475,0.0647,0.0591,0.0753,0.0098,0.0684,0.1487,0.1156,0.1654,0.3833,0.3598,0.1713,0.1136,0.0349,0.3796,0.7401,0.9925,0.9802,0.889,0.6712,0.4286,0.3374,0.7366,0.9611,0.7353,0.4856,0.1594,0.3007,0.4096,0.317,0.3305,0.3408,0.2186,0.2463,0.2726,0.168,0.2792,0.2558,0.174,0.2121,0.1099,0.0985,0.1271,0.1459,0.1164,0.0777,0.0439,0.0061,0.0145,0.0128,0.0145,0.0058,0.0049,0.0065,0.0093,0.0059,0.0022,R
 11 | 0.0164,0.0173,0.0347,0.007,0.0187,0.0671,0.1056,0.0697,0.0962,0.0251,0.0801,0.1056,0.1266,0.089,0.0198,0.1133,0.2826,0.3234,0.3238,0.4333,0.6068,0.7652,0.9203,0.9719,0.9207,0.7545,0.8289,0.8907,0.7309,0.6896,0.5829,0.4935,0.3101,0.0306,0.0244,0.1108,0.1594,0.1371,0.0696,0.0452,0.062,0.1421,0.1597,0.1384,0.0372,0.0688,0.0867,0.0513,0.0092,0.0198,0.0118,0.009,0.0223,0.0179,0.0084,0.0068,0.0032,0.0035,0.0056,0.004,R
 12 | 0.0039,0.0063,0.0152,0.0336,0.031,0.0284,0.0396,0.0272,0.0323,0.0452,0.0492,0.0996,0.1424,0.1194,0.0628,0.0907,0.1177,0.1429,0.1223,0.1104,0.1847,0.3715,0.4382,0.5707,0.6654,0.7476,0.7654,0.8555,0.972,0.9221,0.7502,0.7209,0.7757,0.6055,0.5021,0.4499,0.3947,0.4281,0.4427,0.3749,0.1972,0.0511,0.0793,0.1269,0.1533,0.069,0.0402,0.0534,0.0228,0.0073,0.0062,0.0062,0.012,0.0052,0.0056,0.0093,0.0042,0.0003,0.0053,0.0036,R
 13 | 0.0123,0.0309,0.0169,0.0313,0.0358,0.0102,0.0182,0.0579,0.1122,0.0835,0.0548,0.0847,0.2026,0.2557,0.187,0.2032,0.1463,0.2849,0.5824,0.7728,0.7852,0.8515,0.5312,0.3653,0.5973,0.8275,1.0,0.8673,0.6301,0.4591,0.394,0.2576,0.2817,0.2641,0.2757,0.2698,0.3994,0.4576,0.394,0.2522,0.1782,0.1354,0.0516,0.0337,0.0894,0.0861,0.0872,0.0445,0.0134,0.0217,0.0188,0.0133,0.0265,0.0224,0.0074,0.0118,0.0026,0.0092,0.0009,0.0044,R
 14 | 0.0079,0.0086,0.0055,0.025,0.0344,0.0546,0.0528,0.0958,0.1009,0.124,0.1097,0.1215,0.1874,0.3383,0.3227,0.2723,0.3943,0.6432,0.7271,0.8673,0.9674,0.9847,0.948,0.8036,0.6833,0.5136,0.309,0.0832,0.4019,0.2344,0.1905,0.1235,0.1717,0.2351,0.2489,0.3649,0.3382,0.1589,0.0989,0.1089,0.1043,0.0839,0.1391,0.0819,0.0678,0.0663,0.1202,0.0692,0.0152,0.0266,0.0174,0.0176,0.0127,0.0088,0.0098,0.0019,0.0059,0.0058,0.0059,0.0032,R
 15 | 0.009,0.0062,0.0253,0.0489,0.1197,0.1589,0.1392,0.0987,0.0955,0.1895,0.1896,0.2547,0.4073,0.2988,0.2901,0.5326,0.4022,0.1571,0.3024,0.3907,0.3542,0.4438,0.6414,0.4601,0.6009,0.869,0.8345,0.7669,0.5081,0.462,0.538,0.5375,0.3844,0.3601,0.7402,0.7761,0.3858,0.0667,0.3684,0.6114,0.351,0.2312,0.2195,0.3051,0.1937,0.157,0.0479,0.0538,0.0146,0.0068,0.0187,0.0059,0.0095,0.0194,0.008,0.0152,0.0158,0.0053,0.0189,0.0102,R
 16 | 0.0124,0.0433,0.0604,0.0449,0.0597,0.0355,0.0531,0.0343,0.1052,0.212,0.164,0.1901,0.3026,0.2019,0.0592,0.239,0.3657,0.3809,0.5929,0.6299,0.5801,0.4574,0.4449,0.3691,0.6446,0.894,0.8978,0.498,0.3333,0.235,0.1553,0.3666,0.434,0.3082,0.3024,0.4109,0.5501,0.4129,0.5499,0.5018,0.3132,0.2802,0.2351,0.2298,0.1155,0.0724,0.0621,0.0318,0.045,0.0167,0.0078,0.0083,0.0057,0.0174,0.0188,0.0054,0.0114,0.0196,0.0147,0.0062,R
 17 | 0.0298,0.0615,0.065,0.0921,0.1615,0.2294,0.2176,0.2033,0.1459,0.0852,0.2476,0.3645,0.2777,0.2826,0.3237,0.4335,0.5638,0.4555,0.4348,0.6433,0.3932,0.1989,0.354,0.9165,0.9371,0.462,0.2771,0.6613,0.8028,0.42,0.5192,0.6962,0.5792,0.8889,0.7863,0.7133,0.7615,0.4401,0.3009,0.3163,0.2809,0.2898,0.0526,0.1867,0.1553,0.1633,0.1252,0.0748,0.0452,0.0064,0.0154,0.0031,0.0153,0.0071,0.0212,0.0076,0.0152,0.0049,0.02,0.0073,R
 18 | 0.0352,0.0116,0.0191,0.0469,0.0737,0.1185,0.1683,0.1541,0.1466,0.2912,0.2328,0.2237,0.247,0.156,0.3491,0.3308,0.2299,0.2203,0.2493,0.4128,0.3158,0.6191,0.5854,0.3395,0.2561,0.5599,0.8145,0.6941,0.6985,0.866,0.593,0.3664,0.675,0.8697,0.7837,0.7552,0.5789,0.4713,0.1252,0.6087,0.7322,0.5977,0.3431,0.1803,0.2378,0.3424,0.2303,0.0689,0.0216,0.0469,0.0426,0.0346,0.0158,0.0154,0.0109,0.0048,0.0095,0.0015,0.0073,0.0067,R
 19 | 0.0192,0.0607,0.0378,0.0774,0.1388,0.0809,0.0568,0.0219,0.1037,0.1186,0.1237,0.1601,0.352,0.4479,0.3769,0.5761,0.6426,0.679,0.7157,0.5466,0.5399,0.6362,0.7849,0.7756,0.578,0.4862,0.4181,0.2457,0.0716,0.0613,0.1816,0.4493,0.5976,0.3785,0.2495,0.5771,0.8852,0.8409,0.357,0.3133,0.6096,0.6378,0.2709,0.1419,0.126,0.1288,0.079,0.0829,0.052,0.0216,0.036,0.0331,0.0131,0.012,0.0108,0.0024,0.0045,0.0037,0.0112,0.0075,R
 20 | 0.027,0.0092,0.0145,0.0278,0.0412,0.0757,0.1026,0.1138,0.0794,0.152,0.1675,0.137,0.1361,0.1345,0.2144,0.5354,0.683,0.56,0.3093,0.3226,0.443,0.5573,0.5782,0.6173,0.8132,0.9819,0.9823,0.9166,0.7423,0.7736,0.8473,0.7352,0.6671,0.6083,0.6239,0.5972,0.5715,0.5242,0.2924,0.1536,0.2003,0.2031,0.2207,0.1778,0.1353,0.1373,0.0749,0.0472,0.0325,0.0179,0.0045,0.0084,0.001,0.0018,0.0068,0.0039,0.012,0.0132,0.007,0.0088,R
 21 | 0.0126,0.0149,0.0641,0.1732,0.2565,0.2559,0.2947,0.411,0.4983,0.592,0.5832,0.5419,0.5472,0.5314,0.4981,0.6985,0.8292,0.7839,0.8215,0.9363,1.0,0.9224,0.7839,0.547,0.4562,0.5922,0.5448,0.3971,0.0882,0.2385,0.2005,0.0587,0.2544,0.2009,0.0329,0.1547,0.1212,0.2446,0.3171,0.3195,0.3051,0.0836,0.1266,0.1381,0.1136,0.0516,0.0073,0.0278,0.0372,0.0121,0.0153,0.0092,0.0035,0.0098,0.0121,0.0006,0.0181,0.0094,0.0116,0.0063,R
 22 | 0.0473,0.0509,0.0819,0.1252,0.1783,0.307,0.3008,0.2362,0.383,0.3759,0.3021,0.2909,0.2301,0.1411,0.1582,0.243,0.4474,0.5964,0.6744,0.7969,0.8319,0.7813,0.8626,0.7369,0.4122,0.2596,0.3392,0.3788,0.4488,0.6281,0.7449,0.7328,0.7704,0.787,0.6048,0.586,0.6385,0.7279,0.6286,0.5316,0.4069,0.1791,0.1625,0.2527,0.1903,0.1643,0.0604,0.0209,0.0436,0.0175,0.0107,0.0193,0.0118,0.0064,0.0042,0.0054,0.0049,0.0082,0.0028,0.0027,R
 23 | 0.0664,0.0575,0.0842,0.0372,0.0458,0.0771,0.0771,0.113,0.2353,0.1838,0.2869,0.4129,0.3647,0.1984,0.284,0.4039,0.5837,0.6792,0.6086,0.4858,0.3246,0.2013,0.2082,0.1686,0.2484,0.2736,0.2984,0.4655,0.699,0.7474,0.7956,0.7981,0.6715,0.6942,0.744,0.8169,0.8912,1.0,0.8753,0.7061,0.6803,0.5898,0.4618,0.3639,0.1492,0.1216,0.1306,0.1198,0.0578,0.0235,0.0135,0.0141,0.019,0.0043,0.0036,0.0026,0.0024,0.0162,0.0109,0.0079,R
 24 | 0.0099,0.0484,0.0299,0.0297,0.0652,0.1077,0.2363,0.2385,0.0075,0.1882,0.1456,0.1892,0.3176,0.134,0.2169,0.2458,0.2589,0.2786,0.2298,0.0656,0.1441,0.1179,0.1668,0.1783,0.2476,0.257,0.1036,0.5356,0.7124,0.6291,0.4756,0.6015,0.7208,0.6234,0.5725,0.7523,0.8712,0.9252,0.9709,0.9297,0.8995,0.7911,0.56,0.2838,0.4407,0.5507,0.4331,0.2905,0.1981,0.0779,0.0396,0.0173,0.0149,0.0115,0.0202,0.0139,0.0029,0.016,0.0106,0.0134,R
 25 | 0.0115,0.015,0.0136,0.0076,0.0211,0.1058,0.1023,0.044,0.0931,0.0734,0.074,0.0622,0.1055,0.1183,0.1721,0.2584,0.3232,0.3817,0.4243,0.4217,0.4449,0.4075,0.3306,0.4012,0.4466,0.5218,0.7552,0.9503,1.0,0.9084,0.8283,0.7571,0.7262,0.6152,0.568,0.5757,0.5324,0.3672,0.1669,0.0866,0.0646,0.1891,0.2683,0.2887,0.2341,0.1668,0.1015,0.1195,0.0704,0.0167,0.0107,0.0091,0.0016,0.0084,0.0064,0.0026,0.0029,0.0037,0.007,0.0041,R
 26 | 0.0293,0.0644,0.039,0.0173,0.0476,0.0816,0.0993,0.0315,0.0736,0.086,0.0414,0.0472,0.0835,0.0938,0.1466,0.0809,0.1179,0.2179,0.3326,0.3258,0.2111,0.2302,0.3361,0.4259,0.4609,0.2606,0.0874,0.2862,0.5606,0.8344,0.8096,0.725,0.8048,0.9435,1.0,0.896,0.5516,0.3037,0.2338,0.2382,0.3318,0.3821,0.1575,0.2228,0.1582,0.1433,0.1634,0.1133,0.0567,0.0133,0.017,0.0035,0.0052,0.0083,0.0078,0.0075,0.0105,0.016,0.0095,0.0011,R
 27 | 0.0201,0.0026,0.0138,0.0062,0.0133,0.0151,0.0541,0.021,0.0505,0.1097,0.0841,0.0942,0.1204,0.042,0.0031,0.0162,0.0624,0.2127,0.3436,0.3813,0.3825,0.4764,0.6313,0.7523,0.8675,0.8788,0.7901,0.8357,0.9631,0.9619,0.9236,0.8903,0.9708,0.9647,0.7892,0.5307,0.2718,0.1953,0.1374,0.3105,0.379,0.4105,0.3355,0.2998,0.2748,0.2024,0.1043,0.0453,0.0337,0.0122,0.0072,0.0108,0.007,0.0063,0.003,0.0011,0.0007,0.0024,0.0057,0.0044,R
 28 | 0.0151,0.032,0.0599,0.105,0.1163,0.1734,0.1679,0.1119,0.0889,0.1205,0.0847,0.1518,0.2305,0.2793,0.3404,0.4527,0.695,0.8807,0.9154,0.7542,0.6736,0.7146,0.8335,0.7701,0.6993,0.6543,0.504,0.4926,0.4992,0.4161,0.1631,0.0404,0.0637,0.2962,0.3609,0.1866,0.0476,0.1497,0.2405,0.198,0.3175,0.2379,0.1716,0.1559,0.1556,0.0422,0.0493,0.0476,0.0219,0.0059,0.0086,0.0061,0.0015,0.0084,0.0128,0.0054,0.0011,0.0019,0.0023,0.0062,R
 29 | 0.0177,0.03,0.0288,0.0394,0.063,0.0526,0.0688,0.0633,0.0624,0.0613,0.168,0.3476,0.4561,0.5188,0.6308,0.7201,0.5153,0.3818,0.2644,0.3345,0.4865,0.6628,0.7389,0.9213,1.0,0.775,0.5593,0.6172,0.8635,0.6592,0.477,0.4983,0.333,0.3076,0.2876,0.2226,0.0794,0.0603,0.1049,0.0606,0.153,0.0983,0.1643,0.1901,0.1107,0.1917,0.1467,0.0392,0.0356,0.027,0.0168,0.0102,0.0122,0.0044,0.0075,0.0124,0.0099,0.0057,0.0032,0.0019,R
 30 | 0.01,0.0275,0.019,0.0371,0.0416,0.0201,0.0314,0.0651,0.1896,0.2668,0.3376,0.3282,0.2432,0.1268,0.1278,0.4441,0.6795,0.7051,0.7966,0.9401,0.9857,0.8193,0.5789,0.6394,0.7043,0.6875,0.4081,0.1811,0.2064,0.3917,0.3791,0.2042,0.2227,0.3341,0.3984,0.5077,0.5534,0.3352,0.2723,0.2278,0.2044,0.1986,0.0835,0.0908,0.138,0.1948,0.1211,0.0843,0.0589,0.0247,0.0118,0.0088,0.0104,0.0036,0.0088,0.0047,0.0117,0.002,0.0091,0.0058,R
 31 | 0.0189,0.0308,0.0197,0.0622,0.008,0.0789,0.144,0.1451,0.1789,0.2522,0.2607,0.371,0.3906,0.2672,0.2716,0.4183,0.6988,0.5733,0.2226,0.2631,0.7473,0.7263,0.3393,0.2824,0.6053,0.5897,0.4967,0.8616,0.8339,0.4084,0.2268,0.1745,0.0507,0.1588,0.304,0.1369,0.1605,0.2061,0.0734,0.0202,0.1638,0.1583,0.183,0.1886,0.1008,0.0663,0.0183,0.0404,0.0108,0.0143,0.0091,0.0038,0.0096,0.0142,0.019,0.014,0.0099,0.0092,0.0052,0.0075,R
 32 | 0.024,0.0218,0.0324,0.0569,0.033,0.0513,0.0897,0.0713,0.0569,0.0389,0.1934,0.2434,0.2906,0.2606,0.3811,0.4997,0.3015,0.3655,0.6791,0.7307,0.5053,0.4441,0.6987,0.8133,0.7781,0.8943,0.8929,0.8913,0.861,0.8063,0.554,0.2446,0.3459,0.1615,0.2467,0.5564,0.4681,0.0979,0.1582,0.0751,0.3321,0.3745,0.2666,0.1078,0.1418,0.1687,0.0738,0.0634,0.0144,0.0226,0.0061,0.0162,0.0146,0.0093,0.0112,0.0094,0.0054,0.0019,0.0066,0.0023,R
 33 | 0.0084,0.0153,0.0291,0.0432,0.0951,0.0752,0.0414,0.0259,0.0692,0.1753,0.197,0.1167,0.1683,0.0814,0.2179,0.5121,0.7231,0.7776,0.6222,0.3501,0.3733,0.2622,0.3776,0.7361,0.8673,0.8223,0.7772,0.7862,0.5652,0.3635,0.3534,0.3865,0.337,0.1693,0.2627,0.3195,0.1388,0.1048,0.1681,0.191,0.1174,0.0933,0.0856,0.0951,0.0986,0.0956,0.0426,0.0407,0.0106,0.0179,0.0056,0.0236,0.0114,0.0136,0.0117,0.006,0.0058,0.0031,0.0072,0.0045,R
 34 | 0.0195,0.0213,0.0058,0.019,0.0319,0.0571,0.1004,0.0668,0.0691,0.0242,0.0728,0.0639,0.3002,0.3854,0.4767,0.4602,0.3175,0.416,0.6428,1.0,0.8631,0.5212,0.3156,0.5952,0.7732,0.6042,0.4375,0.5487,0.472,0.6235,0.3851,0.159,0.3891,0.5294,0.3504,0.448,0.4041,0.5031,0.6475,0.5493,0.3548,0.2028,0.1882,0.0845,0.1315,0.159,0.0562,0.0617,0.0343,0.037,0.0261,0.0157,0.0074,0.0271,0.0203,0.0089,0.0095,0.0095,0.0021,0.0053,R
 35 | 0.0442,0.0477,0.0049,0.0581,0.0278,0.0678,0.1664,0.149,0.0974,0.1268,0.1109,0.2375,0.2007,0.214,0.1109,0.2036,0.2468,0.6682,0.8345,0.8252,0.8017,0.8982,0.9664,0.8515,0.6626,0.3241,0.2054,0.5669,0.5726,0.4877,0.7532,0.76,0.5185,0.412,0.556,0.5569,0.1336,0.3831,0.4611,0.433,0.2556,0.1466,0.3489,0.2659,0.0944,0.137,0.1344,0.0416,0.0719,0.0637,0.021,0.0204,0.0216,0.0135,0.0055,0.0073,0.008,0.0105,0.0059,0.0105,R
 36 | 0.0311,0.0491,0.0692,0.0831,0.0079,0.02,0.0981,0.1016,0.2025,0.0767,0.1767,0.2555,0.2812,0.2722,0.3227,0.3463,0.5395,0.7911,0.9064,0.8701,0.7672,0.2957,0.4148,0.6043,0.3178,0.3482,0.6158,0.8049,0.6289,0.4999,0.583,0.666,0.4124,0.126,0.2487,0.4676,0.5382,0.315,0.2139,0.1848,0.1679,0.2328,0.1015,0.0713,0.0615,0.0779,0.0761,0.0845,0.0592,0.0068,0.0089,0.0087,0.0032,0.013,0.0188,0.0101,0.0229,0.0182,0.0046,0.0038,R
 37 | 0.0206,0.0132,0.0533,0.0569,0.0647,0.1432,0.1344,0.2041,0.1571,0.1573,0.2327,0.1785,0.1507,0.1916,0.2061,0.2307,0.236,0.1299,0.3812,0.5858,0.4497,0.4876,1.0,0.8675,0.4718,0.5341,0.6197,0.7143,0.5605,0.3728,0.2481,0.1921,0.1386,0.3325,0.2883,0.3228,0.2607,0.204,0.2396,0.1319,0.0683,0.0334,0.0716,0.0976,0.0787,0.0522,0.05,0.0231,0.0221,0.0144,0.0307,0.0386,0.0147,0.0018,0.01,0.0096,0.0077,0.018,0.0109,0.007,R
