-
Notifications
You must be signed in to change notification settings - Fork 0
/
build_al_data_file.py
526 lines (432 loc) · 27.8 KB
/
build_al_data_file.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
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
import os
import numpy as np
import pandas as pd
from scipy.stats import zscore
class ALDataBuild:
def __init__(self, folder_dls, folder_formulations):
self.folder = folder_dls
self.files = []
self.dls_data = pd.DataFrame()
self.folder_formulation = folder_formulations
self.files_formulations = []
self.random_dataframe = pd.DataFrame()
self.output_dataframe = pd.DataFrame()
def load_prev_gguid(self,save_path, recent_folder):
prev_gguid_path = os.path.join(save_path,recent_folder,'iteration_prev_guid.csv')
self.prev_gguid_df = pd.read_csv(prev_gguid_path)
self.recent_guid = list(self.prev_gguid_df['guid'])[-1]
return self.prev_gguid_df
def collect_csv(self):
'''
This function will call on the folder that was initalised when the class was called, and 'walk'
through it to collect all files, place them in a list. Once collected, it will check for only csv/CSV
files. These csv files will be read in by Pandas (encoding is due to the DLS data being in Hebrew) and that
data will be appended to the main data frame.
:return: A dataframe of the DLS data
'''
for (dirpath, dirnames, filenames) in os.walk(self.folder):
self.files.extend(filenames)
for f in self.files:
if f.endswith(('.csv', '.CSV')):
if str(self.recent_guid) in f:
print(str(f))
dw = pd.read_csv(os.path.join(self.folder, f), encoding="ISO-8859-8")
self.dls_data = self.dls_data.append(dw, ignore_index=True)
return self.dls_data
def clean_dls_data(self):
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.strip() # Clean up white spaces
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(?=)(\s)(\d)", "",
regex=True) # Remove the numbers proceeding the D(value)
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(?<=\d)(?=\-)", " ",
regex=True)
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(\s)(?<=)(\-)(?=)(\s)", "_",
regex=True) # Change the - into an _
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(?<=[A-Z]|\d)(\s)(?=\D)", "_",
regex=True) # Put an underscore between the GUID and D(value)
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(_FILTERED)", "",
regex=True) # Unique instance of putting the
# word filtered in dls naming
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.replace("(_FILTERED\d+)", "", regex=True)
self.dls_data['Sample Name'] = self.dls_data['Sample Name'].str.strip() # For good measure
self.list_dls_unique_scans = list(self.dls_data['Sample Name'].unique())
print("Number of unique DLS scans:", self.dls_data['Sample Name'].nunique())
print("DLS Samples Scanned: ", *iter(self.list_dls_unique_scans), sep=' | ')
def filter_out_D_iteration(self):
# Create the filter list of GUID
guid_list = list(self.dls_data['Sample Name'].str.replace("(?:\_\D\d)", "", regex=True).unique())
self.temp_dls_data = pd.DataFrame()
removal = 0
for index in guid_list:
temp_df = self.dls_data[self.dls_data['Sample Name'].str.contains(str(index))]
scan_iterations = list(temp_df['Sample Name'].str.replace("(\d+_)(?=\D)", "", regex=True).unique())
sorted_scan_iterations = sorted(scan_iterations, key=lambda x: int("".join([i for i in x if i.isdigit()])),
reverse=True)
temp_df = temp_df[temp_df['Sample Name'].str.contains(str(sorted_scan_iterations[0]))]
self.temp_dls_data = pd.concat([self.temp_dls_data, temp_df], ignore_index=True)
removal += (len(sorted_scan_iterations) - 1)
print("Number of scans removed: ", removal)
self.dls_data = self.temp_dls_data
self.list_dls_unique_scans = list(self.dls_data['Sample Name'].unique())
print("Number of unique DLS scans[Updated] :", self.dls_data['Sample Name'].nunique())
def z_scoring(self, threshold: float = 1.0):
'''
Create a DataFrame with the Z-Average (d.nm) [Average], PDI [Average] and PDI Width [Average]. Z score is
used to determine whether the DLS iterration scan values are within the threshold. If not, they are removed
and the mean is once again determined. A mean vs std is utilised when determining whether to use the Z score.
