-
Notifications
You must be signed in to change notification settings - Fork 2
/
superconduct_regression.py
358 lines (239 loc) · 15.8 KB
/
superconduct_regression.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
# from two_TrAdaBoostR2 import TwoStageTrAdaBoostR2 ##STrAdaBoost.R2
# from TwoStageTrAdaBoostR2 import TwoStageTrAdaBoostR2 ##two-stage TrAdaBoost.R2
import pandas as pd
import sys
import numpy as np
from pandas import DataFrame
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import AdaBoostRegressor
from keras.models import Sequential, load_model, Model
from keras.layers import Input, Dense, Activation, Conv2D, Dropout, Flatten
from keras import optimizers, utils, initializers, regularizers
import keras.backend as K
from sklearn.metrics import mean_squared_error
from sklearn.metrics import r2_score
from sklearn.preprocessing import StandardScaler #Importing the StandardScaler
from itertools import combinations
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats.stats import pearsonr
from math import sqrt
#Geo plotting libraries
import geopandas as gdp
from matplotlib.colors import ListedColormap
# import geoplot as glpt
import xgboost as xgb
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import Ridge
from sklearn import linear_model
from keras.models import Sequential
from keras.callbacks import ModelCheckpoint
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.model_selection import KFold
import matplotlib.lines as mlines
import statistics
from scipy.stats import rv_continuous
from scipy.stats import *
from statistics import mean
from sklearn.cluster import KMeans
from scipy.spatial import distance
from sklearn.model_selection import KFold
######### Instance Transfer repositories ####################
from adapt.instance_based import TwoStageTrAdaBoostR2
print("Repositories uploaded!!")
from adapt.instance_based import TrAdaBoost, TrAdaBoostR2, TwoStageTrAdaBoostR2
from sklearn.model_selection import GridSearchCV
from adapt.instance_based import KMM
print("Second Upload Completed!!")
################################### superconduct ###########################################################################################################
################################### Superconductivity ###########################################################################################################
## Target Data: critical_temp
## Correlation col: gmean_ElectronAffinity
## Cuts at: 40.0 and 60.0
##########################################################################################################################################################
superconduct_df = pd.read_csv("Scientific_data/superconduct/train.csv")
superconduct_unq_df = pd.read_csv("Scientific_data/superconduct/train.csv")
print("Superconductivity Data")
print("-------------------------------------------")
print(superconduct_df.shape)
# print("The correlation matrix is: ")
# superconduct_df_corr = superconduct_df.corr()['critical_temp'].abs().sort_values()
# print(superconduct_df_corr.to_string()) ### Helps to print the entire series
# to find where to split the data
# print('Min: ', superconduct_df['gmean_ElectronAffinity'].min())
# print('Max: ', superconduct_df['gmean_ElectronAffinity'].max())
ss = StandardScaler()
drop_col_superconduct = ['gmean_ElectronAffinity']