 38 | 0.0094,0.0166,0.0398,0.0359,0.0681,0.0706,0.102,0.0893,0.0381,0.1328,0.1303,0.0273,0.0644,0.0712,0.1204,0.0717,0.1224,0.2349,0.3684,0.3918,0.4925,0.8793,0.9606,0.8786,0.6905,0.6937,0.5674,0.654,0.7802,0.7575,0.5836,0.6316,0.8108,0.9039,0.8647,0.6695,0.4027,0.237,0.2685,0.3662,0.3267,0.22,0.2996,0.2205,0.1163,0.0635,0.0465,0.0422,0.0174,0.0172,0.0134,0.0141,0.0191,0.0145,0.0065,0.0129,0.0217,0.0087,0.0077,0.0122,R
 39 | 0.0333,0.0221,0.027,0.0481,0.0679,0.0981,0.0843,0.1172,0.0759,0.092,0.1475,0.0522,0.1119,0.097,0.1174,0.1678,0.1642,0.1205,0.0494,0.1544,0.3485,0.6146,0.9146,0.9364,0.8677,0.8772,0.8553,0.8833,1.0,0.8296,0.6601,0.5499,0.5716,0.6859,0.6825,0.5142,0.275,0.1358,0.1551,0.2646,0.1994,0.1883,0.2746,0.1651,0.0575,0.0695,0.0598,0.0456,0.0021,0.0068,0.0036,0.0022,0.0032,0.006,0.0054,0.0063,0.0143,0.0132,0.0051,0.0041,R
 40 | 0.0123,0.0022,0.0196,0.0206,0.018,0.0492,0.0033,0.0398,0.0791,0.0475,0.1152,0.052,0.1192,0.1943,0.184,0.2077,0.1956,0.163,0.1218,0.1017,0.1354,0.3157,0.4645,0.5906,0.6776,0.8119,0.8594,0.9228,0.8387,0.7238,0.6292,0.5181,0.4629,0.5255,0.5147,0.3929,0.1279,0.0411,0.0859,0.1131,0.1306,0.1757,0.2648,0.1955,0.0656,0.058,0.0319,0.0301,0.0272,0.0074,0.0149,0.0125,0.0134,0.0026,0.0038,0.0018,0.0113,0.0058,0.0047,0.0071,R
 41 | 0.0091,0.0213,0.0206,0.0505,0.0657,0.0795,0.097,0.0872,0.0743,0.0837,0.1579,0.0898,0.0309,0.1856,0.2969,0.2032,0.1264,0.1655,0.1661,0.2091,0.231,0.446,0.6634,0.6933,0.7663,0.8206,0.7049,0.756,0.7466,0.6387,0.4846,0.3328,0.5356,0.8741,0.8573,0.6718,0.3446,0.315,0.2702,0.2598,0.2742,0.3594,0.4382,0.246,0.0758,0.0187,0.0797,0.0748,0.0367,0.0155,0.03,0.0112,0.0112,0.0102,0.0026,0.0097,0.0098,0.0043,0.0071,0.0108,R
 42 | 0.0068,0.0232,0.0513,0.0444,0.0249,0.0637,0.0422,0.113,0.1911,0.2475,0.1606,0.0922,0.2398,0.322,0.4295,0.2652,0.0666,0.1442,0.2373,0.2595,0.2493,0.3903,0.6384,0.8037,0.7026,0.6874,0.6997,0.8558,1.0,0.9621,0.8996,0.7575,0.6902,0.5686,0.4396,0.4546,0.2959,0.1587,0.1681,0.0842,0.1173,0.1754,0.2728,0.1705,0.0194,0.0213,0.0354,0.042,0.0093,0.0204,0.0199,0.0173,0.0163,0.0055,0.0045,0.0068,0.0041,0.0052,0.0194,0.0105,R
 43 | 0.0093,0.0185,0.0056,0.0064,0.026,0.0458,0.047,0.0057,0.0425,0.064,0.0888,0.1599,0.1541,0.2768,0.2176,0.2799,0.3491,0.2824,0.2479,0.3005,0.43,0.4684,0.452,0.5026,0.6217,0.6571,0.6632,0.7321,0.8534,1.0,0.8448,0.6354,0.6308,0.6211,0.6976,0.5868,0.4889,0.3683,0.2043,0.1469,0.222,0.1449,0.149,0.1211,0.1144,0.0791,0.0365,0.0152,0.0085,0.012,0.0022,0.0069,0.0064,0.0129,0.0114,0.0054,0.0089,0.005,0.0058,0.0025,R
 44 | 0.0211,0.0319,0.0415,0.0286,0.0121,0.0438,0.1299,0.139,0.0695,0.0568,0.0869,0.1935,0.1478,0.1871,0.1994,0.3283,0.6861,0.5814,0.25,0.1734,0.3363,0.5588,0.6592,0.7012,0.8099,0.8901,0.8745,0.7887,0.8725,0.9376,0.892,0.7508,0.6832,0.761,0.9017,1.0,0.9123,0.7388,0.5915,0.4057,0.3019,0.2331,0.2931,0.2298,0.2391,0.191,0.1096,0.03,0.0171,0.0383,0.0053,0.009,0.0042,0.0153,0.0106,0.002,0.0105,0.0049,0.007,0.008,R
 45 | 0.0093,0.0269,0.0217,0.0339,0.0305,0.1172,0.145,0.0638,0.074,0.136,0.2132,0.3738,0.3738,0.2673,0.2333,0.5367,0.7312,0.7659,0.6271,0.4395,0.433,0.4326,0.5544,0.736,0.8589,0.8989,0.942,0.9401,0.9379,0.8575,0.7284,0.67,0.7547,0.8773,0.9919,0.9922,0.9419,0.8388,0.6605,0.4816,0.2917,0.1769,0.1136,0.0701,0.1578,0.1938,0.1106,0.0693,0.0176,0.0205,0.0309,0.0212,0.0091,0.0056,0.0086,0.0092,0.007,0.0116,0.006,0.011,R
 46 | 0.0257,0.0447,0.0388,0.0239,0.1315,0.1323,0.1608,0.2145,0.0847,0.0561,0.0891,0.0861,0.1531,0.1524,0.1849,0.2871,0.2009,0.2748,0.5017,0.2172,0.4978,0.5265,0.3647,0.5768,0.5161,0.5715,0.4006,0.365,0.6685,0.8659,0.8052,0.4082,0.3379,0.5092,0.6776,0.7313,0.6062,0.704,0.8849,0.8979,0.7751,0.7247,0.7733,0.7762,0.6009,0.4514,0.3096,0.1859,0.0956,0.0206,0.0206,0.0096,0.0153,0.0096,0.0131,0.0198,0.0025,0.0199,0.0255,0.018,R
 47 | 0.0408,0.0653,0.0397,0.0604,0.0496,0.1817,0.1178,0.1024,0.0583,0.2176,0.2459,0.3332,0.3087,0.2613,0.3232,0.3731,0.4203,0.5364,0.7062,0.8196,0.8835,0.8299,0.7609,0.7605,0.8367,0.8905,0.7652,0.5897,0.3037,0.0823,0.2787,0.7241,0.8032,0.805,0.7676,0.7468,0.6253,0.173,0.2916,0.5003,0.522,0.4824,0.4004,0.3877,0.1651,0.0442,0.0663,0.0418,0.0475,0.0235,0.0066,0.0062,0.0129,0.0184,0.0069,0.0198,0.0199,0.0102,0.007,0.0055,R
 48 | 0.0308,0.0339,0.0202,0.0889,0.157,0.175,0.092,0.1353,0.1593,0.2795,0.3336,0.294,0.1608,0.3335,0.4985,0.7295,0.735,0.8253,0.8793,0.9657,1.0,0.8707,0.6471,0.5973,0.8218,0.7755,0.6111,0.4195,0.299,0.1354,0.2438,0.5624,0.5555,0.6963,0.7298,0.7022,0.5468,0.1421,0.4738,0.641,0.4375,0.3178,0.2377,0.2808,0.1374,0.1136,0.1034,0.0688,0.0422,0.0117,0.007,0.0167,0.0127,0.0138,0.009,0.0051,0.0029,0.0122,0.0056,0.002,R
 49 | 0.0373,0.0281,0.0232,0.0225,0.0179,0.0733,0.0841,0.1031,0.0993,0.0802,0.1564,0.2565,0.2624,0.1179,0.0597,0.1563,0.2241,0.3586,0.1792,0.3256,0.6079,0.6988,0.8391,0.8553,0.771,0.6215,0.5736,0.4402,0.4056,0.4411,0.513,0.5965,0.7272,0.6539,0.5902,0.5393,0.4897,0.4081,0.4145,0.6003,0.7196,0.6633,0.6287,0.4087,0.3212,0.2518,0.1482,0.0988,0.0317,0.0269,0.0066,0.0008,0.0045,0.0024,0.0006,0.0073,0.0096,0.0054,0.0085,0.006,R
 50 | 0.019,0.0038,0.0642,0.0452,0.0333,0.069,0.0901,0.1454,0.074,0.0349,0.1459,0.3473,0.3197,0.2823,0.0166,0.0572,0.2164,0.4563,0.3819,0.5627,0.6484,0.7235,0.8242,0.8766,1.0,0.8582,0.6563,0.5087,0.4817,0.453,0.4521,0.4532,0.5385,0.5308,0.5356,0.5271,0.426,0.2436,0.1205,0.3845,0.4107,0.5067,0.4216,0.2479,0.1586,0.1124,0.0651,0.0789,0.0325,0.007,0.0026,0.0093,0.0118,0.0112,0.0094,0.014,0.0072,0.0022,0.0055,0.0122,R
 51 | 0.0119,0.0582,0.0623,0.06,0.1397,0.1883,0.1422,0.1447,0.0487,0.0864,0.2143,0.372,0.2665,0.2113,0.1103,0.1136,0.1934,0.4142,0.3279,0.6222,0.7468,0.7676,0.7867,0.8253,1.0,0.9481,0.7539,0.6008,0.5437,0.5387,0.5619,0.5141,0.6084,0.5621,0.5956,0.6078,0.5025,0.2829,0.0477,0.2811,0.3422,0.5147,0.4372,0.247,0.1708,0.1343,0.0838,0.0755,0.0304,0.0074,0.0069,0.0025,0.0103,0.0074,0.0123,0.0069,0.0076,0.0073,0.003,0.0138,R
 52 | 0.0353,0.0713,0.0326,0.0272,0.037,0.0792,0.1083,0.0687,0.0298,0.088,0.1078,0.0979,0.225,0.2819,0.2099,0.124,0.1699,0.0939,0.1091,0.141,0.1268,0.3151,0.143,0.2264,0.5756,0.7876,0.7158,0.5998,0.5583,0.6295,0.7659,0.894,0.8436,0.6807,0.838,1.0,0.9497,0.7866,0.5647,0.348,0.2585,0.2304,0.2948,0.3363,0.3017,0.2193,0.1316,0.1078,0.0559,0.0035,0.0098,0.0163,0.0242,0.0043,0.0202,0.0108,0.0037,0.0096,0.0093,0.0053,R
 53 | 0.0131,0.0068,0.0308,0.0311,0.0085,0.0767,0.0771,0.064,0.0726,0.0901,0.075,0.0844,0.1226,0.1619,0.2317,0.2934,0.3526,0.3657,0.3221,0.3093,0.4084,0.4285,0.4663,0.5956,0.6948,0.8386,0.8875,0.6404,0.3308,0.3425,0.492,0.4592,0.3034,0.4366,0.5175,0.5122,0.4746,0.4902,0.4603,0.446,0.4196,0.2873,0.2296,0.0949,0.0095,0.0527,0.0383,0.0107,0.0108,0.0077,0.0109,0.0062,0.0028,0.004,0.0075,0.0039,0.0053,0.0013,0.0052,0.0023,R
 54 | 0.0087,0.0046,0.0081,0.023,0.0586,0.0682,0.0993,0.0717,0.0576,0.0818,0.1315,0.1862,0.2789,0.2579,0.224,0.2568,0.2933,0.2991,0.3924,0.4691,0.5665,0.6464,0.6774,0.7577,0.8856,0.9419,1.0,0.8564,0.679,0.5587,0.4147,0.2946,0.2025,0.0688,0.1171,0.2157,0.2216,0.2776,0.2309,0.1444,0.1513,0.1745,0.1756,0.1424,0.0908,0.0138,0.0469,0.048,0.0159,0.0045,0.0015,0.0052,0.0038,0.0079,0.0114,0.005,0.003,0.0064,0.0058,0.003,R
 55 | 0.0293,0.0378,0.0257,0.0062,0.013,0.0612,0.0895,0.1107,0.0973,0.0751,0.0528,0.1209,0.1763,0.2039,0.2727,0.2321,0.2676,0.2934,0.3295,0.491,0.5402,0.6257,0.6826,0.7527,0.8504,0.8938,0.9928,0.9134,0.708,0.6318,0.6126,0.4638,0.2797,0.1721,0.1665,0.2561,0.2735,0.3209,0.2724,0.188,0.1552,0.2522,0.2121,0.1801,0.1473,0.0681,0.1091,0.0919,0.0397,0.0093,0.0076,0.0065,0.0072,0.0108,0.0051,0.0102,0.0041,0.0055,0.005,0.0087,R
 56 | 0.0132,0.008,0.0188,0.0141,0.0436,0.0668,0.0609,0.0131,0.0899,0.0922,0.1445,0.1475,0.2087,0.2558,0.2603,0.1985,0.2394,0.3134,0.4077,0.4529,0.4893,0.5666,0.6234,0.6741,0.8282,0.8823,0.9196,0.8965,0.7549,0.6736,0.6463,0.5007,0.3663,0.2298,0.1362,0.2123,0.2395,0.2673,0.2865,0.206,0.1659,0.2633,0.2552,0.1696,0.1467,0.1286,0.0926,0.0716,0.0325,0.0258,0.0136,0.0044,0.0028,0.0021,0.0022,0.0048,0.0138,0.014,0.0028,0.0064,R
 57 | 0.0201,0.0116,0.0123,0.0245,0.0547,0.0208,0.0891,0.0836,0.1335,0.1199,0.1742,0.1387,0.2042,0.258,0.2616,0.2097,0.2532,0.3213,0.4327,0.476,0.5328,0.6057,0.6696,0.7476,0.893,0.9405,1.0,0.9785,0.8473,0.7639,0.6701,0.4989,0.3718,0.2196,0.1416,0.268,0.263,0.3104,0.3392,0.2123,0.117,0.2655,0.2203,0.1541,0.1464,0.1044,0.1225,0.0745,0.049,0.0224,0.0032,0.0076,0.0045,0.0056,0.0075,0.0037,0.0045,0.0029,0.0008,0.0018,R
 58 | 0.0152,0.0102,0.0113,0.0263,0.0097,0.0391,0.0857,0.0915,0.0949,0.1504,0.1911,0.2115,0.2249,0.2573,0.1701,0.2023,0.2538,0.3417,0.4026,0.4553,0.5525,0.5991,0.5854,0.7114,0.95,0.9858,1.0,0.9578,0.8642,0.7128,0.5893,0.4323,0.2897,0.1744,0.077,0.2297,0.2459,0.3101,0.3312,0.222,0.0871,0.2064,0.1808,0.1624,0.112,0.0815,0.1117,0.095,0.0412,0.012,0.0048,0.0049,0.0041,0.0036,0.0013,0.0046,0.0037,0.0011,0.0034,0.0033,R
 59 | 0.0216,0.0124,0.0174,0.0152,0.0608,0.1026,0.1139,0.0877,0.116,0.0866,0.1564,0.078,0.0997,0.0915,0.0662,0.1134,0.174,0.2573,0.3294,0.391,0.5438,0.6115,0.7022,0.761,0.7973,0.9105,0.8807,0.7949,0.799,0.718,0.6407,0.6312,0.5929,0.6168,0.6498,0.6764,0.6253,0.5117,0.389,0.3273,0.2509,0.153,0.1323,0.1657,0.1215,0.0978,0.0452,0.0273,0.0179,0.0092,0.0018,0.0052,0.0049,0.0096,0.0134,0.0122,0.0047,0.0018,0.0006,0.0023,R
 60 | 0.0225,0.0019,0.0075,0.0097,0.0445,0.0906,0.0889,0.0655,0.1624,0.1452,0.1442,0.0948,0.0618,0.1641,0.0708,0.0844,0.259,0.2679,0.3094,0.4678,0.5958,0.7245,0.8773,0.9214,0.9282,0.9942,1.0,0.9071,0.8545,0.7293,0.6499,0.6071,0.5588,0.5967,0.6275,0.5459,0.4786,0.3965,0.2087,0.1651,0.1836,0.0652,0.0758,0.0486,0.0353,0.0297,0.0241,0.0379,0.0119,0.0073,0.0051,0.0034,0.0129,0.01,0.0044,0.0057,0.003,0.0035,0.0021,0.0027,R
 61 | 0.0125,0.0152,0.0218,0.0175,0.0362,0.0696,0.0873,0.0616,0.1252,0.1302,0.0888,0.05,0.0628,0.1274,0.0801,0.0742,0.2048,0.295,0.3193,0.4567,0.5959,0.7101,0.8225,0.8425,0.9065,0.9802,1.0,0.8752,0.7583,0.6616,0.5786,0.5128,0.4776,0.4994,0.5197,0.5071,0.4577,0.3505,0.1845,0.189,0.1967,0.1041,0.055,0.0492,0.0622,0.0505,0.0247,0.0219,0.0102,0.0047,0.0019,0.0041,0.0074,0.003,0.005,0.0048,0.0017,0.0041,0.0086,0.0058,R
 62 | 0.013,0.0006,0.0088,0.0456,0.0525,0.0778,0.0931,0.0941,0.1711,0.1483,0.1532,0.11,0.089,0.1236,0.1197,0.1145,0.2137,0.2838,0.364,0.543,0.6673,0.7979,0.9273,0.9027,0.9192,1.0,0.9821,0.9092,0.8184,0.6962,0.59,0.5447,0.5142,0.5389,0.5531,0.5318,0.4826,0.379,0.1831,0.175,0.1679,0.0674,0.0609,0.0375,0.0533,0.0278,0.0179,0.0114,0.0073,0.0116,0.0092,0.0078,0.0041,0.0013,0.0011,0.0045,0.0039,0.0022,0.0023,0.0016,R
 63 | 0.0135,0.0045,0.0051,0.0289,0.0561,0.0929,0.1031,0.0883,0.1596,0.1908,0.1576,0.1112,0.1197,0.1174,0.1415,0.2215,0.2658,0.2713,0.3862,0.5717,0.6797,0.8747,1.0,0.8948,0.842,0.9174,0.9307,0.905,0.8228,0.6986,0.5831,0.4924,0.4563,0.5159,0.567,0.5284,0.5144,0.3742,0.2282,0.1193,0.1088,0.0431,0.107,0.0583,0.0046,0.0473,0.0408,0.029,0.0192,0.0094,0.0025,0.0037,0.0084,0.0102,0.0096,0.0024,0.0037,0.0028,0.003,0.003,R
 64 | 0.0086,0.0215,0.0242,0.0445,0.0667,0.0771,0.0499,0.0906,0.1229,0.1185,0.0775,0.1101,0.1042,0.0853,0.0456,0.1304,0.269,0.2947,0.3669,0.4948,0.6275,0.8162,0.9237,0.871,0.8052,0.8756,1.0,0.9858,0.9427,0.8114,0.6987,0.681,0.6591,0.6954,0.729,0.668,0.5917,0.4899,0.3439,0.2366,0.1716,0.1013,0.0766,0.0845,0.026,0.0333,0.0205,0.0309,0.0101,0.0095,0.0047,0.0072,0.0054,0.0022,0.0016,0.0029,0.0058,0.005,0.0024,0.003,R
 65 | 0.0067,0.0096,0.0024,0.0058,0.0197,0.0618,0.0432,0.0951,0.0836,0.118,0.0978,0.0909,0.0656,0.0593,0.0832,0.1297,0.2038,0.3811,0.4451,0.5224,0.5911,0.6566,0.6308,0.5998,0.4958,0.5647,0.6906,0.8513,1.0,0.9166,0.7676,0.6177,0.5468,0.5516,0.5463,0.5515,0.4561,0.3466,0.3384,0.2853,0.2502,0.1641,0.1605,0.1491,0.1326,0.0687,0.0602,0.0561,0.0306,0.0154,0.0029,0.0048,0.0023,0.002,0.004,0.0019,0.0034,0.0034,0.0051,0.0031,R