A Standard Deviation should not be larger than the Mean. Please note that the filtered out block of code is
the original code that does not take into account the Z score. If deemed better for results, then utilise/
PdI Width may be useless and there is argument to also bring in the volume and count values
:return:
'''
columns = [col for col in self.dls_data.columns if
'Z-Average' in col or 'Sample Name' in col or 'PdI' in col or 'PdI Width' in col]
dlsdata_average = pd.DataFrame(columns=columns)
z_score_filter_df = pd.DataFrame(columns=['Sample Name', 'Z-Average', 'PdI', 'PdI_Width'])
for index in self.list_dls_unique_scans:
temp_df = self.dls_data[self.dls_data['Sample Name'] == index]
temp_df = temp_df[['Z-Average (d.nm)', 'PdI', 'PdI Width (d.nm)']]
# create variable for checking
location_s = None
location_pdi = None
location_pdi_width = None
# Z-Average
temp_df_mean = temp_df['Z-Average (d.nm)'].mean()
temp_df_std = temp_df['Z-Average (d.nm)'].std()
if (temp_df_mean < temp_df_std) == True:
z_s = np.abs(zscore(temp_df['Z-Average (d.nm)']))
location_s = np.where(z_s > threshold)
for i in range(len(location_s[0])):
print("Dropping size: ", len(location_s[0]), str(index))
temp_df_s = temp_df.reset_index(drop=True).drop([location_s[0][i]])
temp_df_mean = temp_df_s['Z-Average (d.nm)'].mean()
if location_s is not None:
z_ave_cnt = len(location_s[0])
else:
z_ave_cnt = 0
# PdI
temp_df_mean_p = temp_df['PdI'].mean()
temp_df_std_p = temp_df['PdI'].std()
if (temp_df_mean_p < temp_df_std_p) == True:
z_pdi = np.abs(zscore(temp_df['PdI']))
location_pdi = np.where(z_pdi > threshold)
for i in range(len(location_pdi[0])):
print("Dropping PdI: ", len(location_pdi[0]), str(index))
temp_df_s = temp_df.reset_index(drop=True).drop([location_pdi[0][i]])
temp_df_mean_p = temp_df_s['PdI'].mean()
if location_pdi is not None:
pdi_ave_cnt = len(location_pdi[0])
else:
pdi_ave_cnt = 0
# PdI Width
temp_df_mean_pw = temp_df['PdI Width (d.nm)'].mean()
temp_df_std_pw = temp_df['PdI Width (d.nm)'].std()
if (temp_df_mean_pw < temp_df_std_pw) == True:
z_pdi_width = np.abs(zscore(temp_df['PdI Width (d.nm)']))
location_pdi_width = np.where(z_pdi_width > threshold)
for i in range(len(location_pdi_width[0])):
print("Dropping PdI Width: ", len(location_pdi_width[0]), str(index))
temp_df_s = temp_df.reset_index(drop=True).drop([location_pdi_width[0][i]])
temp_df_mean_pw = temp_df_s['PdI Width (d.nm)'].mean()
if location_pdi_width is not None:
pdiw_ave_cnt = len(location_pdi_width[0])
else:
pdiw_ave_cnt = 0
z_score_filter_df = z_score_filter_df.append({'Sample Name': str(index),
'Z-Average': int(z_ave_cnt),
'PdI': int(pdi_ave_cnt),
'PdI_Width': int(pdiw_ave_cnt)}, ignore_index=True)
dlsdata_average = dlsdata_average.append({'Sample Name': str(index),
'Z-Average (d.nm)': float(temp_df_mean),
'PdI': float(temp_df_mean_p),
'PdI Width (d.nm)': float(temp_df_mean_pw)}, ignore_index=True)
print(z_score_filter_df)
dlsdata_average['Sample Name'] = dlsdata_average['Sample Name'].str.replace("(?<!\_)(\D+\d)", "",
regex=True)
dlsdata_average['Sample Name'] = pd.to_numeric(dlsdata_average['Sample Name'])
self.dlsdata_average = dlsdata_average