# superconduct_df['gmean_ElectronAffinity'].sort_values()
superconduct_train_df = superconduct_df.loc[(superconduct_df['gmean_ElectronAffinity'] > 40.0) & (superconduct_df['gmean_ElectronAffinity'] <= 60.0)]
superconduct_train_df = superconduct_train_df.drop(drop_col_superconduct, axis = 1)
superconduct_train_df = superconduct_train_df.reset_index(drop = True)
print("Training Set: ", superconduct_train_df.shape)
superconduct_source1_df = superconduct_df.loc[(superconduct_df['gmean_ElectronAffinity'] > 60.0)]
superconduct_source1_df = superconduct_source1_df.drop(drop_col_superconduct, axis = 1)
superconduct_source1_df = superconduct_source1_df.reset_index(drop = True)
print("Source Set 1: ", superconduct_source1_df.shape)
superconduct_source2_df = superconduct_df.loc[(superconduct_df['gmean_ElectronAffinity'] <= 40.0)]
superconduct_source2_df = superconduct_source2_df.drop(drop_col_superconduct, axis = 1)
superconduct_source2_df = superconduct_source2_df.reset_index(drop = True)
print("Source Set 2: ",superconduct_source2_df.shape)
superconduct_source_df = pd.concat([superconduct_source1_df, superconduct_source2_df], ignore_index=True)
superconduct_source_df = superconduct_source_df.sample(frac = 0.3, replace = True, random_state = 1)
print("Final Source Set: ",superconduct_source_df.shape)
#################### Splitting into features and target ####################
target_column_superconduct = ['critical_temp']
superconduct_train_df_y = superconduct_train_df[target_column_superconduct]
superconduct_train_df_X = superconduct_train_df.drop(target_column_superconduct, axis = 1)
superconduct_cols = superconduct_train_df_X.columns
superconduct_train_df_X[superconduct_cols] = ss.fit_transform(superconduct_train_df_X[superconduct_cols])
superconduct_source_df_y = superconduct_source_df[target_column_superconduct]
superconduct_source_df_X = superconduct_source_df.drop(target_column_superconduct, axis = 1)
superconduct_cols = superconduct_source_df_X.columns
superconduct_source_df_X[superconduct_cols] = ss.fit_transform(superconduct_source_df_X[superconduct_cols])
########################### Transfer Learning superconduct #####################################################
from sklearn.ensemble import AdaBoostRegressor
def get_estimator(**kwargs):
return DecisionTreeRegressor(max_depth = 6)
kwargs_TwoTrAda = {'steps': 30,
'fold': 10,
'learning_rate': 0.1}
print("Adaboost.R2 Transfer Learning (M + H, L)")
print("-------------------------------------------")
r2scorelist_AdaTL_superconduct = []
rmselist_AdaTL_superconduct = []
r2scorelist_Ada_superconduct = []
rmselist_Ada_superconduct = []
r2scorelist_KMM_superconduct = []
rmselist_KMM_superconduct = []
r2scorelist_GBRTL_superconduct = []
rmselist_GBRTL_superconduct = []
r2scorelist_GBR_superconduct = []
rmselist_GBR_superconduct = []
r2scorelist_TwoTrAda_superconduct = []
rmselist_TwoTrAda_superconduct = []
r2scorelist_stradaboost_superconduct = []
rmselist_stradaboost_superconduct = []
kfold = KFold(n_splits = 20, random_state = 42, shuffle=False)
for train_ix, test_ix in kfold.split(superconduct_train_df_X):
############### get data ###############
superconduct_test_df_X, superconduct_tgt_df_X = superconduct_train_df_X.iloc[train_ix], superconduct_train_df_X.iloc[test_ix] #### Make it opposite, so target size is small.
superconduct_test_df_y, superconduct_tgt_df_y = superconduct_train_df_y.iloc[train_ix], superconduct_train_df_y.iloc[test_ix] #### Make it opposite, so target size is small.