 66 | 0.0071,0.0103,0.0135,0.0494,0.0253,0.0806,0.0701,0.0738,0.0117,0.0898,0.0289,0.1554,0.1437,0.1035,0.1424,0.1227,0.0892,0.2047,0.0827,0.1524,0.3031,0.1608,0.0667,0.1426,0.0395,0.1653,0.3399,0.4855,0.5206,0.5508,0.6102,0.5989,0.6764,0.8897,1.0,0.9517,0.8459,0.7073,0.6697,0.6326,0.5102,0.4161,0.2816,0.1705,0.1421,0.0971,0.0879,0.0863,0.0355,0.0233,0.0252,0.0043,0.0048,0.0076,0.0124,0.0105,0.0054,0.0032,0.0073,0.0063,R
 67 | 0.0176,0.0172,0.0501,0.0285,0.0262,0.0351,0.0362,0.0535,0.0258,0.0474,0.0526,0.1854,0.104,0.0948,0.0912,0.1688,0.1568,0.0375,0.1316,0.2086,0.1976,0.0946,0.1965,0.1242,0.0616,0.2141,0.4642,0.6471,0.634,0.6107,0.7046,0.5376,0.5934,0.8443,0.9481,0.9705,0.7766,0.6313,0.576,0.6148,0.545,0.4813,0.3406,0.1916,0.1134,0.064,0.0911,0.098,0.0563,0.0187,0.0088,0.0042,0.0175,0.0171,0.0079,0.005,0.0112,0.0179,0.0294,0.0063,R
 68 | 0.0265,0.044,0.0137,0.0084,0.0305,0.0438,0.0341,0.078,0.0844,0.0779,0.0327,0.206,0.1908,0.1065,0.1457,0.2232,0.207,0.1105,0.1078,0.1165,0.2224,0.0689,0.206,0.2384,0.0904,0.2278,0.5872,0.8457,0.8467,0.7679,0.8055,0.626,0.6545,0.8747,0.9885,0.9348,0.696,0.5733,0.5872,0.6663,0.5651,0.5247,0.3684,0.1997,0.1512,0.0508,0.0931,0.0982,0.0524,0.0188,0.01,0.0038,0.0187,0.0156,0.0068,0.0097,0.0073,0.0081,0.0086,0.0095,R
 69 | 0.0368,0.0403,0.0317,0.0293,0.082,0.1342,0.1161,0.0663,0.0155,0.0506,0.0906,0.2545,0.1464,0.1272,0.1223,0.1669,0.1424,0.1285,0.1857,0.1136,0.2069,0.0219,0.24,0.2547,0.024,0.1923,0.4753,0.7003,0.6825,0.6443,0.7063,0.5373,0.6601,0.8708,0.9518,0.9605,0.7712,0.6772,0.6431,0.672,0.6035,0.5155,0.3802,0.2278,0.1522,0.0801,0.0804,0.0752,0.0566,0.0175,0.0058,0.0091,0.016,0.016,0.0081,0.007,0.0135,0.0067,0.0078,0.0068,R
 70 | 0.0195,0.0142,0.0181,0.0406,0.0391,0.0249,0.0892,0.0973,0.084,0.1191,0.1522,0.1322,0.1434,0.1244,0.0653,0.089,0.1226,0.1846,0.388,0.3658,0.2297,0.261,0.4193,0.5848,0.5643,0.5448,0.4772,0.6897,0.9797,1.0,0.9546,0.8835,0.7662,0.6547,0.5447,0.4593,0.4679,0.1987,0.0699,0.1493,0.1713,0.1654,0.26,0.3846,0.3754,0.2414,0.1077,0.0224,0.0155,0.0187,0.0125,0.0028,0.0067,0.012,0.0012,0.0022,0.0058,0.0042,0.0067,0.0012,R
 71 | 0.0216,0.0215,0.0273,0.0139,0.0357,0.0785,0.0906,0.0908,0.1151,0.0973,0.1203,0.1102,0.1192,0.1762,0.239,0.2138,0.1929,0.1765,0.0746,0.1265,0.2005,0.1571,0.2605,0.5386,0.844,1.0,0.8684,0.6742,0.5537,0.4638,0.3609,0.2055,0.162,0.2092,0.31,0.2344,0.1058,0.0383,0.0528,0.1291,0.2241,0.1915,0.1587,0.0942,0.084,0.067,0.0342,0.0469,0.0357,0.0136,0.0082,0.014,0.0044,0.0052,0.0073,0.0021,0.0047,0.0024,0.0009,0.0017,R
 72 | 0.0065,0.0122,0.0068,0.0108,0.0217,0.0284,0.0527,0.0575,0.1054,0.1109,0.0937,0.0827,0.092,0.0911,0.1487,0.1666,0.1268,0.1374,0.1095,0.1286,0.2146,0.2889,0.4238,0.6168,0.8167,0.9622,0.828,0.5816,0.4667,0.3539,0.2727,0.141,0.1863,0.2176,0.236,0.1725,0.0589,0.0621,0.1847,0.2452,0.2984,0.3041,0.2275,0.148,0.1102,0.1178,0.0608,0.0333,0.0276,0.01,0.0023,0.0069,0.0025,0.0027,0.0052,0.0036,0.0026,0.0036,0.0006,0.0035,R
 73 | 0.0036,0.0078,0.0092,0.0387,0.053,0.1197,0.1243,0.1026,0.1239,0.0888,0.0937,0.1245,0.1599,0.1542,0.1846,0.1732,0.1477,0.1748,0.1455,0.1579,0.2257,0.1975,0.3368,0.5828,0.8505,1.0,0.8457,0.6624,0.5564,0.3925,0.3233,0.2054,0.192,0.2227,0.3147,0.2268,0.0795,0.0748,0.1166,0.1969,0.2619,0.2507,0.1983,0.0948,0.0931,0.0965,0.0381,0.0435,0.0336,0.0055,0.0079,0.0119,0.0055,0.0035,0.0036,0.0004,0.0018,0.0049,0.0024,0.0016,R
 74 | 0.0208,0.0186,0.0131,0.0211,0.061,0.0613,0.0612,0.0506,0.0989,0.1093,0.1063,0.1179,0.1291,0.1591,0.168,0.1918,0.1615,0.1647,0.1397,0.1426,0.2429,0.2816,0.429,0.6443,0.9061,1.0,0.8087,0.6119,0.526,0.3677,0.2746,0.102,0.1339,0.1582,0.1952,0.1787,0.0429,0.1096,0.1762,0.2481,0.315,0.292,0.1902,0.0696,0.0758,0.091,0.0441,0.0244,0.0265,0.0095,0.014,0.0074,0.0063,0.0081,0.0087,0.0044,0.0028,0.0019,0.0049,0.0023,R
 75 | 0.0139,0.0222,0.0089,0.0108,0.0215,0.0136,0.0659,0.0954,0.0786,0.1015,0.1261,0.0828,0.0493,0.0848,0.1514,0.1396,0.1066,0.1923,0.2991,0.3247,0.3797,0.5658,0.7483,0.8757,0.9048,0.7511,0.6858,0.7043,0.5864,0.3773,0.2206,0.2628,0.2672,0.2907,0.1982,0.2288,0.3186,0.2871,0.2921,0.2806,0.2682,0.2112,0.1513,0.1789,0.185,0.1717,0.0898,0.0656,0.0445,0.011,0.0024,0.0062,0.0072,0.0113,0.0012,0.0022,0.0025,0.0059,0.0039,0.0048,R
 76 | 0.0109,0.0093,0.0121,0.0378,0.0679,0.0863,0.1004,0.0664,0.0941,0.1036,0.0972,0.0501,0.1546,0.3404,0.4804,0.657,0.7738,0.7827,0.8152,0.8129,0.8297,0.8535,0.887,0.8894,0.898,0.9667,1.0,0.9134,0.6762,0.4659,0.2895,0.2959,0.1746,0.2112,0.2569,0.2276,0.2149,0.1601,0.0371,0.0117,0.0488,0.0288,0.0597,0.0431,0.0369,0.0025,0.0327,0.0257,0.0182,0.0108,0.0124,0.0077,0.0023,0.0117,0.0053,0.0077,0.0076,0.0056,0.0055,0.0039,R
 77 | 0.0202,0.0104,0.0325,0.0239,0.0807,0.1529,0.1154,0.0608,0.1317,0.137,0.0843,0.0269,0.1254,0.3046,0.5584,0.7973,0.8341,0.8057,0.8616,0.8769,0.9413,0.9403,0.9409,1.0,0.9725,0.9309,0.9351,0.7317,0.4421,0.3244,0.4161,0.4611,0.4031,0.3,0.2459,0.1348,0.2541,0.2255,0.1598,0.1485,0.0845,0.0569,0.0855,0.1262,0.1153,0.057,0.0426,0.0425,0.0235,0.0006,0.0188,0.0127,0.0081,0.0067,0.0043,0.0065,0.0049,0.0054,0.0073,0.0054,R
 78 | 0.0239,0.0189,0.0466,0.044,0.0657,0.0742,0.138,0.1099,0.1384,0.1376,0.0938,0.0259,0.1499,0.2851,0.5743,0.8278,0.8669,0.8131,0.9045,0.9046,1.0,0.9976,0.9872,0.9761,0.9009,0.9724,0.9675,0.7633,0.4434,0.3822,0.4727,0.4007,0.3381,0.3172,0.2222,0.0733,0.2692,0.1888,0.0712,0.1062,0.0694,0.03,0.0893,0.1459,0.1348,0.0391,0.0546,0.0469,0.0201,0.0095,0.0155,0.0091,0.0151,0.008,0.0018,0.0078,0.0045,0.0026,0.0036,0.0024,R
 79 | 0.0336,0.0294,0.0476,0.0539,0.0794,0.0804,0.1136,0.1228,0.1235,0.0842,0.0357,0.0689,0.1705,0.3257,0.4602,0.6225,0.7327,0.7843,0.7988,0.8261,1.0,0.9814,0.962,0.9601,0.9118,0.9086,0.7931,0.5877,0.3474,0.4235,0.4633,0.341,0.2849,0.2847,0.1742,0.0549,0.1192,0.1154,0.0855,0.1811,0.1264,0.0799,0.0378,0.1268,0.1125,0.0505,0.0949,0.0677,0.0259,0.017,0.0033,0.015,0.0111,0.0032,0.0035,0.0169,0.0137,0.0015,0.0069,0.0051,R
 80 | 0.0231,0.0351,0.003,0.0304,0.0339,0.086,0.1738,0.1351,0.1063,0.0347,0.0575,0.1382,0.2274,0.4038,0.5223,0.6847,0.7521,0.776,0.7708,0.8627,1.0,0.8873,0.8057,0.876,0.9066,0.943,0.8846,0.65,0.297,0.2423,0.2992,0.2285,0.2277,0.1529,0.1037,0.0352,0.1073,0.1373,0.1331,0.1454,0.1115,0.044,0.0762,0.1381,0.0831,0.0654,0.0844,0.0595,0.0497,0.0313,0.0154,0.0106,0.0097,0.0022,0.0052,0.0072,0.0056,0.0038,0.0043,0.003,R
 81 | 0.0108,0.0086,0.0058,0.046,0.0752,0.0887,0.1015,0.0494,0.0472,0.0393,0.1106,0.1412,0.2202,0.2976,0.4116,0.4754,0.539,0.6279,0.706,0.7918,0.9493,1.0,0.9645,0.9432,0.8658,0.7895,0.6501,0.4492,0.4739,0.6153,0.4929,0.3195,0.3735,0.3336,0.1052,0.0671,0.0379,0.0461,0.1694,0.2169,0.1677,0.0644,0.0159,0.0778,0.0653,0.021,0.0509,0.0387,0.0262,0.0101,0.0161,0.0029,0.0078,0.0114,0.0083,0.0058,0.0003,0.0023,0.0026,0.0027,R
 82 | 0.0229,0.0369,0.004,0.0375,0.0455,0.1452,0.2211,0.1188,0.075,0.1631,0.2709,0.3358,0.4091,0.44,0.5485,0.7213,0.8137,0.9185,1.0,0.9418,0.9116,0.9349,0.7484,0.5146,0.4106,0.3443,0.6981,0.8713,0.9013,0.8014,0.438,0.1319,0.1709,0.2484,0.3044,0.2312,0.1338,0.2056,0.2474,0.279,0.161,0.0056,0.0351,0.1148,0.1331,0.0276,0.0763,0.0631,0.0309,0.024,0.0115,0.0064,0.0022,0.0122,0.0151,0.0056,0.0026,0.0029,0.0104,0.0163,R
 83 | 0.01,0.0194,0.0155,0.0489,0.0839,0.1009,0.1627,0.2071,0.2696,0.299,0.3242,0.3565,0.3951,0.5201,0.6953,0.8468,1.0,0.9278,0.851,0.801,0.8142,0.8825,0.7302,0.6107,0.7159,0.8458,0.6319,0.4808,0.6291,0.7152,0.6005,0.4235,0.4106,0.3992,0.173,0.1975,0.237,0.1339,0.1583,0.3151,0.1968,0.2054,0.1272,0.1129,0.1946,0.2195,0.193,0.1498,0.0773,0.0196,0.0122,0.013,0.0073,0.0077,0.0075,0.006,0.008,0.0019,0.0053,0.0019,R
 84 | 0.0409,0.0421,0.0573,0.013,0.0183,0.1019,0.1054,0.107,0.2302,0.2259,0.2373,0.3323,0.3827,0.484,0.6812,0.7555,0.9522,0.9826,0.8871,0.8268,0.7561,0.8217,0.6967,0.6444,0.6948,0.8014,0.6053,0.6084,0.8877,0.8557,0.5563,0.2897,0.3638,0.4786,0.2908,0.0899,0.2043,0.1707,0.0407,0.1286,0.1581,0.2191,0.1701,0.0971,0.2217,0.2732,0.1874,0.1062,0.0665,0.0405,0.0113,0.0028,0.0036,0.0105,0.012,0.0087,0.0061,0.0061,0.003,0.0078,R
 85 | 0.0217,0.034,0.0392,0.0236,0.1081,0.1164,0.1398,0.1009,0.1147,0.1777,0.4079,0.4113,0.3973,0.5078,0.6509,0.8073,0.9819,1.0,0.9407,0.8452,0.8106,0.846,0.6212,0.5815,0.7745,0.8204,0.5601,0.2989,0.5009,0.6628,0.5753,0.4055,0.3746,0.3481,0.158,0.1422,0.213,0.1866,0.1003,0.2396,0.2241,0.2029,0.071,0.1606,0.1669,0.17,0.1829,0.1403,0.0506,0.0224,0.0095,0.0031,0.0103,0.0078,0.0077,0.0094,0.0031,0.003,0.0013,0.0069,R
 86 | 0.0378,0.0318,0.0423,0.035,0.1787,0.1635,0.0887,0.0817,0.1779,0.2053,0.3135,0.3118,0.3686,0.3885,0.585,0.7868,0.9739,1.0,0.9843,0.861,0.8443,0.9061,0.5847,0.4033,0.5946,0.6793,0.6389,0.5002,0.5578,0.4831,0.4729,0.3318,0.3969,0.3894,0.2314,0.1036,0.1312,0.0864,0.2569,0.3179,0.2649,0.2714,0.1713,0.0584,0.123,0.22,0.2198,0.1074,0.0423,0.0162,0.0093,0.0046,0.0044,0.0078,0.0102,0.0065,0.0061,0.0062,0.0043,0.0053,R
 87 | 0.0365,0.1632,0.1636,0.1421,0.113,0.1306,0.2112,0.2268,0.2992,0.3735,0.3042,0.0387,0.2679,0.5397,0.6204,0.7257,0.835,0.6888,0.445,0.3921,0.5605,0.7545,0.8311,1.0,0.8762,0.7092,0.7009,0.5014,0.3942,0.4456,0.4072,0.0773,0.1423,0.0401,0.3597,0.6847,0.7076,0.3597,0.0612,0.3027,0.3966,0.3868,0.238,0.2059,0.2288,0.1704,0.1587,0.1792,0.1022,0.0151,0.0223,0.011,0.0071,0.0205,0.0164,0.0063,0.0078,0.0094,0.011,0.0068,R
 88 | 0.0188,0.037,0.0953,0.0824,0.0249,0.0488,0.1424,0.1972,0.1873,0.1806,0.2139,0.1523,0.1975,0.4844,0.7298,0.7807,0.7906,0.6122,0.42,0.2807,0.5148,0.7569,0.8596,1.0,0.8457,0.6797,0.6971,0.5843,0.4772,0.5201,0.4241,0.1592,0.1668,0.0588,0.3967,0.7147,0.7319,0.3509,0.0589,0.269,0.42,0.3874,0.244,0.2,0.2307,0.1886,0.196,0.1701,0.1366,0.0398,0.0143,0.0093,0.0033,0.0113,0.003,0.0057,0.009,0.0057,0.0068,0.0024,R
 89 | 0.0856,0.0454,0.0382,0.0203,0.0385,0.0534,0.214,0.311,0.2837,0.2751,0.2707,0.0946,0.102,0.4519,0.6737,0.6699,0.7066,0.5632,0.3785,0.2721,0.5297,0.7697,0.8643,0.9304,0.9372,0.6247,0.6024,0.681,0.5047,0.5775,0.4754,0.24,0.2779,0.1997,0.5305,0.7409,0.7775,0.4424,0.1416,0.3508,0.4482,0.4208,0.3054,0.2235,0.2611,0.2798,0.2392,0.2021,0.1326,0.0358,0.0128,0.0172,0.0138,0.0079,0.0037,0.0051,0.0258,0.0102,0.0037,0.0037,R
 90 | 0.0274,0.0242,0.0621,0.056,0.1129,0.0973,0.1823,0.1745,0.144,0.1808,0.2366,0.0906,0.1749,0.4012,0.5187,0.7312,0.9062,0.926,0.7434,0.4463,0.5103,0.6952,0.7755,0.8364,0.7283,0.6399,0.5759,0.4146,0.3495,0.4437,0.2665,0.2024,0.1942,0.0765,0.3725,0.5843,0.4827,0.2347,0.0999,0.3244,0.399,0.2975,0.1684,0.1761,0.1683,0.0729,0.119,0.1297,0.0748,0.0067,0.0255,0.0113,0.0108,0.0085,0.0047,0.0074,0.0104,0.0161,0.022,0.0173,R
 91 | 0.0235,0.0291,0.0749,0.0519,0.0227,0.0834,0.0677,0.2002,0.2876,0.3674,0.2974,0.0837,0.1912,0.504,0.6352,0.6804,0.7505,0.6595,0.4509,0.2964,0.4019,0.6794,0.8297,1.0,0.824,0.7115,0.7726,0.6124,0.4936,0.5648,0.4906,0.182,0.1811,0.1107,0.4603,0.665,0.6423,0.2166,0.1951,0.4947,0.4925,0.4041,0.2402,0.1392,0.1779,0.1946,0.1723,0.1522,0.0929,0.0179,0.0242,0.0083,0.0037,0.0095,0.0105,0.003,0.0132,0.0068,0.0108,0.009,R
 92 | 0.0126,0.0519,0.0621,0.0518,0.1072,0.2587,0.2304,0.2067,0.3416,0.4284,0.3015,0.1207,0.3299,0.5707,0.6962,0.9751,1.0,0.9293,0.621,0.4586,0.5001,0.5032,0.7082,0.842,0.8109,0.769,0.8105,0.6203,0.2356,0.2595,0.6299,0.6762,0.2903,0.4393,0.8529,0.718,0.4801,0.5856,0.4993,0.2866,0.0601,0.1167,0.2737,0.2812,0.2078,0.066,0.0491,0.0345,0.0172,0.0287,0.0027,0.0208,0.0048,0.0199,0.0126,0.0022,0.0037,0.0034,0.0114,0.0077,R
 93 | 0.0253,0.0808,0.0507,0.0244,0.1724,0.3823,0.3729,0.3583,0.3429,0.2197,0.2653,0.3223,0.5582,0.6916,0.7943,0.7152,0.3512,0.2008,0.2676,0.4299,0.528,0.3489,0.143,0.5453,0.6338,0.7712,0.6838,0.8015,0.8073,0.831,0.7792,0.5049,0.1413,0.2767,0.5084,0.4787,0.1356,0.2299,0.2789,0.3833,0.2933,0.1155,0.1705,0.1294,0.0909,0.08,0.0567,0.0198,0.0114,0.0151,0.0085,0.0178,0.0073,0.0079,0.0038,0.0116,0.0033,0.0039,0.0081,0.0053,R