def collect_formulations(self):
for (dirpath, dirnames, filenames) in os.walk(self.folder_formulation):
dirnames[:] = [d for d in dirnames if '_complete' in d]
for file in filenames:
if file.endswith(('.xlsx', '.XLSX')):
self.files_formulations.append(os.path.join(dirpath, file))
###Need to remove 'cv_results' as these are large files and grinds the for loop to a halt
self.files_formulations[:] = [d for d in self.files_formulations if 'cv_results' not in d]
for f in self.files_formulations:
if f.endswith(('.xlsx', '.XLSX')):
print(f)
if 'random' in f:
# print(str(f))
tdf_random = pd.read_excel(f)
self.random_dataframe = pd.concat([self.random_dataframe, tdf_random], ignore_index=True)
# self.dls_data = self.dls_data.append(dw, ignore_index=True)
elif 'Ouput' in f:
# Need to hardcode the skiprows for Output
tdf_output = pd.read_excel(f, skiprows=np.arange(13))
self.output_dataframe = pd.concat([self.output_dataframe, tdf_output], ignore_index=True)
# header=0)
# print(tdf_output)
self.output_dataframe = self.output_dataframe[['original_index',
'Concentration_1 (mM)',
'Ratio_1',
'Overall_Concentration_2',
'Ratio_2',
'Concentration_4',
'Ratio_4',
'Final_Vol',
'Lipid_Vol_Pcnt',
'Dispense_Speed_uls',
'mw_cp_1',
'xlogp_cp_1',
'complexity_cp_1',
'heavy_atom_count_cp_1',
'single_bond_cp_1',
'double_bond_cp_1',
'mw_cp_2',
'h_bond_acceptor_count_cp_2',
'xlogp_cp_2',
'complexity_cp_2',
'heavy_atom_count_cp_2',
'tpsa_cp_2',
'ssr_cp_2',
'single_bond_cp_2',
'double_bond_cp_2',
'aromatic_bond_cp_2',
'sample_scoring']]
#I have no idea where this issue came from but it has been persistent throughout
self.output_dataframe.loc[self.output_dataframe['mw_cp_2'] == 430.6999999999999, 'mw_cp_2'] = 430.7
self.output_dataframe.loc[self.output_dataframe['mw_cp_2'] == 386.699999999999, 'mw_cp_2'] = 386.7
self.output_dataframe.loc[self.output_dataframe['mw_cp_2'] == 386.6999999999999, 'mw_cp_2'] = 386.7
self.output_dataframe.loc[self.output_dataframe['complexity_cp_1'] == 856.9999999999999, 'complexity_cp_1'] = 857
self.output_dataframe.loc[self.output_dataframe['Ratio_2'] == 0.44999999999999996, 'Ratio_2'] = 0.45
self.output_dataframe.loc[self.output_dataframe['Ratio_2'] == 0.4499999999999999, 'Ratio_2'] = 0.45
self.output_dataframe_cols_rnd = self.output_dataframe.columns.difference(['Lipid_Vol_Pcnt'])
self.output_dataframe[self.output_dataframe_cols_rnd] = np.round(self.output_dataframe[self.output_dataframe_cols_rnd], 2)
self.output_dataframe.dropna(inplace=True)
self.random_dataframe.dropna(how='all', inplace=True)
self.random_dataframe.drop(columns=['original_index'], inplace=True)
# Random is missing this column name for some reason
self.random_dataframe.rename(columns={'Unnamed: 0': 'original_index'}, inplace=True)
def merge_dls_data(self):
# self.dlsdata_average['Sample Name'] = self.dlsdata_average['Sample Name'].str.strip()
# self.output_dataframe['original_index'] = self.output_dataframe['original_index'].str.strip()
self.merged_dataframe_AL = pd.merge(self.output_dataframe, self.dlsdata_average,
how="inner",
left_on="original_index",
right_on="Sample Name").reset_index(drop=True)
print(self.merged_dataframe_AL)