print(superconduct_tgt_df_X.shape, superconduct_test_df_X.shape)
############### Merging the datasets ##########################################
superconduct_X_df = pd.concat([superconduct_tgt_df_X, superconduct_source_df_X], ignore_index=True)
superconduct_y_df = pd.concat([superconduct_tgt_df_y, superconduct_source_df_y], ignore_index=True)
superconduct_np_train_X = superconduct_X_df.to_numpy()
superconduct_np_train_y = superconduct_y_df.to_numpy()
superconduct_np_test_X = superconduct_test_df_X.to_numpy()
superconduct_np_test_y = superconduct_test_df_y.to_numpy()
superconduct_np_train_y_list = superconduct_np_train_y.ravel()
superconduct_np_test_y_list = superconduct_np_test_y.ravel()
src_size_superconduct = len(superconduct_source_df_y)
tgt_size_superconduct = len(superconduct_tgt_df_y)
src_idx = np.arange(start=0, stop=(src_size_superconduct - 1), step=1)
tgt_idx = np.arange(start=src_size_superconduct, stop=((src_size_superconduct + tgt_size_superconduct)-1), step=1)
################### AdaBoost Tl ###################
model_AdaTL_superconduct = AdaBoostRegressor(DecisionTreeRegressor(max_depth = 8), learning_rate=0.01, n_estimators=500)
model_AdaTL_superconduct.fit(superconduct_np_train_X, superconduct_np_train_y_list)
y_pred_AdaTL_superconduct = model_AdaTL_superconduct.predict(superconduct_np_test_X)
mse_AdaTL_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_AdaTL_superconduct))
rmselist_AdaTL_superconduct.append(mse_AdaTL_superconduct)
r2_score_AdaTL_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_AdaTL_superconduct)
r2_score_AdaTL_superconduct = (r2_score_AdaTL_superconduct[0])**2
r2scorelist_AdaTL_superconduct.append(r2_score_AdaTL_superconduct)
################### AdaBoost ###################
model_Ada_superconduct = AdaBoostRegressor(DecisionTreeRegressor(max_depth = 8), learning_rate=0.01, n_estimators=500)
model_Ada_superconduct.fit(superconduct_tgt_df_X, superconduct_tgt_df_y)
y_pred_ada_superconduct = model_Ada_superconduct.predict(superconduct_np_test_X)
mse_Ada_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_ada_superconduct))
rmselist_Ada_superconduct.append(mse_Ada_superconduct)
r2_score_Ada_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_ada_superconduct)
r2_score_Ada_superconduct = (r2_score_Ada_superconduct[0])**2
r2scorelist_Ada_superconduct.append(r2_score_Ada_superconduct)
################### KMM ###################
model_KMM_superconduct = KMM(get_estimator = get_estimator)
model_KMM_superconduct.fit(superconduct_np_train_X, superconduct_np_train_y_list, src_idx, tgt_idx)
y_pred_KMM_superconduct = model_KMM_superconduct.predict(superconduct_test_df_X) ##Using dataframe instead of the numpy matrix
mse_KMM_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_KMM_superconduct))
rmselist_KMM_superconduct.append(mse_KMM_superconduct)
r2_score_KMM_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_KMM_superconduct)
r2_score_KMM_superconduct = (r2_score_KMM_superconduct[0])**2
r2scorelist_KMM_superconduct.append(r2_score_KMM_superconduct)
################### GBRTL ###################
model_GBRTL_superconduct = GradientBoostingRegressor(learning_rate = 0.01, max_depth = 4, n_estimators = 1000, subsample = 0.5)
model_GBRTL_superconduct.fit(superconduct_np_train_X, superconduct_np_train_y_list)
y_pred_GBRTL_superconduct = model_GBRTL_superconduct.predict(superconduct_test_df_X) ##Using dataframe instead of the numpy matrix
mse_GBRTL_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_GBRTL_superconduct))
rmselist_GBRTL_superconduct.append(mse_GBRTL_superconduct)
r2_score_GBRTL_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_GBRTL_superconduct)
r2_score_GBRTL_superconduct = (r2_score_GBRTL_superconduct[0])**2
r2scorelist_GBRTL_superconduct.append(r2_score_GBRTL_superconduct)
################### GBR ###################
model_GBR_superconduct = GradientBoostingRegressor(learning_rate=0.01, max_depth=4, n_estimators=1000, subsample=0.5)
model_GBR_superconduct.fit(superconduct_tgt_df_X, superconduct_tgt_df_y)
y_pred_GBR_superconduct = model_GBR_superconduct.predict(superconduct_test_df_X) ##Using dataframe instead of the numpy matrix
mse_GBR_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_GBR_superconduct))