 94 | 0.026,0.0192,0.0254,0.0061,0.0352,0.0701,0.1263,0.108,0.1523,0.163,0.103,0.2187,0.1542,0.263,0.294,0.2978,0.0699,0.1401,0.299,0.3915,0.3598,0.2403,0.4208,0.5675,0.6094,0.6323,0.6549,0.7673,1.0,0.8463,0.5509,0.4444,0.5169,0.4268,0.1802,0.0791,0.0535,0.1906,0.2561,0.2153,0.2769,0.2841,0.1733,0.0815,0.0335,0.0933,0.1018,0.0309,0.0208,0.0318,0.0132,0.0118,0.012,0.0051,0.007,0.0015,0.0035,0.0008,0.0044,0.0077,R
 95 | 0.0459,0.0437,0.0347,0.0456,0.0067,0.089,0.1798,0.1741,0.1598,0.1408,0.2693,0.3259,0.4545,0.5785,0.4471,0.2231,0.2164,0.3201,0.2915,0.4235,0.446,0.238,0.6415,0.8966,0.8918,0.7529,0.6838,0.839,1.0,0.8362,0.5427,0.4577,0.8067,0.6973,0.3915,0.1558,0.1598,0.2161,0.5178,0.4782,0.2344,0.3599,0.2785,0.1807,0.0352,0.0473,0.0322,0.0408,0.0163,0.0088,0.0121,0.0067,0.0032,0.0109,0.0164,0.0151,0.007,0.0085,0.0117,0.0056,R
 96 | 0.0025,0.0309,0.0171,0.0228,0.0434,0.1224,0.1947,0.1661,0.1368,0.143,0.0994,0.225,0.2444,0.3239,0.3039,0.241,0.0367,0.1672,0.3038,0.4069,0.3613,0.1994,0.4611,0.6849,0.7272,0.7152,0.7102,0.8516,1.0,0.769,0.4841,0.3717,0.6096,0.511,0.2586,0.0916,0.0947,0.2287,0.348,0.2095,0.1901,0.2941,0.2211,0.1524,0.0746,0.0606,0.0692,0.0446,0.0344,0.0082,0.0108,0.0149,0.0077,0.0036,0.0114,0.0085,0.0101,0.0016,0.0028,0.0014,R
 97 | 0.0291,0.04,0.0771,0.0809,0.0521,0.1051,0.0145,0.0674,0.1294,0.1146,0.0942,0.0794,0.0252,0.1191,0.1045,0.205,0.1556,0.269,0.3784,0.4024,0.347,0.1395,0.1208,0.2827,0.15,0.2626,0.4468,0.752,0.9036,0.7812,0.4766,0.2483,0.5372,0.6279,0.3647,0.4572,0.6359,0.6474,0.552,0.3253,0.2292,0.0653,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056,0.0237,0.0204,0.005,0.0137,0.0164,0.0081,0.0139,0.0111,R
 98 | 0.0181,0.0146,0.0026,0.0141,0.0421,0.0473,0.0361,0.0741,0.1398,0.1045,0.0904,0.0671,0.0997,0.1056,0.0346,0.1231,0.1626,0.3652,0.3262,0.2995,0.2109,0.2104,0.2085,0.2282,0.0747,0.1969,0.4086,0.6385,0.797,0.7508,0.5517,0.2214,0.4672,0.4479,0.2297,0.3235,0.448,0.5581,0.652,0.5354,0.2478,0.2268,0.1788,0.0898,0.0536,0.0374,0.099,0.0956,0.0317,0.0142,0.0076,0.0223,0.0255,0.0145,0.0233,0.0041,0.0018,0.0048,0.0089,0.0085,R
 99 | 0.0491,0.0279,0.0592,0.127,0.1772,0.1908,0.2217,0.0768,0.1246,0.2028,0.0947,0.2497,0.2209,0.3195,0.334,0.3323,0.278,0.2975,0.2948,0.1729,0.3264,0.3834,0.3523,0.541,0.5228,0.4475,0.534,0.5323,0.3907,0.3456,0.4091,0.4639,0.558,0.5727,0.6355,0.7563,0.6903,0.6176,0.5379,0.5622,0.6508,0.4797,0.3736,0.2804,0.1982,0.2438,0.1789,0.1706,0.0762,0.0238,0.0268,0.0081,0.0129,0.0161,0.0063,0.0119,0.0194,0.014,0.0332,0.0439,M
100 | 0.1313,0.2339,0.3059,0.4264,0.401,0.1791,0.1853,0.0055,0.1929,0.2231,0.2907,0.2259,0.3136,0.3302,0.366,0.3956,0.4386,0.467,0.5255,0.3735,0.2243,0.1973,0.4337,0.6532,0.507,0.2796,0.4163,0.595,0.5242,0.4178,0.3714,0.2375,0.0863,0.1437,0.2896,0.4577,0.3725,0.3372,0.3803,0.4181,0.3603,0.2711,0.1653,0.1951,0.2811,0.2246,0.1921,0.15,0.0665,0.0193,0.0156,0.0362,0.021,0.0154,0.018,0.0013,0.0106,0.0127,0.0178,0.0231,M
101 | 0.0201,0.0423,0.0554,0.0783,0.062,0.0871,0.1201,0.2707,0.1206,0.0279,0.2251,0.2615,0.177,0.3709,0.4533,0.5553,0.4616,0.3797,0.345,0.2665,0.2395,0.1127,0.2556,0.5169,0.3779,0.4082,0.5353,0.5116,0.4544,0.4258,0.3869,0.3939,0.4661,0.3974,0.2194,0.1816,0.1023,0.2108,0.3253,0.3697,0.2912,0.301,0.2563,0.1927,0.2062,0.1751,0.0841,0.1035,0.0641,0.0153,0.0081,0.0191,0.0182,0.016,0.029,0.009,0.0242,0.0224,0.019,0.0096,M
102 | 0.0629,0.1065,0.1526,0.1229,0.1437,0.119,0.0884,0.0907,0.2107,0.3597,0.5466,0.5205,0.5127,0.5395,0.6558,0.8705,0.9786,0.9335,0.7917,0.7383,0.6908,0.385,0.0671,0.0502,0.2717,0.2839,0.2234,0.1911,0.0408,0.2531,0.1979,0.1891,0.2433,0.1956,0.2667,0.134,0.1073,0.2023,0.1794,0.0227,0.1313,0.1775,0.1549,0.1626,0.0708,0.0129,0.0795,0.0762,0.0117,0.0061,0.0257,0.0089,0.0262,0.0108,0.0138,0.0187,0.023,0.0057,0.0113,0.0131,M
103 | 0.0335,0.0134,0.0696,0.118,0.0348,0.118,0.1948,0.1607,0.3036,0.4372,0.5533,0.5771,0.7022,0.7067,0.7367,0.7391,0.8622,0.9458,0.8782,0.7913,0.576,0.3061,0.0563,0.0239,0.2554,0.4862,0.5027,0.4402,0.2847,0.1797,0.356,0.3522,0.3321,0.3112,0.3638,0.0754,0.1834,0.182,0.1815,0.1593,0.0576,0.0954,0.1086,0.0812,0.0784,0.0487,0.0439,0.0586,0.037,0.0185,0.0302,0.0244,0.0232,0.0093,0.0159,0.0193,0.0032,0.0377,0.0126,0.0156,M
104 | 0.0587,0.121,0.1268,0.1498,0.1436,0.0561,0.0832,0.0672,0.1372,0.2352,0.3208,0.4257,0.5201,0.4914,0.595,0.7221,0.9039,0.9111,0.8723,0.7686,0.7326,0.5222,0.3097,0.3172,0.227,0.164,0.1746,0.1835,0.2048,0.1674,0.2767,0.3104,0.3399,0.4441,0.5046,0.2814,0.1681,0.2633,0.3198,0.1933,0.0934,0.0443,0.078,0.0722,0.0405,0.0553,0.1081,0.1139,0.0767,0.0265,0.0215,0.0331,0.0111,0.0088,0.0158,0.0122,0.0038,0.0101,0.0228,0.0124,M
105 | 0.0162,0.0253,0.0262,0.0386,0.0645,0.0472,0.1056,0.1388,0.0598,0.1334,0.2969,0.4754,0.5677,0.569,0.6421,0.7487,0.8999,1.0,0.969,0.9032,0.7685,0.6998,0.6644,0.5964,0.3711,0.0921,0.0481,0.0876,0.104,0.1714,0.3264,0.4612,0.3939,0.505,0.4833,0.3511,0.2319,0.4029,0.3676,0.151,0.0745,0.1395,0.1552,0.0377,0.0636,0.0443,0.0264,0.0223,0.0187,0.0077,0.0137,0.0071,0.0082,0.0232,0.0198,0.0074,0.0035,0.01,0.0048,0.0019,M
106 | 0.0307,0.0523,0.0653,0.0521,0.0611,0.0577,0.0665,0.0664,0.146,0.2792,0.3877,0.4992,0.4981,0.4972,0.5607,0.7339,0.823,0.9173,0.9975,0.9911,0.824,0.6498,0.598,0.4862,0.315,0.1543,0.0989,0.0284,0.1008,0.2636,0.2694,0.293,0.2925,0.3998,0.366,0.3172,0.4609,0.4374,0.182,0.3376,0.6202,0.4448,0.1863,0.142,0.0589,0.0576,0.0672,0.0269,0.0245,0.019,0.0063,0.0321,0.0189,0.0137,0.0277,0.0152,0.0052,0.0121,0.0124,0.0055,M
107 | 0.0116,0.0179,0.0449,0.1096,0.1913,0.0924,0.0761,0.1092,0.0757,0.1006,0.25,0.3988,0.3809,0.4753,0.6165,0.6464,0.8024,0.9208,0.9832,0.9634,0.8646,0.8325,0.8276,0.8007,0.6102,0.4853,0.4355,0.4307,0.4399,0.3833,0.3032,0.3035,0.3197,0.2292,0.2131,0.2347,0.3201,0.4455,0.3655,0.2715,0.1747,0.1781,0.2199,0.1056,0.0573,0.0307,0.0237,0.047,0.0102,0.0057,0.0031,0.0163,0.0099,0.0084,0.027,0.0277,0.0097,0.0054,0.0148,0.0092,M
108 | 0.0331,0.0423,0.0474,0.0818,0.0835,0.0756,0.0374,0.0961,0.0548,0.0193,0.0897,0.1734,0.1936,0.2803,0.3313,0.502,0.636,0.7096,0.8333,0.873,0.8073,0.7507,0.7526,0.7298,0.6177,0.4946,0.4531,0.4099,0.454,0.4124,0.3139,0.3194,0.3692,0.3776,0.4469,0.4777,0.4716,0.4664,0.3893,0.4255,0.4064,0.3712,0.3863,0.2802,0.1283,0.1117,0.1303,0.0787,0.0436,0.0224,0.0133,0.0078,0.0174,0.0176,0.0038,0.0129,0.0066,0.0044,0.0134,0.0092,M
109 | 0.0428,0.0555,0.0708,0.0618,0.1215,0.1524,0.1543,0.0391,0.061,0.0113,0.1255,0.2473,0.3011,0.3747,0.452,0.5392,0.6588,0.7113,0.7602,0.8672,0.8416,0.7974,0.8385,0.9317,0.8555,0.6162,0.4139,0.3269,0.3108,0.2554,0.3367,0.4465,0.5,0.5111,0.5194,0.4619,0.4234,0.4372,0.4277,0.4433,0.37,0.3324,0.2564,0.2527,0.2137,0.1789,0.101,0.0528,0.0453,0.0118,0.0009,0.0142,0.0179,0.0079,0.006,0.0131,0.0089,0.0084,0.0113,0.0049,M
110 | 0.0599,0.0474,0.0498,0.0387,0.1026,0.0773,0.0853,0.0447,0.1094,0.0351,0.1582,0.2023,0.2268,0.2829,0.3819,0.4665,0.6687,0.8647,0.9361,0.9367,0.9144,0.9162,0.9311,0.8604,0.7327,0.5763,0.4162,0.4113,0.4146,0.3149,0.2936,0.3169,0.3149,0.4132,0.3994,0.4195,0.4532,0.4419,0.4737,0.3431,0.3194,0.337,0.2493,0.265,0.1748,0.0932,0.053,0.0081,0.0342,0.0137,0.0028,0.0013,0.0005,0.0227,0.0209,0.0081,0.0117,0.0114,0.0112,0.01,M
111 | 0.0264,0.0071,0.0342,0.0793,0.1043,0.0783,0.1417,0.1176,0.0453,0.0945,0.1132,0.084,0.0717,0.1968,0.2633,0.4191,0.505,0.6711,0.7922,0.8381,0.8759,0.9422,1.0,0.9931,0.9575,0.8647,0.7215,0.5801,0.4964,0.4886,0.4079,0.2443,0.1768,0.2472,0.3518,0.3762,0.2909,0.2311,0.3168,0.3554,0.3741,0.4443,0.3261,0.1963,0.0864,0.1688,0.1991,0.1217,0.0628,0.0323,0.0253,0.0214,0.0262,0.0177,0.0037,0.0068,0.0121,0.0077,0.0078,0.0066,M
112 | 0.021,0.0121,0.0203,0.1036,0.1675,0.0418,0.0723,0.0828,0.0494,0.0686,0.1125,0.1741,0.271,0.3087,0.3575,0.4998,0.6011,0.647,0.8067,0.9008,0.8906,0.9338,1.0,0.9102,0.8496,0.7867,0.7688,0.7718,0.6268,0.4301,0.2077,0.1198,0.166,0.2618,0.3862,0.3958,0.3248,0.2302,0.325,0.4022,0.4344,0.4008,0.337,0.2518,0.2101,0.1181,0.115,0.055,0.0293,0.0183,0.0104,0.0117,0.0101,0.0061,0.0031,0.0099,0.008,0.0107,0.0161,0.0133,M
113 | 0.053,0.0885,0.1997,0.2604,0.3225,0.2247,0.0617,0.2287,0.095,0.074,0.161,0.2226,0.2703,0.3365,0.4266,0.4144,0.5655,0.6921,0.8547,0.9234,0.9171,1.0,0.9532,0.9101,0.8337,0.7053,0.6534,0.4483,0.246,0.202,0.1446,0.0994,0.151,0.2392,0.4434,0.5023,0.4441,0.4571,0.3927,0.29,0.3408,0.499,0.3632,0.1387,0.18,0.1299,0.0523,0.0817,0.0469,0.0114,0.0299,0.0244,0.0199,0.0257,0.0082,0.0151,0.0171,0.0146,0.0134,0.0056,M
114 | 0.0454,0.0472,0.0697,0.1021,0.1397,0.1493,0.1487,0.0771,0.1171,0.1675,0.2799,0.3323,0.4012,0.4296,0.535,0.5411,0.687,0.8045,0.9194,0.9169,1.0,0.9972,0.9093,0.7918,0.6705,0.5324,0.3572,0.2484,0.3161,0.3775,0.3138,0.1713,0.2937,0.5234,0.5926,0.5437,0.4516,0.3379,0.3215,0.2178,0.1674,0.2634,0.298,0.2037,0.1155,0.0919,0.0882,0.0228,0.038,0.0142,0.0137,0.012,0.0042,0.0238,0.0129,0.0084,0.0218,0.0321,0.0154,0.0053,M
115 | 0.0283,0.0599,0.0656,0.0229,0.0839,0.1673,0.1154,0.1098,0.137,0.1767,0.1995,0.2869,0.3275,0.3769,0.4169,0.5036,0.618,0.8025,0.9333,0.9399,0.9275,0.945,0.8328,0.7773,0.7007,0.6154,0.581,0.4454,0.3707,0.2891,0.2185,0.1711,0.3578,0.3947,0.2867,0.2401,0.3619,0.3314,0.3763,0.4767,0.4059,0.3661,0.232,0.145,0.1017,0.1111,0.0655,0.0271,0.0244,0.0179,0.0109,0.0147,0.017,0.0158,0.0046,0.0073,0.0054,0.0033,0.0045,0.0079,M
116 | 0.0114,0.0222,0.0269,0.0384,0.1217,0.2062,0.1489,0.0929,0.135,0.1799,0.2486,0.2973,0.3672,0.4394,0.5258,0.6755,0.7402,0.8284,0.9033,0.9584,1.0,0.9982,0.8899,0.7493,0.6367,0.6744,0.7207,0.6821,0.5512,0.4789,0.3924,0.2533,0.1089,0.139,0.2551,0.3301,0.2818,0.2142,0.2266,0.2142,0.2354,0.2871,0.2596,0.1925,0.1256,0.1003,0.0951,0.121,0.0728,0.0174,0.0213,0.0269,0.0152,0.0257,0.0097,0.0041,0.005,0.0145,0.0103,0.0025,M
117 | 0.0414,0.0436,0.0447,0.0844,0.0419,0.1215,0.2002,0.1516,0.0818,0.1975,0.2309,0.3025,0.3938,0.505,0.5872,0.661,0.7417,0.8006,0.8456,0.7939,0.8804,0.8384,0.7852,0.8479,0.7434,0.6433,0.5514,0.3519,0.3168,0.3346,0.2056,0.1032,0.3168,0.404,0.4282,0.4538,0.3704,0.3741,0.3839,0.3494,0.438,0.4265,0.2854,0.2808,0.2395,0.0369,0.0805,0.0541,0.0177,0.0065,0.0222,0.0045,0.0136,0.0113,0.0053,0.0165,0.0141,0.0077,0.0246,0.0198,M
118 | 0.0094,0.0333,0.0306,0.0376,0.1296,0.1795,0.1909,0.1692,0.187,0.1725,0.2228,0.3106,0.4144,0.5157,0.5369,0.5107,0.6441,0.7326,0.8164,0.8856,0.9891,1.0,0.875,0.8631,0.9074,0.8674,0.775,0.66,0.5615,0.4016,0.2331,0.1164,0.1095,0.0431,0.0619,0.1956,0.212,0.3242,0.4102,0.2939,0.1911,0.1702,0.101,0.1512,0.1427,0.1097,0.1173,0.0972,0.0703,0.0281,0.0216,0.0153,0.0112,0.0241,0.0164,0.0055,0.0078,0.0055,0.0091,0.0067,M
119 | 0.0228,0.0106,0.013,0.0842,0.1117,0.1506,0.1776,0.0997,0.1428,0.2227,0.2621,0.3109,0.2859,0.3316,0.3755,0.4499,0.4765,0.6254,0.7304,0.8702,0.9349,0.9614,0.9126,0.9443,1.0,0.9455,0.8815,0.752,0.7068,0.5986,0.3857,0.251,0.2162,0.0968,0.1323,0.1344,0.225,0.3244,0.3939,0.3806,0.3258,0.3654,0.2983,0.1779,0.1535,0.1199,0.0959,0.0765,0.0649,0.0313,0.0185,0.0098,0.0178,0.0077,0.0074,0.0095,0.0055,0.0045,0.0063,0.0039,M
120 | 0.0363,0.0478,0.0298,0.021,0.1409,0.1916,0.1349,0.1613,0.1703,0.1444,0.1989,0.2154,0.2863,0.357,0.398,0.4359,0.5334,0.6304,0.6995,0.7435,0.8379,0.8641,0.9014,0.9432,0.9536,1.0,0.9547,0.9745,0.8962,0.7196,0.5462,0.3156,0.2525,0.1969,0.2189,0.1533,0.0711,0.1498,0.1755,0.2276,0.1322,0.1056,0.1973,0.1692,0.1881,0.1177,0.0779,0.0495,0.0492,0.0194,0.025,0.0115,0.019,0.0055,0.0096,0.005,0.0066,0.0114,0.0073,0.0033,M