self.merged_dataframe_random = pd.merge(self.random_dataframe, self.dlsdata_average,
how="inner",
left_on="original_index",
right_on="Sample Name").reset_index(drop=True)
###Don't know why this is happpening but quick fix
#self.merged_dataframe_AL['ethanol_dil'] = 0.00
#self.merged_dataframe_random['ethanol_dil'] = 0.00
print(self.merged_dataframe_random)
def load_x_y_prev_run(self, save_path, recent_folder):
recent_iteration_run = recent_folder
path_of_file = os.path.join(save_path, recent_iteration_run, "Added_Data.csv")
if os.path.exists(path_of_file):
prev_x_y_df = pd.read_csv(path_of_file)
prev_x_y_df['mw_cp_2'] = prev_x_y_df['mw_cp_2'].round(2)
prev_x_y_df.loc[prev_x_y_df['Ratio_2'] == 0.44999999999999996] = 0.45
prev_x_df = prev_x_y_df.drop(columns=['Z-Average (d.nm)']).to_numpy()
prev_y_df = prev_x_y_df['Z-Average (d.nm)'].to_numpy().reshape(-1)
else:
prev_x_df = None
prev_y_df = None
return prev_x_df, prev_y_df
def load_x_val_prev_run(self, save_path, recent_folder):
# Need most recent folder name
# Then load the X_val file if exists.
x_val_prev_index = None
recent_iteration_run = recent_folder
path_of_file = os.path.join(save_path, recent_iteration_run, "Unlabeled_Data.csv")
if os.path.exists(path_of_file):
X_val_df = pd.read_csv(path_of_file)
X_val_df.loc[X_val_df['Ratio_2'] == 0.44999999999999996] = 0.45
if 'original_index' in X_val_df.columns:
X_val_df.drop(columns=['original_index'], inplace=True)
if 'Unnamed: 0' in X_val_df:
x_val_prev_index = X_val_df['Unnamed: 0'].copy()
X_val_df.drop(columns=['Unnamed: 0'], inplace=True)
print("Ingest X_VAL_PREV_RUN")
print(X_val_df['mw_cp_2'].value_counts())
X_val = X_val_df.to_numpy()
X_val_column_names = X_val_df.columns
return X_val, X_val_column_names, x_val_prev_index
def remove_AL_from_unlabelled(self, X_val, X_val_columns_names,x_prev_index, count):
X_val = pd.DataFrame(X_val, columns=X_val_columns_names)
if count<1:
temporary_AL_dataframe = self.merged_dataframe_AL.drop(columns=['original_index', 'sample_scoring',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
merge_dfs = X_val.merge(temporary_AL_dataframe.drop_duplicates(), on=list(temporary_AL_dataframe),
how='left', indicator=True)
merge_dfs = merge_dfs[merge_dfs['_merge'] == 'left_only']
merge_dfs.drop(columns=['_merge'], inplace=True)
print("No. of removed unlabelled formulations: ", X_val.shape[0] - merge_dfs.shape[0])
X_val = merge_dfs.to_numpy()
else:
temporary_AL_dataframe = self.merged_dataframe_AL.drop(columns=['sample_scoring',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
X_val['original_index'] = x_prev_index
merge_dfs = X_val.merge(temporary_AL_dataframe['original_index'], on=['original_index'], how='left',
indicator=True)
merge_dfs = merge_dfs[merge_dfs['_merge'] == 'left_only']
merge_dfs.drop(columns=['_merge','original_index'], inplace=True)
print("No. of removed unlabelled formulations: ", X_val.shape[0] - merge_dfs.shape[0])
X_val = merge_dfs.to_numpy()
return X_val, X_val_columns_names
def return_AL_data(self):
pre_x_train = self.merged_dataframe_AL.drop(columns=['original_index', 'sample_scoring',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
self.X_train_AL = pre_x_train.to_numpy()