rmselist_GBR_superconduct.append(mse_GBR_superconduct)
r2_score_GBR_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_GBR_superconduct)
r2_score_GBR_superconduct = (r2_score_GBR_superconduct[0])**2
r2scorelist_GBR_superconduct.append(r2_score_GBR_superconduct)
################### Two-TrAdaBoost ###################
from adapt.instance_based import TrAdaBoost, TrAdaBoostR2, TwoStageTrAdaBoostR2
model_TwoTrAda_superconduct = TwoStageTrAdaBoostR2(get_estimator = get_estimator, n_estimators = 1000, cv=10) #, kwargs_TwoTrAda)
model_TwoTrAda_superconduct.fit(superconduct_np_train_X, superconduct_np_train_y_list, src_idx, tgt_idx)
y_pred_TwoTrAda_superconduct = model_TwoTrAda_superconduct.predict(superconduct_np_test_X)
mse_TwoTrAda_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_TwoTrAda_superconduct))
rmselist_TwoTrAda_superconduct.append(mse_TwoTrAda_superconduct)
r2_score_TwoTrAda_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_TwoTrAda_superconduct)
r2_score_TwoTrAda_superconduct = (r2_score_TwoTrAda_superconduct[0])**2
r2scorelist_TwoTrAda_superconduct.append(r2_score_TwoTrAda_superconduct)
################### STrAdaBoost ###################
from two_TrAdaBoostR2 import TwoStageTrAdaBoostR2
sample_size = [len(superconduct_tgt_df_X), len(superconduct_source_df_X)]
n_estimators = 100
steps = 30
fold = 10
random_state = np.random.RandomState(1)
model_stradaboost_superconduct = TwoStageTrAdaBoostR2(DecisionTreeRegressor(max_depth=6),
n_estimators = n_estimators, sample_size = sample_size,
steps = steps, fold = fold, random_state = random_state)
model_stradaboost_superconduct.fit(superconduct_np_train_X, superconduct_np_train_y_list)
y_pred_stradaboost_superconduct = model_stradaboost_superconduct.predict(superconduct_np_test_X)
mse_stradaboost_superconduct = sqrt(mean_squared_error(superconduct_np_test_y, y_pred_stradaboost_superconduct))
rmselist_stradaboost_superconduct.append(mse_stradaboost_superconduct)
r2_score_stradaboost_superconduct = pearsonr(superconduct_np_test_y_list, y_pred_stradaboost_superconduct)
r2_score_stradaboost_superconduct = (r2_score_stradaboost_superconduct[0])**2
r2scorelist_stradaboost_superconduct.append(r2_score_stradaboost_superconduct)
with open('superconduct_rmse.txt', 'w') as superconduct_handle_rmse:
superconduct_handle_rmse.write("AdaBoost TL:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_AdaTL_superconduct)
superconduct_handle_rmse.write("\n\nAdaBoost:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_Ada_superconduct)
superconduct_handle_rmse.write("\n\nKMM:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_KMM_superconduct)
superconduct_handle_rmse.write("\n\nGBRT:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_GBRTL_superconduct)
superconduct_handle_rmse.write("\n\nGBR:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_GBR_superconduct)
superconduct_handle_rmse.write("\n\nTrAdaBoost:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_TwoTrAda_superconduct)
superconduct_handle_rmse.write("\n\nSTrAdaBoost:\n ")
superconduct_handle_rmse.writelines("%s\n" % ele for ele in rmselist_stradaboost_superconduct)
with open('superconduct_r2.txt', 'w') as superconduct_handle_r2:
superconduct_handle_r2.write("AdaBoost TL:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_AdaTL_superconduct)
superconduct_handle_r2.write("\n\nAdaBoost:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_Ada_superconduct)
superconduct_handle_r2.write("\n\nKMM:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_KMM_superconduct)
superconduct_handle_r2.write("\n\nGBRT:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_GBRTL_superconduct)
superconduct_handle_r2.write("\n\nGBR:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_GBR_superconduct)
superconduct_handle_r2.write("\n\nTrAdaBoost:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_TwoTrAda_superconduct)
superconduct_handle_r2.write("\n\nSTrAdaBoost:\n ")
superconduct_handle_r2.writelines("%s\n" % ele for ele in r2scorelist_stradaboost_superconduct)
######################################################################################
print("-------------------------------------------")