121 | 0.0261,0.0266,0.0223,0.0749,0.1364,0.1513,0.1316,0.1654,0.1864,0.2013,0.289,0.365,0.351,0.3495,0.4325,0.5398,0.6237,0.6876,0.7329,0.8107,0.8396,0.8632,0.8747,0.9607,0.9716,0.9121,0.8576,0.8798,0.772,0.5711,0.4264,0.286,0.3114,0.2066,0.1165,0.0185,0.1302,0.248,0.1637,0.1103,0.2144,0.2033,0.1887,0.137,0.1376,0.0307,0.0373,0.0606,0.0399,0.0169,0.0135,0.0222,0.0175,0.0127,0.0022,0.0124,0.0054,0.0021,0.0028,0.0023,M
122 | 0.0346,0.0509,0.0079,0.0243,0.0432,0.0735,0.0938,0.1134,0.1228,0.1508,0.1809,0.239,0.2947,0.2866,0.401,0.5325,0.5486,0.5823,0.6041,0.6749,0.7084,0.789,0.9284,0.9781,0.9738,1.0,0.9702,0.9956,0.8235,0.602,0.5342,0.4867,0.3526,0.1566,0.0946,0.1613,0.2824,0.339,0.3019,0.2945,0.2978,0.2676,0.2055,0.2069,0.1625,0.1216,0.1013,0.0744,0.0386,0.005,0.0146,0.004,0.0122,0.0107,0.0112,0.0102,0.0052,0.0024,0.0079,0.0031,M
123 | 0.0162,0.0041,0.0239,0.0441,0.063,0.0921,0.1368,0.1078,0.1552,0.1779,0.2164,0.2568,0.3089,0.3829,0.4393,0.5335,0.5996,0.6728,0.7309,0.8092,0.8941,0.9668,1.0,0.9893,0.9376,0.8991,0.9184,0.9128,0.7811,0.6018,0.3765,0.33,0.228,0.0212,0.1117,0.1788,0.2373,0.2843,0.2241,0.2715,0.3363,0.2546,0.1867,0.216,0.1278,0.0768,0.107,0.0946,0.0636,0.0227,0.0128,0.0173,0.0135,0.0114,0.0062,0.0157,0.0088,0.0036,0.0053,0.003,M
124 | 0.0249,0.0119,0.0277,0.076,0.1218,0.1538,0.1192,0.1229,0.2119,0.2531,0.2855,0.2961,0.3341,0.4287,0.5205,0.6087,0.7236,0.7577,0.7726,0.8098,0.8995,0.9247,0.9365,0.9853,0.9776,1.0,0.9896,0.9076,0.7306,0.5758,0.4469,0.3719,0.2079,0.0955,0.0488,0.1406,0.2554,0.2054,0.1614,0.2232,0.1773,0.2293,0.2521,0.1464,0.0673,0.0965,0.1492,0.1128,0.0463,0.0193,0.014,0.0027,0.0068,0.015,0.0012,0.0133,0.0048,0.0244,0.0077,0.0074,M
125 | 0.027,0.0163,0.0341,0.0247,0.0822,0.1256,0.1323,0.1584,0.2017,0.2122,0.221,0.2399,0.2964,0.4061,0.5095,0.5512,0.6613,0.6804,0.652,0.6788,0.7811,0.8369,0.8969,0.9856,1.0,0.9395,0.8917,0.8105,0.6828,0.5572,0.4301,0.3339,0.2035,0.0798,0.0809,0.1525,0.2626,0.2456,0.198,0.2412,0.2409,0.1901,0.2077,0.1767,0.1119,0.0779,0.1344,0.096,0.0598,0.033,0.0197,0.0189,0.0204,0.0085,0.0043,0.0092,0.0138,0.0094,0.0105,0.0093,M
126 | 0.0388,0.0324,0.0688,0.0898,0.1267,0.1515,0.2134,0.2613,0.2832,0.2718,0.3645,0.3934,0.3843,0.4677,0.5364,0.4823,0.4835,0.5862,0.7579,0.6997,0.6918,0.8633,0.9107,0.9346,0.7884,0.8585,0.9261,0.708,0.5779,0.5215,0.4505,0.3129,0.1448,0.1046,0.182,0.1519,0.1017,0.1438,0.1986,0.2039,0.2778,0.2879,0.1331,0.114,0.131,0.1433,0.0624,0.01,0.0098,0.0131,0.0152,0.0255,0.0071,0.0263,0.0079,0.0111,0.0107,0.0068,0.0097,0.0067,M
127 | 0.0228,0.0853,0.1,0.0428,0.1117,0.1651,0.1597,0.2116,0.3295,0.3517,0.333,0.3643,0.402,0.4731,0.5196,0.6573,0.8426,0.8476,0.8344,0.8453,0.7999,0.8537,0.9642,1.0,0.9357,0.9409,0.907,0.7104,0.632,0.5667,0.3501,0.2447,0.1698,0.329,0.3674,0.2331,0.2413,0.2556,0.1892,0.194,0.3074,0.2785,0.0308,0.1238,0.1854,0.1753,0.1079,0.0728,0.0242,0.0191,0.0159,0.0172,0.0191,0.026,0.014,0.0125,0.0116,0.0093,0.0012,0.0036,M
128 | 0.0715,0.0849,0.0587,0.0218,0.0862,0.1801,0.1916,0.1896,0.296,0.4186,0.4867,0.5249,0.5959,0.6855,0.8573,0.9718,0.8693,0.8711,0.8954,0.9922,0.898,0.8158,0.8373,0.7541,0.5893,0.5488,0.5643,0.5406,0.4783,0.4439,0.3698,0.2574,0.1478,0.1743,0.1229,0.1588,0.1803,0.1436,0.1667,0.263,0.2234,0.1239,0.0869,0.2092,0.1499,0.0676,0.0899,0.0927,0.0658,0.0086,0.0216,0.0153,0.0121,0.0096,0.0196,0.0042,0.0066,0.0099,0.0083,0.0124,M
129 | 0.0209,0.0261,0.012,0.0768,0.1064,0.168,0.3016,0.346,0.3314,0.4125,0.3943,0.1334,0.4622,0.997,0.9137,0.8292,0.6994,0.7825,0.8789,0.8501,0.892,0.9473,1.0,0.8975,0.7806,0.8321,0.6502,0.4548,0.4732,0.3391,0.2747,0.0978,0.0477,0.1403,0.1834,0.2148,0.1271,0.1912,0.3391,0.3444,0.2369,0.1195,0.2665,0.2587,0.1393,0.1083,0.1383,0.1321,0.1069,0.0325,0.0316,0.0057,0.0159,0.0085,0.0372,0.0101,0.0127,0.0288,0.0129,0.0023,M
130 | 0.0374,0.0586,0.0628,0.0534,0.0255,0.1422,0.2072,0.2734,0.307,0.2597,0.3483,0.3999,0.4574,0.595,0.7924,0.8272,0.8087,0.8977,0.9828,0.8982,0.889,0.9367,0.9122,0.7936,0.6718,0.6318,0.4865,0.3388,0.4832,0.3822,0.3075,0.1267,0.0743,0.151,0.1906,0.1817,0.1709,0.0946,0.2829,0.3006,0.1602,0.1483,0.2875,0.2047,0.1064,0.1395,0.1065,0.0527,0.0395,0.0183,0.0353,0.0118,0.0063,0.0237,0.0032,0.0087,0.0124,0.0113,0.0098,0.0126,M
131 | 0.1371,0.1226,0.1385,0.1484,0.1776,0.1428,0.1773,0.2161,0.163,0.2067,0.4257,0.5484,0.7131,0.7003,0.6777,0.7939,0.9382,0.8925,0.9146,0.7832,0.796,0.7983,0.7716,0.6615,0.486,0.5572,0.4697,0.564,0.4517,0.3369,0.2684,0.2339,0.3052,0.3016,0.2753,0.1041,0.1757,0.3156,0.3603,0.2736,0.1301,0.2458,0.3404,0.1753,0.0679,0.1062,0.0643,0.0532,0.0531,0.0272,0.0171,0.0118,0.0129,0.0344,0.0065,0.0067,0.0022,0.0079,0.0146,0.0051,M
132 | 0.0443,0.0446,0.0235,0.1008,0.2252,0.2611,0.2061,0.1668,0.1801,0.3083,0.3794,0.5364,0.6173,0.7842,0.8392,0.9016,1.0,0.8911,0.8753,0.7886,0.7156,0.7581,0.6372,0.321,0.2076,0.2279,0.3309,0.2847,0.1949,0.1671,0.1025,0.1362,0.2212,0.1124,0.1677,0.1039,0.2562,0.2624,0.2236,0.118,0.1103,0.2831,0.2385,0.0255,0.1967,0.1483,0.0434,0.0627,0.0513,0.0473,0.0248,0.0274,0.0205,0.0141,0.0185,0.0055,0.0045,0.0115,0.0152,0.01,M
133 | 0.115,0.1163,0.0866,0.0358,0.0232,0.1267,0.2417,0.2661,0.4346,0.5378,0.3816,0.0991,0.0616,0.1795,0.3907,0.3602,0.3041,0.2428,0.406,0.8395,0.9777,0.468,0.061,0.2143,0.1348,0.2854,0.1617,0.2649,0.4565,0.6502,0.2848,0.3296,0.537,0.6627,0.8626,0.8547,0.7848,0.9016,0.8827,0.6086,0.281,0.0906,0.1177,0.2694,0.5214,0.4232,0.234,0.1928,0.1092,0.0507,0.0228,0.0099,0.0065,0.0085,0.0166,0.011,0.019,0.0141,0.0068,0.0086,M
134 | 0.0968,0.0821,0.0629,0.0608,0.0617,0.1207,0.0944,0.4223,0.5744,0.5025,0.3488,0.17,0.2076,0.3087,0.4224,0.5312,0.2436,0.1884,0.1908,0.8321,1.0,0.4076,0.096,0.1928,0.2419,0.379,0.2893,0.3451,0.3777,0.5213,0.2316,0.3335,0.4781,0.6116,0.6705,0.7375,0.7356,0.7792,0.6788,0.5259,0.2762,0.1545,0.2019,0.2231,0.4221,0.3067,0.1329,0.1349,0.1057,0.0499,0.0206,0.0073,0.0081,0.0303,0.019,0.0212,0.0126,0.0201,0.021,0.0041,M
135 | 0.079,0.0707,0.0352,0.166,0.133,0.0226,0.0771,0.2678,0.5664,0.6609,0.5002,0.2583,0.165,0.4347,0.4515,0.4579,0.3366,0.4,0.5325,0.901,0.9939,0.3689,0.1012,0.0248,0.2318,0.3981,0.2259,0.5247,0.6898,0.8316,0.4326,0.3741,0.5756,0.8043,0.7963,0.7174,0.7056,0.8148,0.7601,0.6034,0.4554,0.4729,0.4478,0.3722,0.4693,0.3839,0.0768,0.1467,0.0777,0.0469,0.0193,0.0298,0.039,0.0294,0.0175,0.0249,0.0141,0.0073,0.0025,0.0101,M
136 | 0.1083,0.107,0.0257,0.0837,0.0748,0.1125,0.3322,0.459,0.5526,0.5966,0.5304,0.2251,0.2402,0.2689,0.6646,0.6632,0.1674,0.0837,0.4331,0.8718,0.7992,0.3712,0.1703,0.1611,0.2086,0.2847,0.2211,0.6134,0.5807,0.6925,0.3825,0.4303,0.7791,0.8703,1.0,0.9212,0.9386,0.9303,0.7314,0.4791,0.2087,0.2016,0.1669,0.2872,0.4374,0.3097,0.1578,0.0553,0.0334,0.0209,0.0172,0.018,0.011,0.0234,0.0276,0.0032,0.0084,0.0122,0.0082,0.0143,M
137 | 0.0094,0.0611,0.1136,0.1203,0.0403,0.1227,0.2495,0.4566,0.6587,0.5079,0.335,0.0834,0.3004,0.3957,0.3769,0.3828,0.1247,0.1363,0.2678,0.9188,0.9779,0.3236,0.1944,0.1874,0.0885,0.3443,0.2953,0.5908,0.4564,0.7334,0.1969,0.279,0.6212,0.8681,0.8621,0.938,0.8327,0.948,0.6721,0.4436,0.5163,0.3809,0.1557,0.1449,0.2662,0.1806,0.1699,0.2559,0.1129,0.0201,0.048,0.0234,0.0175,0.0352,0.0158,0.0326,0.0201,0.0168,0.0245,0.0154,M
138 | 0.1088,0.1278,0.0926,0.1234,0.1276,0.1731,0.1948,0.4262,0.6828,0.5761,0.4733,0.2362,0.1023,0.2904,0.4713,0.4659,0.1415,0.0849,0.3257,0.9007,0.9312,0.4856,0.1346,0.1604,0.2737,0.5609,0.3654,0.6139,0.547,0.8474,0.5638,0.5443,0.5086,0.6253,0.8497,0.8406,0.842,0.9136,0.7713,0.4882,0.3724,0.4469,0.4586,0.4491,0.5616,0.4305,0.0945,0.0794,0.0274,0.0154,0.014,0.0455,0.0213,0.0082,0.0124,0.0167,0.0103,0.0205,0.0178,0.0187,M
139 | 0.043,0.0902,0.0833,0.0813,0.0165,0.0277,0.0569,0.2057,0.3887,0.7106,0.7342,0.5033,0.3,0.1951,0.2767,0.3737,0.2507,0.2507,0.3292,0.4871,0.6527,0.8454,0.9739,1.0,0.6665,0.5323,0.4024,0.3444,0.4239,0.4182,0.4393,0.1162,0.4336,0.6553,0.6172,0.4373,0.4118,0.3641,0.4572,0.4367,0.2964,0.4312,0.4155,0.1824,0.1487,0.0138,0.1164,0.2052,0.1069,0.0199,0.0208,0.0176,0.0197,0.021,0.0141,0.0049,0.0027,0.0162,0.0059,0.0021,M
140 | 0.0731,0.1249,0.1665,0.1496,0.1443,0.277,0.2555,0.1712,0.0466,0.1114,0.1739,0.316,0.3249,0.2164,0.2031,0.258,0.1796,0.2422,0.3609,0.181,0.2604,0.6572,0.9734,0.9757,0.8079,0.6521,0.4915,0.5363,0.7649,0.525,0.5101,0.4219,0.416,0.1906,0.0223,0.4219,0.5496,0.2483,0.2034,0.2729,0.2837,0.4463,0.3178,0.0807,0.1192,0.2134,0.3241,0.2945,0.1474,0.0211,0.0361,0.0444,0.023,0.029,0.0141,0.0161,0.0177,0.0194,0.0207,0.0057,M
141 | 0.0164,0.0627,0.0738,0.0608,0.0233,0.1048,0.1338,0.0644,0.1522,0.078,0.1791,0.2681,0.1788,0.1039,0.198,0.3234,0.3748,0.2586,0.368,0.3508,0.5606,0.5231,0.5469,0.6954,0.6352,0.6757,0.8499,0.8025,0.6563,0.8591,0.6655,0.5369,0.3118,0.3763,0.2801,0.0875,0.3319,0.4237,0.1801,0.3743,0.4627,0.1614,0.2494,0.3202,0.2265,0.1146,0.0476,0.0943,0.0824,0.0171,0.0244,0.0258,0.0143,0.0226,0.0187,0.0185,0.011,0.0094,0.0078,0.0112,M
142 | 0.0412,0.1135,0.0518,0.0232,0.0646,0.1124,0.1787,0.2407,0.2682,0.2058,0.1546,0.2671,0.3141,0.2904,0.3531,0.5079,0.4639,0.1859,0.4474,0.4079,0.54,0.4786,0.4332,0.6113,0.5091,0.4606,0.7243,0.8987,0.8826,0.9201,0.8005,0.6033,0.212,0.2866,0.4033,0.2803,0.3087,0.355,0.2545,0.1432,0.5869,0.6431,0.5826,0.4286,0.4894,0.5777,0.4315,0.264,0.1794,0.0772,0.0798,0.0376,0.0143,0.0272,0.0127,0.0166,0.0095,0.0225,0.0098,0.0085,M
143 | 0.0707,0.1252,0.1447,0.1644,0.1693,0.0844,0.0715,0.0947,0.1583,0.1247,0.234,0.1764,0.2284,0.3115,0.4725,0.5543,0.5386,0.3746,0.4583,0.5961,0.7464,0.7644,0.5711,0.6257,0.6695,0.7131,0.7567,0.8077,0.8477,0.9289,0.9513,0.7995,0.4362,0.4048,0.4952,0.1712,0.3652,0.3763,0.2841,0.0427,0.5331,0.6952,0.4288,0.3063,0.5835,0.5692,0.263,0.1196,0.0983,0.0374,0.0291,0.0156,0.0197,0.0135,0.0127,0.0138,0.0133,0.0131,0.0154,0.0218,M
144 | 0.0526,0.0563,0.1219,0.1206,0.0246,0.1022,0.0539,0.0439,0.2291,0.1632,0.2544,0.2807,0.3011,0.3361,0.3024,0.2285,0.291,0.1316,0.1151,0.3404,0.5562,0.6379,0.6553,0.7384,0.6534,0.5423,0.6877,0.7325,0.7726,0.8229,0.8787,0.9108,0.6705,0.6092,0.7505,0.4775,0.1666,0.3749,0.3776,0.2106,0.5886,0.5628,0.2577,0.5245,0.6149,0.5123,0.3385,0.1499,0.0546,0.027,0.038,0.0339,0.0149,0.0335,0.0376,0.0174,0.0132,0.0103,0.0364,0.0208,M
145 | 0.0516,0.0944,0.0622,0.0415,0.0995,0.2431,0.1777,0.2018,0.2611,0.1294,0.2646,0.2778,0.4432,0.3672,0.2035,0.2764,0.3252,0.1536,0.2784,0.3508,0.5187,0.7052,0.7143,0.6814,0.51,0.5308,0.6131,0.8388,0.9031,0.8607,0.9656,0.9168,0.7132,0.6898,0.731,0.4134,0.158,0.1819,0.1381,0.296,0.6935,0.8246,0.5351,0.4403,0.6448,0.6214,0.3016,0.1379,0.0364,0.0355,0.0456,0.0432,0.0274,0.0152,0.012,0.0129,0.002,0.0109,0.0074,0.0078,M
146 | 0.0299,0.0688,0.0992,0.1021,0.08,0.0629,0.013,0.0813,0.1761,0.0998,0.0523,0.0904,0.2655,0.3099,0.352,0.3892,0.3962,0.2449,0.2355,0.3045,0.3112,0.4698,0.5534,0.4532,0.4464,0.467,0.4621,0.6988,0.7626,0.7025,0.7382,0.7446,0.7927,0.5227,0.3967,0.3042,0.1309,0.2408,0.178,0.1598,0.5657,0.6443,0.4241,0.4567,0.576,0.5293,0.3287,0.1283,0.0698,0.0334,0.0342,0.0459,0.0277,0.0172,0.0087,0.0046,0.0203,0.013,0.0115,0.0015,M
147 | 0.0721,0.1574,0.1112,0.1085,0.0666,0.18,0.1108,0.2794,0.1408,0.0795,0.2534,0.392,0.3375,0.161,0.1889,0.3308,0.2282,0.2177,0.1853,0.5167,0.5342,0.6298,0.8437,0.6756,0.5825,0.6141,0.8809,0.8375,0.3869,0.5051,0.5455,0.4241,0.1534,0.495,0.6983,0.7109,0.5647,0.487,0.5515,0.4433,0.525,0.6075,0.5251,0.1359,0.4268,0.4442,0.2193,0.09,0.12,0.0628,0.0234,0.0309,0.0127,0.0082,0.0281,0.0117,0.0092,0.0147,0.0157,0.0129,M
148 | 0.1021,0.083,0.0577,0.0627,0.0635,0.1328,0.0988,0.1787,0.1199,0.1369,0.2509,0.2631,0.2796,0.2977,0.3823,0.3129,0.3956,0.2093,0.3218,0.3345,0.3184,0.2887,0.361,0.2566,0.4106,0.4591,0.4722,0.7278,0.7591,0.6579,0.7514,0.6666,0.4903,0.5962,0.6552,0.4014,0.1188,0.3245,0.3107,0.1354,0.5109,0.7988,0.7517,0.5508,0.5858,0.7292,0.5522,0.3339,0.1608,0.0475,0.1004,0.0709,0.0317,0.0309,0.0252,0.0087,0.0177,0.0214,0.0227,0.0106,M