pre_y_train = self.merged_dataframe_AL[['Z-Average (d.nm)']]
self.y_train_AL = pre_y_train.to_numpy().reshape(-1)
return self.X_train_AL, self.y_train_AL
def add_AL_to_train(self, X_train_initial, y_train_initial,
X_train_prev, y_train_prev):
if X_train_prev is not None and y_train_prev is not None:
X_train = np.vstack((X_train_initial, X_train_prev, self.X_train_AL))
y_train = np.hstack((y_train_initial, y_train_prev, self.y_train_AL)).astype(float)
else:
X_train = np.vstack((X_train_initial, self.X_train_AL))
y_train = np.hstack((y_train_initial, self.y_train_AL)).astype(float)
return X_train, y_train
def remove_random_from_unlabelled(self, X_val, X_val_columns_names, x_prev_index, count):
X_val = pd.DataFrame(X_val, columns=X_val_columns_names)
if 'sample_scoring' in self.merged_dataframe_random.columns:
temporary_random_dataframe = self.merged_dataframe_random.drop(columns=['original_index', 'sample_scoring',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
else:
temporary_random_dataframe = self.merged_dataframe_random.drop(columns=['original_index',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
if count <= 1:
merge_dfs = X_val.merge(temporary_random_dataframe.drop_duplicates(), on=list(temporary_random_dataframe),
how='left', indicator=True)
merge_dfs = merge_dfs[merge_dfs['_merge'] == 'left_only']
merge_dfs.drop(columns=['_merge'], inplace=True)
print("No. of removed unlabelled formulations: ", X_val.shape[0] - merge_dfs.shape[0])
X_val = merge_dfs.to_numpy()
else:
X_val['original_index'] = x_prev_index
merge_dfs = X_val.merge(temporary_random_dataframe['original_index'], on=['original_index'], how='left',
indicator=True)
merge_dfs = merge_dfs[merge_dfs['_merge'] == 'left_only']
merge_dfs.drop(columns=['_merge','original_index'], inplace=True)
print("No. of removed unlabelled formulations: ", X_val.shape[0] - merge_dfs.shape[0])
X_val = merge_dfs.to_numpy()
return X_val, X_val_columns_names
def return_random_data(self):
if 'sample_scoring' in self.merged_dataframe_random.columns:
pre_x_train = self.merged_dataframe_random.drop(columns=['original_index', 'sample_scoring',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
else:
pre_x_train = self.merged_dataframe_random.drop(columns=['original_index',
'Sample Name', 'Z-Average (d.nm)',
'PdI', 'PdI Width (d.nm)'])
self.X_train_random = pre_x_train.to_numpy()
pre_y_train = self.merged_dataframe_random[['Z-Average (d.nm)']]
self.y_train_random = pre_y_train.to_numpy().reshape(-1)
return self.X_train_random, self.y_train_random
def add_random_to_train(self, X_train_initial, y_train_initial,
X_train_prev, y_train_prev):
if X_train_prev is not None and y_train_prev is not None:
X_train = np.vstack((X_train_initial, X_train_prev, self.X_train_random))
y_train = np.hstack((y_train_initial, y_train_prev, self.y_train_random)).astype(float)
else:
X_train = np.vstack((X_train_initial, self.X_train_random))
y_train = np.hstack((y_train_initial, self.y_train_random)).astype(float)
return X_train, y_train
def add_random_adjusted_to_train(self, X_train_initial, y_train_initial,
X_train_prev, y_train_prev):