149 | 0.0654,0.0649,0.0737,0.1132,0.2482,0.1257,0.1797,0.0989,0.246,0.3422,0.2128,0.1377,0.4032,0.5684,0.2398,0.4331,0.5954,0.5772,0.8176,0.8835,0.5248,0.6373,0.8375,0.6699,0.7756,0.875,0.83,0.6896,0.3372,0.6405,0.7138,0.8202,0.6657,0.5254,0.296,0.0704,0.097,0.3941,0.6028,0.3521,0.3924,0.4808,0.4602,0.4164,0.5438,0.5649,0.3195,0.2484,0.1299,0.0825,0.0243,0.021,0.0361,0.0239,0.0447,0.0394,0.0355,0.044,0.0243,0.0098,M
150 | 0.0712,0.0901,0.1276,0.1497,0.1284,0.1165,0.1285,0.1684,0.183,0.2127,0.2891,0.3985,0.4576,0.5821,0.5027,0.193,0.2579,0.3177,0.2745,0.6186,0.8958,0.7442,0.5188,0.2811,0.1773,0.6607,0.7576,0.5122,0.4701,0.5479,0.4347,0.1276,0.0846,0.0927,0.0313,0.0998,0.1781,0.1586,0.3001,0.2208,0.1455,0.2895,0.3203,0.1414,0.0629,0.0734,0.0805,0.0608,0.0565,0.0286,0.0154,0.0154,0.0156,0.0054,0.003,0.0048,0.0087,0.0101,0.0095,0.0068,M
151 | 0.0207,0.0535,0.0334,0.0818,0.074,0.0324,0.0918,0.107,0.1553,0.1234,0.1796,0.1787,0.1247,0.2577,0.337,0.399,0.1647,0.2266,0.3219,0.5356,0.8159,1.0,0.8701,0.6889,0.6299,0.5738,0.5707,0.5976,0.4301,0.2058,0.1,0.2247,0.2308,0.3977,0.3317,0.1726,0.1429,0.2168,0.1967,0.214,0.3674,0.2023,0.0778,0.0925,0.2388,0.34,0.2594,0.1102,0.0911,0.0462,0.0171,0.0033,0.005,0.019,0.0103,0.0121,0.0042,0.009,0.007,0.0099,M
152 | 0.0209,0.0278,0.0115,0.0445,0.0427,0.0766,0.1458,0.143,0.1894,0.1853,0.1748,0.1556,0.1476,0.1378,0.2584,0.3827,0.4784,0.536,0.6192,0.7912,0.9264,1.0,0.908,0.7435,0.5557,0.3172,0.1295,0.0598,0.2722,0.3616,0.3293,0.4855,0.3936,0.1845,0.0342,0.2489,0.3837,0.3514,0.2654,0.176,0.1599,0.0866,0.059,0.0813,0.0492,0.0417,0.0495,0.0367,0.0115,0.0118,0.0133,0.0096,0.0014,0.0049,0.0039,0.0029,0.0078,0.0047,0.0021,0.0011,M
153 | 0.0231,0.0315,0.017,0.0226,0.041,0.0116,0.0223,0.0805,0.2365,0.2461,0.2245,0.152,0.1732,0.3099,0.438,0.5595,0.682,0.6164,0.6803,0.8435,0.9921,1.0,0.7983,0.5426,0.3952,0.5179,0.565,0.3042,0.1881,0.396,0.2286,0.3544,0.4187,0.2398,0.1847,0.376,0.4331,0.3626,0.2519,0.187,0.1046,0.2339,0.1991,0.11,0.0684,0.0303,0.0674,0.0785,0.0455,0.0246,0.0151,0.0125,0.0036,0.0123,0.0043,0.0114,0.0052,0.0091,0.0008,0.0092,M
154 | 0.0131,0.0201,0.0045,0.0217,0.023,0.0481,0.0742,0.0333,0.1369,0.2079,0.2295,0.199,0.1184,0.1891,0.2949,0.5343,0.685,0.7923,0.822,0.729,0.7352,0.7918,0.8057,0.4898,0.1934,0.2924,0.6255,0.8546,0.8966,0.7821,0.5168,0.484,0.4038,0.3411,0.2849,0.2353,0.2699,0.4442,0.4323,0.3314,0.1195,0.1669,0.3702,0.3072,0.0945,0.1545,0.1394,0.0772,0.0615,0.023,0.0111,0.0168,0.0086,0.0045,0.0062,0.0065,0.003,0.0066,0.0029,0.0053,M
155 | 0.0233,0.0394,0.0416,0.0547,0.0993,0.1515,0.1674,0.1513,0.1723,0.2078,0.1239,0.0236,0.1771,0.3115,0.499,0.6707,0.7655,0.8485,0.9805,1.0,1.0,0.9992,0.9067,0.6803,0.5103,0.4716,0.498,0.6196,0.7171,0.6316,0.3554,0.2897,0.4316,0.3791,0.2421,0.0944,0.0351,0.0844,0.0436,0.113,0.2045,0.1937,0.0834,0.1502,0.1675,0.1058,0.1111,0.0849,0.0596,0.0201,0.0071,0.0104,0.0062,0.0026,0.0025,0.0061,0.0038,0.0101,0.0078,0.0006,M
156 | 0.0117,0.0069,0.0279,0.0583,0.0915,0.1267,0.1577,0.1927,0.2361,0.2169,0.118,0.0754,0.2782,0.3758,0.5093,0.6592,0.7071,0.7532,0.8357,0.8593,0.9615,0.9838,0.8705,0.6403,0.5067,0.5395,0.6934,0.8487,0.8213,0.5962,0.295,0.2758,0.2885,0.1893,0.1446,0.0955,0.0888,0.0836,0.0894,0.1547,0.2318,0.2225,0.1035,0.1721,0.2017,0.1787,0.1112,0.0398,0.0305,0.0084,0.0039,0.0053,0.0029,0.002,0.0013,0.0029,0.002,0.0062,0.0026,0.0052,M
157 | 0.0211,0.0128,0.0015,0.045,0.0711,0.1563,0.1518,0.1206,0.1666,0.1345,0.0785,0.0367,0.1227,0.2614,0.428,0.6122,0.7435,0.813,0.9006,0.9603,0.9162,0.914,0.7851,0.5134,0.3439,0.329,0.2571,0.3685,0.5765,0.619,0.4613,0.3615,0.4434,0.3864,0.3093,0.2138,0.1112,0.1386,0.1523,0.0996,0.1644,0.1902,0.1313,0.1776,0.2,0.0765,0.0727,0.0749,0.0449,0.0134,0.0174,0.0117,0.0023,0.0047,0.0049,0.0031,0.0024,0.0039,0.0051,0.0015,M
158 | 0.0047,0.0059,0.008,0.0554,0.0883,0.1278,0.1674,0.1373,0.2922,0.3469,0.3265,0.3263,0.2301,0.1253,0.2102,0.2401,0.1928,0.1673,0.1228,0.0902,0.1557,0.3291,0.5268,0.674,0.7906,0.8938,0.9395,0.9493,0.904,0.9151,0.8828,0.8086,0.718,0.672,0.6447,0.6879,0.6241,0.4936,0.4144,0.424,0.4546,0.4392,0.4323,0.4921,0.471,0.3196,0.2241,0.1806,0.099,0.0251,0.0129,0.0095,0.0126,0.0069,0.0039,0.0068,0.006,0.0045,0.0002,0.0029,M
159 | 0.0201,0.0178,0.0274,0.0232,0.0724,0.0833,0.1232,0.1298,0.2085,0.272,0.2188,0.3037,0.2959,0.2059,0.0906,0.161,0.18,0.218,0.2026,0.1506,0.0521,0.2143,0.4333,0.5943,0.6926,0.7576,0.8787,0.906,0.8528,0.9087,0.9657,0.9306,0.7774,0.6643,0.6604,0.6884,0.6938,0.5932,0.5774,0.6223,0.5841,0.4527,0.4911,0.5762,0.5013,0.4042,0.3123,0.2232,0.1085,0.0414,0.0253,0.0131,0.0049,0.0104,0.0102,0.0092,0.0083,0.002,0.0048,0.0036,M
160 | 0.0107,0.0453,0.0289,0.0713,0.1075,0.1019,0.1606,0.2119,0.3061,0.2936,0.3104,0.3431,0.2456,0.1887,0.1184,0.208,0.2736,0.3274,0.2344,0.126,0.0576,0.1241,0.3239,0.4357,0.5734,0.7825,0.9252,0.9349,0.9348,1.0,0.9308,0.8478,0.7605,0.704,0.7539,0.799,0.7673,0.5955,0.4731,0.484,0.434,0.3954,0.4837,0.5379,0.4485,0.2674,0.1541,0.1359,0.0941,0.0261,0.0079,0.0164,0.012,0.0113,0.0021,0.0097,0.0072,0.006,0.0017,0.0036,M
161 | 0.0235,0.022,0.0167,0.0516,0.0746,0.1121,0.1258,0.1717,0.3074,0.3199,0.2946,0.2484,0.251,0.1806,0.1413,0.3019,0.3635,0.3887,0.298,0.2219,0.1624,0.1343,0.2046,0.3791,0.5771,0.7545,0.8406,0.8547,0.9036,1.0,0.9646,0.7912,0.6412,0.5986,0.6835,0.7771,0.8084,0.7426,0.6295,0.5708,0.4433,0.3361,0.3795,0.495,0.4373,0.2404,0.1128,0.1654,0.0933,0.0225,0.0214,0.0221,0.0152,0.0083,0.0058,0.0023,0.0057,0.0052,0.0027,0.0021,M
162 | 0.0258,0.0433,0.0547,0.0681,0.0784,0.125,0.1296,0.1729,0.2794,0.2954,0.2506,0.2601,0.2249,0.2115,0.127,0.1193,0.1794,0.2185,0.1646,0.074,0.0625,0.2381,0.4824,0.6372,0.7531,0.8959,0.9941,0.9957,0.9328,0.9344,0.8854,0.769,0.6865,0.639,0.6378,0.6629,0.5983,0.4565,0.3129,0.4158,0.4325,0.4031,0.4201,0.4557,0.3955,0.2966,0.2095,0.1558,0.0884,0.0265,0.0121,0.0091,0.0062,0.0019,0.0045,0.0079,0.0031,0.0063,0.0048,0.005,M
163 | 0.0305,0.0363,0.0214,0.0227,0.0456,0.0665,0.0939,0.0972,0.2535,0.3127,0.2192,0.2621,0.2419,0.2179,0.1159,0.1237,0.0886,0.1755,0.1758,0.154,0.0512,0.1805,0.4039,0.5697,0.6577,0.7474,0.8543,0.9085,0.8668,0.8892,0.9065,0.8522,0.7204,0.62,0.6253,0.6848,0.7337,0.6281,0.5725,0.6119,0.5597,0.4965,0.5027,0.5772,0.5907,0.4803,0.3877,0.2779,0.1427,0.0424,0.0271,0.02,0.007,0.007,0.0086,0.0089,0.0074,0.0042,0.0055,0.0021,M
164 | 0.0217,0.0152,0.0346,0.0346,0.0484,0.0526,0.0773,0.0862,0.1451,0.211,0.2343,0.2087,0.1645,0.1689,0.165,0.1967,0.2934,0.3709,0.4309,0.4161,0.5116,0.6501,0.7717,0.8491,0.9104,0.8912,0.8189,0.6779,0.5368,0.5207,0.5651,0.5749,0.525,0.4255,0.333,0.2331,0.1451,0.1648,0.2694,0.373,0.4467,0.4133,0.3743,0.3021,0.2069,0.179,0.1689,0.1341,0.0769,0.0222,0.0205,0.0123,0.0067,0.0011,0.0026,0.0049,0.0029,0.0022,0.0022,0.0032,M
165 | 0.0072,0.0027,0.0089,0.0061,0.042,0.0865,0.1182,0.0999,0.1976,0.2318,0.2472,0.288,0.2126,0.0708,0.1194,0.2808,0.4221,0.5279,0.5857,0.6153,0.6753,0.7873,0.8974,0.9828,1.0,0.846,0.6055,0.3036,0.0144,0.2526,0.4335,0.4918,0.5409,0.5961,0.5248,0.3777,0.2369,0.172,0.1878,0.325,0.2575,0.2423,0.2706,0.2323,0.1724,0.1457,0.1175,0.0868,0.0392,0.0131,0.0092,0.0078,0.0071,0.0081,0.0034,0.0064,0.0037,0.0036,0.0012,0.0037,M
166 | 0.0163,0.0198,0.0202,0.0386,0.0752,0.1444,0.1487,0.1484,0.2442,0.2822,0.3691,0.375,0.3927,0.3308,0.1085,0.1139,0.3446,0.5441,0.647,0.7276,0.7894,0.8264,0.8697,0.7836,0.714,0.5698,0.2908,0.4636,0.6409,0.7405,0.8069,0.842,1.0,0.9536,0.6755,0.3905,0.1249,0.3629,0.6356,0.8116,0.7664,0.5417,0.2614,0.1723,0.2814,0.2764,0.1985,0.1502,0.1219,0.0493,0.0027,0.0077,0.0026,0.0031,0.0083,0.002,0.0084,0.0108,0.0083,0.0033,M
167 | 0.0221,0.0065,0.0164,0.0487,0.0519,0.0849,0.0812,0.1833,0.2228,0.181,0.2549,0.2984,0.2624,0.1893,0.0668,0.2666,0.4274,0.6291,0.7782,0.7686,0.8099,0.8493,0.944,0.945,0.9655,0.8045,0.4969,0.396,0.3856,0.5574,0.7309,0.8549,0.9425,0.8726,0.6673,0.4694,0.1546,0.1748,0.3607,0.5208,0.5177,0.3702,0.224,0.0816,0.0395,0.0785,0.1052,0.1034,0.0764,0.0216,0.0167,0.0089,0.0051,0.0015,0.0075,0.0058,0.0016,0.007,0.0074,0.0038,M
168 | 0.0411,0.0277,0.0604,0.0525,0.0489,0.0385,0.0611,0.1117,0.1237,0.23,0.137,0.1335,0.2137,0.1526,0.0775,0.1196,0.0903,0.0689,0.2071,0.2975,0.2836,0.3353,0.3622,0.3202,0.3452,0.3562,0.3892,0.6622,0.9254,1.0,0.8528,0.6297,0.525,0.4012,0.2901,0.2007,0.3356,0.4799,0.6147,0.6246,0.4973,0.3492,0.2662,0.3137,0.4282,0.4262,0.3511,0.2458,0.1259,0.0327,0.0181,0.0217,0.0038,0.0019,0.0065,0.0132,0.0108,0.005,0.0085,0.0044,M
169 | 0.0137,0.0297,0.0116,0.0082,0.0241,0.0253,0.0279,0.013,0.0489,0.0874,0.11,0.1084,0.1094,0.1023,0.0601,0.0906,0.1313,0.2758,0.366,0.5269,0.581,0.6181,0.5875,0.4639,0.5424,0.7367,0.9089,1.0,0.8247,0.5441,0.3349,0.0877,0.16,0.4169,0.6576,0.739,0.7963,0.7493,0.6795,0.4713,0.2355,0.1704,0.2728,0.4016,0.4125,0.347,0.2739,0.179,0.0922,0.0276,0.0169,0.0081,0.004,0.0025,0.0036,0.0058,0.0067,0.0035,0.0043,0.0033,M
170 | 0.0015,0.0186,0.0289,0.0195,0.0515,0.0817,0.1005,0.0124,0.1168,0.1476,0.2118,0.2575,0.2354,0.1334,0.0092,0.1951,0.3685,0.4646,0.5418,0.626,0.742,0.8257,0.8609,0.84,0.8949,0.9945,1.0,0.9649,0.8747,0.6257,0.2184,0.2945,0.3645,0.5012,0.7843,0.9361,0.8195,0.6207,0.4513,0.3004,0.2674,0.2241,0.3141,0.3693,0.2986,0.2226,0.0849,0.0359,0.0289,0.0122,0.0045,0.0108,0.0075,0.0089,0.0036,0.0029,0.0013,0.001,0.0032,0.0047,M
171 | 0.013,0.012,0.0436,0.0624,0.0428,0.0349,0.0384,0.0446,0.1318,0.1375,0.2026,0.2389,0.2112,0.1444,0.0742,0.1533,0.3052,0.4116,0.5466,0.5933,0.6663,0.7333,0.7136,0.7014,0.7758,0.9137,0.9964,1.0,0.8881,0.6585,0.2707,0.1746,0.2709,0.4853,0.7184,0.8209,0.7536,0.6496,0.4708,0.3482,0.3508,0.3181,0.3524,0.3659,0.2846,0.1714,0.0694,0.0303,0.0292,0.0116,0.0024,0.0084,0.01,0.0018,0.0035,0.0058,0.0011,0.0009,0.0033,0.0026,M
172 | 0.0134,0.0172,0.0178,0.0363,0.0444,0.0744,0.08,0.0456,0.0368,0.125,0.2405,0.2325,0.2523,0.1472,0.0669,0.11,0.2353,0.3282,0.4416,0.5167,0.6508,0.7793,0.7978,0.7786,0.8587,0.9321,0.9454,0.8645,0.722,0.485,0.1357,0.2951,0.4715,0.6036,0.8083,0.987,0.88,0.6411,0.4276,0.2702,0.2642,0.3342,0.4335,0.4542,0.396,0.2525,0.1084,0.0372,0.0286,0.0099,0.0046,0.0094,0.0048,0.0047,0.0016,0.0008,0.0042,0.0024,0.0027,0.0041,M
173 | 0.0179,0.0136,0.0408,0.0633,0.0596,0.0808,0.209,0.3465,0.5276,0.5965,0.6254,0.4507,0.3693,0.2864,0.1635,0.0422,0.1785,0.4394,0.695,0.8097,0.855,0.8717,0.8601,0.9201,0.8729,0.8084,0.8694,0.8411,0.5793,0.3754,0.3485,0.4639,0.6495,0.6901,0.5666,0.5188,0.506,0.3885,0.3762,0.3738,0.2605,0.1591,0.1875,0.2267,0.1577,0.1211,0.0883,0.085,0.0355,0.0219,0.0086,0.0123,0.006,0.0187,0.0111,0.0126,0.0081,0.0155,0.016,0.0085,M
174 | 0.018,0.0444,0.0476,0.0698,0.1615,0.0887,0.0596,0.1071,0.3175,0.2918,0.3273,0.3035,0.3033,0.2587,0.1682,0.1308,0.2803,0.4519,0.6641,0.7683,0.696,0.4393,0.2432,0.2886,0.4974,0.8172,1.0,0.9238,0.8519,0.7722,0.5772,0.519,0.6824,0.622,0.5054,0.3578,0.3809,0.3813,0.3359,0.2771,0.3648,0.3834,0.3453,0.2096,0.1031,0.0798,0.0701,0.0526,0.0241,0.0117,0.0122,0.0122,0.0114,0.0098,0.0027,0.0025,0.0026,0.005,0.0073,0.0022,M
175 | 0.0329,0.0216,0.0386,0.0627,0.1158,0.1482,0.2054,0.1605,0.2532,0.2672,0.3056,0.3161,0.2314,0.2067,0.1804,0.2808,0.4423,0.5947,0.6601,0.5844,0.4539,0.4789,0.5646,0.5281,0.7115,1.0,0.9564,0.609,0.5112,0.4,0.0482,0.1852,0.2186,0.1436,0.1757,0.1428,0.1644,0.3089,0.3648,0.4441,0.3859,0.2813,0.1238,0.0953,0.1201,0.0825,0.0618,0.0141,0.0108,0.0124,0.0104,0.0095,0.0151,0.0059,0.0015,0.0053,0.0016,0.0042,0.0053,0.0074,M
176 | 0.0191,0.0173,0.0291,0.0301,0.0463,0.069,0.0576,0.1103,0.2423,0.3134,0.4786,0.5239,0.4393,0.344,0.2869,0.3889,0.442,0.3892,0.4088,0.5006,0.7271,0.9385,1.0,0.9831,0.9932,0.9161,0.8237,0.6957,0.4536,0.3281,0.2522,0.3964,0.4154,0.3308,0.1445,0.1923,0.3208,0.3367,0.5683,0.5505,0.3231,0.0448,0.3131,0.3387,0.413,0.3639,0.2069,0.0859,0.06,0.0267,0.0125,0.004,0.0136,0.0137,0.0172,0.0132,0.011,0.0122,0.0114,0.0068,M
177 | 0.0294,0.0123,0.0117,0.0113,0.0497,0.0998,0.1326,0.1117,0.2984,0.3473,0.4231,0.5044,0.5237,0.4398,0.3236,0.2956,0.3286,0.3231,0.4528,0.6339,0.7044,0.8314,0.8449,0.8512,0.9138,0.9985,1.0,0.7544,0.4661,0.3924,0.3849,0.4674,0.4245,0.3095,0.0752,0.2885,0.4072,0.317,0.2863,0.2634,0.0541,0.1874,0.3459,0.4646,0.4366,0.2581,0.1319,0.0505,0.0112,0.0059,0.0041,0.0056,0.0104,0.0079,0.0014,0.0054,0.0015,0.0006,0.0081,0.0043,M
178 | 0.0635,0.0709,0.0453,0.0333,0.0185,0.126,0.1015,0.1918,0.3362,0.39,0.4674,0.5632,0.5506,0.4343,0.3052,0.3492,0.3975,0.3875,0.528,0.7198,0.7702,0.8562,0.8688,0.9236,1.0,0.9662,0.9822,0.736,0.4158,0.2918,0.328,0.369,0.345,0.2863,0.0864,0.3724,0.4649,0.3488,0.1817,0.1142,0.122,0.2621,0.4461,0.4726,0.3263,0.1423,0.039,0.0406,0.0311,0.0086,0.0154,0.0048,0.0025,0.0087,0.0072,0.0095,0.0086,0.0085,0.004,0.0051,M
179 | 0.0201,0.0165,0.0344,0.033,0.0397,0.0443,0.0684,0.0903,0.1739,0.2571,0.2931,0.3108,0.3603,0.3002,0.2718,0.2007,0.1801,0.2234,0.3568,0.5492,0.7209,0.8318,0.8864,0.952,0.9637,1.0,0.9673,0.8664,0.7896,0.6345,0.5351,0.4056,0.2563,0.2894,0.3588,0.4296,0.4773,0.4516,0.3765,0.3051,0.1921,0.1184,0.1984,0.157,0.066,0.1294,0.0797,0.0052,0.0233,0.0152,0.0125,0.0054,0.0057,0.0137,0.0109,0.0035,0.0056,0.0105,0.0082,0.0036,M
180 | 0.0197,0.0394,0.0384,0.0076,0.0251,0.0629,0.0747,0.0578,0.1357,0.1695,0.1734,0.247,0.3141,0.3297,0.2759,0.2056,0.1162,0.1884,0.339,0.3926,0.4282,0.5418,0.6448,0.7223,0.7853,0.7984,0.8847,0.9582,0.899,0.6831,0.6108,0.548,0.5058,0.4476,0.2401,0.1405,0.1772,0.1742,0.3326,0.4021,0.3009,0.2075,0.1206,0.0255,0.0298,0.0691,0.0781,0.0777,0.0369,0.0057,0.0091,0.0134,0.0097,0.0042,0.0058,0.0072,0.0041,0.0045,0.0047,0.0054,M
181 | 0.0394,0.042,0.0446,0.0551,0.0597,0.1416,0.0956,0.0802,0.1618,0.2558,0.3078,0.3404,0.34,0.3951,0.3352,0.2252,0.2086,0.2248,0.3382,0.4578,0.6474,0.6708,0.7007,0.7619,0.7745,0.6767,0.7373,0.7834,0.9619,1.0,0.8086,0.5558,0.5409,0.4988,0.3108,0.2897,0.2244,0.096,0.2287,0.3228,0.3454,0.3882,0.324,0.0926,0.1173,0.0566,0.0766,0.0969,0.0588,0.005,0.0118,0.0146,0.004,0.0114,0.0032,0.0062,0.0101,0.0068,0.0053,0.0087,M
182 | 0.031,0.0221,0.0433,0.0191,0.0964,0.1827,0.1106,0.1702,0.2804,0.4432,0.5222,0.5611,0.5379,0.4048,0.2245,0.1784,0.2297,0.272,0.5209,0.6898,0.8202,0.878,0.76,0.7616,0.7152,0.7288,0.8686,0.9509,0.8348,0.573,0.4363,0.4289,0.424,0.3156,0.1287,0.1477,0.2062,0.24,0.5173,0.5168,0.1491,0.2407,0.3415,0.4494,0.4624,0.2001,0.0775,0.1232,0.0783,0.0089,0.0249,0.0204,0.0059,0.0053,0.0079,0.0037,0.0015,0.0056,0.0067,0.0054,M
183 | 0.0423,0.0321,0.0709,0.0108,0.107,0.0973,0.0961,0.1323,0.2462,0.2696,0.3412,0.4292,0.3682,0.394,0.2965,0.3172,0.2825,0.305,0.2408,0.542,0.6802,0.632,0.5824,0.6805,0.5984,0.8412,0.9911,0.9187,0.8005,0.6713,0.5632,0.7332,0.6038,0.2575,0.0349,0.1799,0.3039,0.476,0.5756,0.4254,0.5046,0.7179,0.6163,0.5663,0.5749,0.3593,0.2526,0.2299,0.1271,0.0356,0.0367,0.0176,0.0035,0.0093,0.0121,0.0075,0.0056,0.0021,0.0043,0.0017,M
184 | 0.0095,0.0308,0.0539,0.0411,0.0613,0.1039,0.1016,0.1394,0.2592,0.3745,0.4229,0.4499,0.5404,0.4303,0.3333,0.3496,0.3426,0.2851,0.4062,0.6833,0.765,0.667,0.5703,0.5995,0.6484,0.8614,0.9819,0.938,0.8435,0.6074,0.5403,0.689,0.5977,0.3244,0.0516,0.3157,0.359,0.3881,0.5716,0.4314,0.3051,0.4393,0.4302,0.4831,0.5084,0.1952,0.1539,0.2037,0.1054,0.0251,0.0357,0.0181,0.0019,0.0102,0.0133,0.004,0.0042,0.003,0.0031,0.0033,M
185 | 0.0096,0.0404,0.0682,0.0688,0.0887,0.0932,0.0955,0.214,0.2546,0.2952,0.4025,0.5148,0.4901,0.4127,0.3575,0.3447,0.3068,0.2945,0.4351,0.7264,0.8147,0.8103,0.6665,0.6958,0.7748,0.8688,1.0,0.9941,0.8793,0.6482,0.5876,0.6408,0.4972,0.2755,0.03,0.3356,0.3167,0.4133,0.6281,0.4977,0.2613,0.4697,0.4806,0.4921,0.5294,0.2216,0.1401,0.1888,0.0947,0.0134,0.031,0.0237,0.0078,0.0144,0.017,0.0012,0.0109,0.0036,0.0043,0.0018,M
186 | 0.0269,0.0383,0.0505,0.0707,0.1313,0.2103,0.2263,0.2524,0.3595,0.5915,0.6675,0.5679,0.5175,0.3334,0.2002,0.2856,0.2937,0.3424,0.5949,0.7526,0.8959,0.8147,0.7109,0.7378,0.7201,0.8254,0.8917,0.982,0.8179,0.4848,0.3203,0.2775,0.2382,0.2911,0.1675,0.3156,0.1869,0.3391,0.5993,0.4124,0.1181,0.3651,0.4655,0.4777,0.3517,0.092,0.1227,0.1785,0.1085,0.03,0.0346,0.0167,0.0199,0.0145,0.0081,0.0045,0.0043,0.0027,0.0055,0.0057,M
187 | 0.034,0.0625,0.0381,0.0257,0.0441,0.1027,0.1287,0.185,0.2647,0.4117,0.5245,0.5341,0.5554,0.3915,0.295,0.3075,0.3021,0.2719,0.5443,0.7932,0.8751,0.8667,0.7107,0.6911,0.7287,0.8792,1.0,0.9816,0.8984,0.6048,0.4934,0.5371,0.4586,0.2908,0.0774,0.2249,0.1602,0.3958,0.6117,0.5196,0.2321,0.437,0.3797,0.4322,0.4892,0.1901,0.094,0.1364,0.0906,0.0144,0.0329,0.0141,0.0019,0.0067,0.0099,0.0042,0.0057,0.0051,0.0033,0.0058,M
188 | 0.0209,0.0191,0.0411,0.0321,0.0698,0.1579,0.1438,0.1402,0.3048,0.3914,0.3504,0.3669,0.3943,0.3311,0.3331,0.3002,0.2324,0.1381,0.345,0.4428,0.489,0.3677,0.4379,0.4864,0.6207,0.7256,0.6624,0.7689,0.7981,0.8577,0.9273,0.7009,0.4851,0.3409,0.1406,0.1147,0.1433,0.182,0.3605,0.5529,0.5988,0.5077,0.5512,0.5027,0.7034,0.5904,0.4069,0.2761,0.1584,0.051,0.0054,0.0078,0.0201,0.0104,0.0039,0.0031,0.0062,0.0087,0.007,0.0042,M
189 | 0.0368,0.0279,0.0103,0.0566,0.0759,0.0679,0.097,0.1473,0.2164,0.2544,0.2936,0.2935,0.2657,0.3187,0.2794,0.2534,0.198,0.1929,0.2826,0.3245,0.3504,0.3324,0.4217,0.4774,0.4808,0.6325,0.8334,0.9458,1.0,0.8425,0.5524,0.4795,0.52,0.3968,0.194,0.1519,0.201,0.1736,0.1029,0.2244,0.3717,0.4449,0.3939,0.203,0.201,0.2187,0.184,0.1477,0.0971,0.0224,0.0151,0.0105,0.0024,0.0018,0.0057,0.0092,0.0009,0.0086,0.011,0.0052,M
190 | 0.0089,0.0274,0.0248,0.0237,0.0224,0.0845,0.1488,0.1224,0.1569,0.2119,0.3003,0.3094,0.2743,0.2547,0.187,0.1452,0.1457,0.2429,0.3259,0.3679,0.3355,0.31,0.3914,0.528,0.6409,0.7707,0.8754,1.0,0.9806,0.6969,0.4973,0.502,0.5359,0.3842,0.1848,0.1149,0.157,0.1311,0.1583,0.2631,0.3103,0.4512,0.3785,0.1269,0.1459,0.1092,0.1485,0.1385,0.0716,0.0176,0.0199,0.0096,0.0103,0.0093,0.0025,0.0044,0.0021,0.0069,0.006,0.0018,M
191 | 0.0158,0.0239,0.015,0.0494,0.0988,0.1425,0.1463,0.1219,0.1697,0.1923,0.2361,0.2719,0.3049,0.2986,0.2226,0.1745,0.2459,0.31,0.3572,0.4283,0.4268,0.3735,0.4585,0.6094,0.7221,0.7595,0.8706,1.0,0.9815,0.7187,0.5848,0.4192,0.3756,0.3263,0.1944,0.1394,0.167,0.1275,0.1666,0.2574,0.2258,0.2777,0.1613,0.1335,0.1976,0.1234,0.1554,0.1057,0.049,0.0097,0.0223,0.0121,0.0108,0.0057,0.0028,0.0079,0.0034,0.0046,0.0022,0.0021,M
192 | 0.0156,0.021,0.0282,0.0596,0.0462,0.0779,0.1365,0.078,0.1038,0.1567,0.2476,0.2783,0.2896,0.2956,0.3189,0.1892,0.173,0.2226,0.2427,0.3149,0.4102,0.3808,0.4896,0.6292,0.7519,0.7985,0.883,0.9915,0.9223,0.6981,0.6167,0.5069,0.3921,0.3524,0.2183,0.1245,0.1592,0.1626,0.2356,0.2483,0.2437,0.2715,0.1184,0.1157,0.1449,0.1883,0.1954,0.1492,0.0511,0.0155,0.0189,0.015,0.006,0.0082,0.0091,0.0038,0.0056,0.0056,0.0048,0.0024,M
193 | 0.0315,0.0252,0.0167,0.0479,0.0902,0.1057,0.1024,0.1209,0.1241,0.1533,0.2128,0.2536,0.2686,0.2803,0.1886,0.1485,0.216,0.2417,0.2989,0.3341,0.3786,0.3956,0.5232,0.6913,0.7868,0.8337,0.9199,1.0,0.899,0.6456,0.5967,0.4355,0.2997,0.2294,0.1866,0.0922,0.1829,0.1743,0.2452,0.2407,0.2518,0.3184,0.1685,0.0675,0.1186,0.1833,0.1878,0.1114,0.031,0.0143,0.0138,0.0108,0.0062,0.0044,0.0072,0.0007,0.0054,0.0035,0.0001,0.0055,M
194 | 0.0056,0.0267,0.0221,0.0561,0.0936,0.1146,0.0706,0.0996,0.1673,0.1859,0.2481,0.2712,0.2934,0.2637,0.188,0.1405,0.2028,0.2613,0.2778,0.3346,0.383,0.4003,0.5114,0.686,0.749,0.7843,0.9021,1.0,0.8888,0.6511,0.6083,0.4463,0.2948,0.1729,0.1488,0.0801,0.177,0.1382,0.2404,0.2046,0.197,0.2778,0.1377,0.0685,0.0664,0.1665,0.1807,0.1245,0.0516,0.0044,0.0185,0.0072,0.0055,0.0074,0.0068,0.0084,0.0037,0.0024,0.0034,0.0007,M
195 | 0.0203,0.0121,0.038,0.0128,0.0537,0.0874,0.1021,0.0852,0.1136,0.1747,0.2198,0.2721,0.2105,0.1727,0.204,0.1786,0.1318,0.226,0.2358,0.3107,0.3906,0.3631,0.4809,0.6531,0.7812,0.8395,0.918,0.9769,0.8937,0.7022,0.65,0.5069,0.3903,0.3009,0.1565,0.0985,0.22,0.2243,0.2736,0.2152,0.2438,0.3154,0.2112,0.0991,0.0594,0.194,0.1937,0.1082,0.0336,0.0177,0.0209,0.0134,0.0094,0.0047,0.0045,0.0042,0.0028,0.0036,0.0013,0.0016,M
196 | 0.0392,0.0108,0.0267,0.0257,0.041,0.0491,0.1053,0.169,0.2105,0.2471,0.268,0.3049,0.2863,0.2294,0.1165,0.2127,0.2062,0.2222,0.3241,0.433,0.5071,0.5944,0.7078,0.7641,0.8878,0.9711,0.988,0.9812,0.9464,0.8542,0.6457,0.3397,0.3828,0.3204,0.1331,0.044,0.1234,0.203,0.1652,0.1043,0.1066,0.211,0.2417,0.1631,0.0769,0.0723,0.0912,0.0812,0.0496,0.0101,0.0089,0.0083,0.008,0.0026,0.0079,0.0042,0.0071,0.0044,0.0022,0.0014,M
197 | 0.0129,0.0141,0.0309,0.0375,0.0767,0.0787,0.0662,0.1108,0.1777,0.2245,0.2431,0.3134,0.3206,0.2917,0.2249,0.2347,0.2143,0.2939,0.4898,0.6127,0.7531,0.7718,0.7432,0.8673,0.9308,0.9836,1.0,0.9595,0.8722,0.6862,0.4901,0.328,0.3115,0.1969,0.1019,0.0317,0.0756,0.0907,0.1066,0.138,0.0665,0.1475,0.247,0.2788,0.2709,0.2283,0.1818,0.1185,0.0546,0.0219,0.0204,0.0124,0.0093,0.0072,0.0019,0.0027,0.0054,0.0017,0.0024,0.0029,M
198 | 0.005,0.0017,0.027,0.045,0.0958,0.083,0.0879,0.122,0.1977,0.2282,0.2521,0.3484,0.3309,0.2614,0.1782,0.2055,0.2298,0.3545,0.6218,0.7265,0.8346,0.8268,0.8366,0.9408,0.951,0.9801,0.9974,1.0,0.9036,0.6409,0.3857,0.2908,0.204,0.1653,0.1769,0.114,0.074,0.0941,0.0621,0.0426,0.0572,0.1068,0.1909,0.2229,0.2203,0.2265,0.1766,0.1097,0.0558,0.0142,0.0281,0.0165,0.0056,0.001,0.0027,0.0062,0.0024,0.0063,0.0017,0.0028,M
199 | 0.0366,0.0421,0.0504,0.025,0.0596,0.0252,0.0958,0.0991,0.1419,0.1847,0.2222,0.2648,0.2508,0.2291,0.1555,0.1863,0.2387,0.3345,0.5233,0.6684,0.7766,0.7928,0.794,0.9129,0.9498,0.9835,1.0,0.9471,0.8237,0.6252,0.4181,0.3209,0.2658,0.2196,0.1588,0.0561,0.0948,0.17,0.1215,0.1282,0.0386,0.1329,0.2331,0.2468,0.196,0.1985,0.157,0.0921,0.0549,0.0194,0.0166,0.0132,0.0027,0.0022,0.0059,0.0016,0.0025,0.0017,0.0027,0.0027,M
200 | 0.0238,0.0318,0.0422,0.0399,0.0788,0.0766,0.0881,0.1143,0.1594,0.2048,0.2652,0.31,0.2381,0.1918,0.143,0.1735,0.1781,0.2852,0.5036,0.6166,0.7616,0.8125,0.7793,0.8788,0.8813,0.947,1.0,0.9739,0.8446,0.6151,0.4302,0.3165,0.2869,0.2017,0.1206,0.0271,0.058,0.1262,0.1072,0.1082,0.036,0.1197,0.2061,0.2054,0.1878,0.2047,0.1716,0.1069,0.0477,0.017,0.0186,0.0096,0.0071,0.0084,0.0038,0.0026,0.0028,0.0013,0.0035,0.006,M
201 | 0.0116,0.0744,0.0367,0.0225,0.0076,0.0545,0.111,0.1069,0.1708,0.2271,0.3171,0.2882,0.2657,0.2307,0.1889,0.1791,0.2298,0.3715,0.6223,0.726,0.7934,0.8045,0.8067,0.9173,0.9327,0.9562,1.0,0.9818,0.8684,0.6381,0.3997,0.3242,0.2835,0.2413,0.2321,0.126,0.0693,0.0701,0.1439,0.1475,0.0438,0.0469,0.1476,0.1742,0.1555,0.1651,0.1181,0.072,0.0321,0.0056,0.0202,0.0141,0.0103,0.01,0.0034,0.0026,0.0037,0.0044,0.0057,0.0035,M
202 | 0.0131,0.0387,0.0329,0.0078,0.0721,0.1341,0.1626,0.1902,0.261,0.3193,0.3468,0.3738,0.3055,0.1926,0.1385,0.2122,0.2758,0.4576,0.6487,0.7154,0.801,0.7924,0.8793,1.0,0.9865,0.9474,0.9474,0.9315,0.8326,0.6213,0.3772,0.2822,0.2042,0.219,0.2223,0.1327,0.0521,0.0618,0.1416,0.146,0.0846,0.1055,0.1639,0.1916,0.2085,0.2335,0.1964,0.13,0.0633,0.0183,0.0137,0.015,0.0076,0.0032,0.0037,0.0071,0.004,0.0009,0.0015,0.0085,M
203 | 0.0335,0.0258,0.0398,0.057,0.0529,0.1091,0.1709,0.1684,0.1865,0.266,0.3188,0.3553,0.3116,0.1965,0.178,0.2794,0.287,0.3969,0.5599,0.6936,0.7969,0.7452,0.8203,0.9261,0.881,0.8814,0.9301,0.9955,0.8576,0.6069,0.3934,0.2464,0.1645,0.114,0.0956,0.008,0.0702,0.0936,0.0894,0.1127,0.0873,0.102,0.1964,0.2256,0.1814,0.2012,0.1688,0.1037,0.0501,0.0136,0.013,0.012,0.0039,0.0053,0.0062,0.0046,0.0045,0.0022,0.0005,0.0031,M
204 | 0.0272,0.0378,0.0488,0.0848,0.1127,0.1103,0.1349,0.2337,0.3113,0.3997,0.3941,0.3309,0.2926,0.176,0.1739,0.2043,0.2088,0.2678,0.2434,0.1839,0.2802,0.6172,0.8015,0.8313,0.844,0.8494,0.9168,1.0,0.7896,0.5371,0.6472,0.6505,0.4959,0.2175,0.099,0.0434,0.1708,0.1979,0.188,0.1108,0.1702,0.0585,0.0638,0.1391,0.0638,0.0581,0.0641,0.1044,0.0732,0.0275,0.0146,0.0091,0.0045,0.0043,0.0043,0.0098,0.0054,0.0051,0.0065,0.0103,M
205 | 0.0187,0.0346,0.0168,0.0177,0.0393,0.163,0.2028,0.1694,0.2328,0.2684,0.3108,0.2933,0.2275,0.0994,0.1801,0.22,0.2732,0.2862,0.2034,0.174,0.413,0.6879,0.812,0.8453,0.8919,0.93,0.9987,1.0,0.8104,0.6199,0.6041,0.5547,0.416,0.1472,0.0849,0.0608,0.0969,0.1411,0.1676,0.12,0.1201,0.1036,0.1977,0.1339,0.0902,0.1085,0.1521,0.1363,0.0858,0.029,0.0203,0.0116,0.0098,0.0199,0.0033,0.0101,0.0065,0.0115,0.0193,0.0157,M
206 | 0.0323,0.0101,0.0298,0.0564,0.076,0.0958,0.099,0.1018,0.103,0.2154,0.3085,0.3425,0.299,0.1402,0.1235,0.1534,0.1901,0.2429,0.212,0.2395,0.3272,0.5949,0.8302,0.9045,0.9888,0.9912,0.9448,1.0,0.9092,0.7412,0.7691,0.7117,0.5304,0.2131,0.0928,0.1297,0.1159,0.1226,0.1768,0.0345,0.1562,0.0824,0.1149,0.1694,0.0954,0.008,0.079,0.1255,0.0647,0.0179,0.0051,0.0061,0.0093,0.0135,0.0063,0.0063,0.0034,0.0032,0.0062,0.0067,M
207 | 0.0522,0.0437,0.018,0.0292,0.0351,0.1171,0.1257,0.1178,0.1258,0.2529,0.2716,0.2374,0.1878,0.0983,0.0683,0.1503,0.1723,0.2339,0.1962,0.1395,0.3164,0.5888,0.7631,0.8473,0.9424,0.9986,0.9699,1.0,0.863,0.6979,0.7717,0.7305,0.5197,0.1786,0.1098,0.1446,0.1066,0.144,0.1929,0.0325,0.149,0.0328,0.0537,0.1309,0.091,0.0757,0.1059,0.1005,0.0535,0.0235,0.0155,0.016,0.0029,0.0051,0.0062,0.0089,0.014,0.0138,0.0077,0.0031,M
208 | 0.0303,0.0353,0.049,0.0608,0.0167,0.1354,0.1465,0.1123,0.1945,0.2354,0.2898,0.2812,0.1578,0.0273,0.0673,0.1444,0.207,0.2645,0.2828,0.4293,0.5685,0.699,0.7246,0.7622,0.9242,1.0,0.9979,0.8297,0.7032,0.7141,0.6893,0.4961,0.2584,0.0969,0.0776,0.0364,0.1572,0.1823,0.1349,0.0849,0.0492,0.1367,0.1552,0.1548,0.1319,0.0985,0.1258,0.0954,0.0489,0.0241,0.0042,0.0086,0.0046,0.0126,0.0036,0.0035,0.0034,0.0079,0.0036,0.0048,M
209 | 0.026,0.0363,0.0136,0.0272,0.0214,0.0338,0.0655,0.14,0.1843,0.2354,0.272,0.2442,0.1665,0.0336,0.1302,0.1708,0.2177,0.3175,0.3714,0.4552,0.57,0.7397,0.8062,0.8837,0.9432,1.0,0.9375,0.7603,0.7123,0.8358,0.7622,0.4567,0.1715,0.1549,0.1641,0.1869,0.2655,0.1713,0.0959,0.0768,0.0847,0.2076,0.2505,0.1862,0.1439,0.147,0.0991,0.0041,0.0154,0.0116,0.0181,0.0146,0.0129,0.0047,0.0039,0.0061,0.004,0.0036,0.0061,0.0115,M
210 | 


--------------------------------------------------------------------------------
/synth_data_experiments.py:
--------------------------------------------------------------------------------
  1 | import os
  2 | from final_model import LGBM_w_Feature_Selector 
  3 | import logging
  4 | import warnings
  5 | from sklearn.exceptions import DataConversionWarning,ConvergenceWarning
  6 | from sklearn.preprocessing import MinMaxScaler, StandardScaler
  7 | import pandas as pd
  8 | import re
  9 | from sklearn.model_selection import train_test_split
 10 | import matplotlib.pyplot as plt
 11 | import numpy as np
 12 | warnings.filterwarnings("ignore", category=DataConversionWarning)
 13 | warnings.filterwarnings("ignore", category=ConvergenceWarning)
 14 | logging.basicConfig(filename='final_exp.log', filemode='w', level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')
 15 | 
 16 | lgbm_param_grid={
 17 | 'num_leaves': [7, 15],#[31, 63],
 18 | 'learning_rate': [0.01, 0.025, 0.05],
 19 | 'n_estimators': [20, 40],
 20 | 'subsample': [0.6, 0.8],
 21 | 'colsample_bytree': [0.6, 0.8],
 22 | # 'reg_alpha': [0.0, 0.1, 0.5],
 23 | # 'reg_lambda': [0.0, 0.1, 0.5],
 24 | 'min_child_samples': [5, 10],
 25 | }
 26 | # lgbm_param_grid={
 27 | # 'num_leaves': [63, 127],
 28 | # 'learning_rate': [0.01, 0.025, 0.05],
 29 | # 'n_estimators': [50, 100],
 30 | # 'subsample': [0.6, 0.8],
 31 | # 'colsample_bytree': [0.6, 0.8],
 32 | # # 'reg_alpha': [0.0, 0.1, 0.5],
 33 | # # 'reg_lambda': [0.0, 0.1, 0.5],
 34 | # #'min_child_samples': [5, 10],
 35 | # }
 36 | mlp_param_grid ={
 37 |     "hidden_layer_sizes": [(40,20),(40),(80),(20,10)],#[(20),(40),(80),(40,10),(80,20),(80,40),(80,20,10),(80,40,20),(80,40,10)], 
 38 |     "activation":['relu','logistic'], 
 39 |     "solver":['adam',], 
 40 |     "alpha":[0.0001, 0.001, 0.01], 
 41 |     "learning_rate_init":[0.001, 0.01], 
 42 |     "warm_start":[False],  
 43 |     "momentum":[0.9], 
 44 | } 
 45 | 
 46 | # Settings
 47 | n_samples = 300
 48 | n_real_features = 10
 49 | n_dummy_features = 90
 50 | np.random.seed(42)
 51 | # Generate real features
 52 | real_features = np.random.rand(n_samples, n_real_features)
 53 | #redundant_features =np.copy(real_features)
 54 | # Generate dummy features
 55 | dummy_features = np.random.rand(n_samples, n_dummy_features)
 56 | 
 57 | # Combine into one dataset
 58 | data = np.hstack((real_features, dummy_features))
 59 | 
 60 | 
 61 | target = np.zeros(n_samples)
 62 | for i in range(n_real_features):
 63 |     target += real_features[:, i] ** 2  # Polynomial term (square)
 64 |     target += np.sin(real_features[:, i])
 65 | target = np.expand_dims(target, 1)
 66 | X_train, X_temp_test, y_train, y_temp_test = train_test_split(data, target, test_size=0.55, random_state=42)
 67 | 
 68 | # Split the temporary test set into true test set and temporary validation set
 69 | X_test, X_temp_val, y_test, y_temp_val = train_test_split(X_temp_test, y_temp_test, test_size=0.73, random_state=42)
 70 | 
 71 | # Finally, split the temporary validation set into two validation sets (val1 and val2)
 72 | X_val1, X_val2, y_val1, y_val2 = train_test_split(X_temp_val, y_temp_val, test_size=0.33, random_state=42)
 73 | 
 74 | network = LGBM_w_Feature_Selector(model="lgbm",problem_type="regression",param_grid=lgbm_param_grid,X_train=X_train, X_test=X_test, slack=0.0, 
 75 |                                         X_val_1=X_val1, y_val_1=y_val1,X_val_2=X_val2, y_val_2=y_val2, y_train=y_train,y_test=y_test,iterations=10) 
 76 | _,all_score, flbmo_score, mo_score, gbm_score, rfe_score, mi_score = network.feature_extraction(5, seed=11 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=True, f_numbers=[6, 10, 15, 20], loss_ts=[1,2], slacks=[ 0.001, 0.005, 0.02, 0.05]) #slacks=[0.005, 0.01])
 77 | plt.scatter(range(len(network.best_ft_nbs)), network.best_ft_nbs, s=10)
 78 | print(network.best_ft_nbs[-1])
 79 | # Adding title and labels (optional)
 80 | plt.title("GBMO Iterations Features Selected")
 81 | plt.xlabel("Iterations")
 82 | plt.ylabel("Number of Selected Features")
 83 | plt.ylim([0, 110])
 84 | # Save the figure
 85 | plt.savefig("gbmo_plot.png")  # You can change the filename and extension as needed
 86 | 
 87 | # Display the plot
 88 | plt.show()
 89 | #plt.scatter(range(len(network.selected_feature_nbs_flbmo)), network.selected_feature_nbs_flbmo, s=10)
 90 | plt.figure()
 91 | plt.scatter(range(len(network.best_ft_val_losses)), network.best_ft_val_losses, s=10)
 92 | 
 93 | ##Adding title and labels (optional)
 94 | plt.title("Validation Losses with respect to Iterations")
 95 | plt.xlabel("Iterations")
 96 | plt.ylabel("Validation Loss")
 97 | 
 98 | ##Save the figure
 99 | plt.savefig("synth_lgbm_gbmo_plot_losses.png") 
100 | 
101 | # Display the plot
102 | plt.show()
103 | 
104 | # # Adding title and labels (optional)
105 | # plt.title("FLBMO Iterations Features Selected")
106 | # plt.xlabel("Iterations")
107 | # plt.ylabel("Number of Selected Features")
108 | 
109 | # # Save the figure
110 | # plt.savefig("flbmo_plot.png")  # You can change the filename and extension as needed
111 | 
112 | # Display the plot
113 | plt.show()
114 | 
115 | # network = LGBM_w_Feature_Selector(model="mlp",problem_type="regression",param_grid=mlp_param_grid,X_train=X_train, X_test=X_test, slack=0.0, 
116 | #                                         X_val_1=X_val1, y_val_1=y_val1,X_val_2=X_val2, y_val_2=y_val2, y_train=y_train,y_test=y_test,iterations=10) 
117 | # # gbmo_score, flbmo_score = network.test_on_synth_data(slacks=[0.001, 0.008], loss_ts=[1,2], f_numbers=[10, 20, 30])
118 | # # print(gbmo_score)
119 | # # print(flbmo_score)
120 | # _,all_score, flbmo_score, mo_score, gbm_score, mi_score = network.feature_extraction(5, seed=42 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=False, f_numbers=[6, 10, 15, 20], loss_ts=[1,2], slacks=[ 0.001, 0.25]) #slacks=[0.005, 0.01])
121 | 
122 | # plt.scatter(range(len(network.selected_feature_nbs_gbmo)), network.selected_feature_nbs_gbmo, s=10)
123 | 
124 | # # Adding title and labels (optional)
125 | # plt.title("GBMO Iterations Features Selected")
126 | # plt.xlabel("Iterations")
127 | # plt.ylabel("Number of Selected Features")
128 | 
129 | # # Save the figure
130 | # plt.savefig("mlp_gbmo_plot.png")  # You can change the filename and extension as needed
131 | 
132 | # # Display the plot
133 | # plt.show()
134 | # plt.scatter(range(len(network.selected_feature_nbs_flbmo)), network.selected_feature_nbs_flbmo, s=10)
135 | # plt.figure()
136 | # plt.scatter(range(len(network.best_ft_val_losses)), network.best_ft_val_losses, s=10)
137 | 
138 | # ##Adding title and labels (optional)
139 | # plt.title("Validation Losses with respect to Iterations")
140 | # plt.xlabel("Iterations")
141 | # plt.ylabel("Validation Loss")
142 | 
143 | # ##Save the figure
144 | # plt.savefig("synth_mlp_gbmo_plot_losses.png") 
145 | # # Display the plot
146 | # plt.show()
147 | 
148 | # # Adding title and labels (optional)
149 | # plt.title("FLBMO Iterations Features Selected")
150 | # plt.xlabel("Iterations")
151 | # plt.ylabel("Number of Selected Features")
152 | 
153 | # # Save the figure
154 | # plt.savefig("flbmo_plot.png")  # You can change the filename and extension as needed
155 | 
156 | # # Display the plot
157 | # plt.show()
158 | 
159 | 
160 | 
161 | 
162 | # target = np.zeros(n_samples)
163 | # for i in range(n_real_features):
164 | #     target += np.cos(real_features[:, i])* real_features[:, i] ** (1/2)
165 | #     target += np.sin(real_features[:, i])
166 | #     # target += np.exp(-real_features[:, i])  # Exponential decay
167 | #     # if real_features[:, i].all() > 0:  # Avoiding log(0)
168 | #     #     target += np.log(real_features[:, i] + 1)  # Logarithmic growth
169 | # # target = (target-np.min(target))/(np.max(target)-np.min(target))
170 | # #target+=np.random.normal(0, 0.3, n_samples)
171 | # #target+=np.random.normal(0, 0.05, n_samples)
172 | # #target = (target-np.min(target))/(np.max(target)-np.min(target))
173 | # target = np.expand_dims(target, 1)
174 | # X_train, X_temp_test, y_train, y_temp_test = train_test_split(data, target, test_size=0.55, random_state=42)
175 | 
176 | # # Split the temporary test set into true test set and temporary validation set
177 | # X_test, X_temp_val, y_test, y_temp_val = train_test_split(X_temp_test, y_temp_test, test_size=0.73, random_state=42)
178 | 
179 | # # Finally, split the temporary validation set into two validation sets (val1 and val2)
180 | # X_val1, X_val2, y_val1, y_val2 = train_test_split(X_temp_val, y_temp_val, test_size=0.33, random_state=42)
181 | 
182 | # network = LGBM_w_Feature_Selector(model="lgbm",problem_type="regression",param_grid=lgbm_param_grid,X_train=X_train, X_test=X_test, slack=0.0, 
183 | #                                         X_val_1=X_val1, y_val_1=y_val1,X_val_2=X_val2, y_val_2=y_val2, y_train=y_train,y_test=y_test,iterations=10) 
184 | # # gbmo_score, flbmo_score = network.test_on_synth_data(slacks=[0.001, 0.005], loss_ts=[1,2], f_numbers=[10, 20, 30])
185 | # # print(gbmo_score)
186 | # # print(flbmo_score)
187 | # _,all_score, flbmo_score, mo_score, gbm_score, rfe_score, mi_score = network.feature_extraction(5, seed=11 , method="convergence", loss_tolerance=1,run_CV=True, include_RFE=True, f_numbers=[6, 10, 15], loss_ts=[1,2], slacks=[ 0.001, 0.005, 0.02, 0.05]) #slacks=[0.005, 0.01])
188 | 


--------------------------------------------------------------------------------