if X_train_prev is not None and y_train_prev is not None:
X_train = np.vstack((X_train_initial, X_train_prev, self.x_random_train))
y_train = np.hstack((y_train_initial, y_train_prev, self.y_random_train)).astype(float)
else:
X_train = np.vstack((X_train_initial, self.x_random_train))
y_train = np.hstack((y_train_initial, self.y_random_train)).astype(float)
return X_train, y_train
def return_random_adjusted(self, column_selection, save_path, folder):
folder_file = os.path.join(save_path, folder, "random_data_iteration.xlsx")
self.random_adjusted_data = pd.read_excel(folder_file)
self.x_random_train = self.random_adjusted_data[column_selection].to_numpy()
self.y_random_train = self.random_adjusted_data['Z-Average (d.nm)'].to_numpy().reshape(-1)
return self.x_random_train, self.y_random_train
##Test
# X_Val Function
# def unlabelled_data(file, method, column_removal_experiment: list=None):
# ul_df = pd.read_csv(file)
# column_drop = ['Duplicate_Check',
# 'PdI Width (d.nm)',
# 'PdI',
# 'Z-Average (d.nm)',
# 'ES_Aggregation']
#
#
# ul_df = ul_df.drop(columns=column_drop)
# #ul_df = ul_df.drop(columns=useless_clm_drop)
#
# #Remove a column(s)
# if column_removal_experiment is not None:
# ul_df.drop(columns=column_removal_experiment, inplace=True)
#
#
# ul_df.replace(np.nan, 'None', inplace=True)
# if "Component_1" and "Component_2" and "Component_3" in ul_df.columns:
# ul_df = pd.get_dummies(ul_df, columns=["Component_1", "Component_2", "Component_3"],
# prefix="", prefix_sep="")
# # if method=='fillna': ul_df['Component_3'] = ul_df['Component_3'].apply(lambda x: None if pd.isnull(x) else x) #TODO This should be transformed into an IF function, thus when the function for unlabelled is filled with a parameter, then activates
#
# ul_df = ul_df.groupby(level=0, axis=1, sort=False).sum()
#
# # print(ul_df.isna().any())
# X_val = ul_df.to_numpy()
# columns_x_val = ul_df.columns
# return X_val, columns_x_val
#
#
# column_removal_experiment = ['xlogp_cp_1', 'xlogp_cp_2', 'xlogp_cp_3',
# 'aromatic_bond_cp_2', 'complexity_cp_3',
# 'complexity_cp_2', 'complexity_cp_1', 'Final_Concentration',
# 'final_lipid_volume', 'component_1_vol', 'component_2_vol',
# 'component_3_vol', 'component_4_vol', 'component_1_vol_conc',
# 'tpsa_cp_3', 'tpsa_cp_2',
# # 'heavy_atom_count_cp_1','heavy_atom_count_cp_2','heavy_atom_count_cp_3',
# # 'single_bond_cp_1','double_bond_cp_1',
# # 'single_bond_cp_2','double_bond_cp_2',
# # 'single_bond_cp_3','double_bond_cp_3',
# # 'h_bond_donor_count_cp_2','h_bond_acceptor_count_cp_2',
# # 'h_bond_donor_count_cp_3','h_bond_acceptor_count_cp_3',
# # 'ssr_cp_2','ssr_cp_3',
# 'Req_Weight_1', 'Ethanol_1',
# 'Req_Weight_2', 'Ethanol_2',
# 'Req_Weight_3',
# 'Req_Weight_4', 'Ethanol_4'
# # 'ethanol_dil',
# # 'component_1_vol_stock',
# # 'component_2_vol_stock',
# # 'component_3_vol_stock'
#
# ]
# X_val, columns_x_val = unlabelled_data(file = r"/Users/calvin/Documents/OneDrive/Documents/2022/Data_Output/unlabelled_data_full.csv",
# method='fillna',
# column_removal_experiment=column_removal_experiment) # TODO need to rename
#
# test = ALDataBuild(
# folder_dls=r'/Users/calvin/Library/CloudStorage/OneDrive-Personal/Documents/2022/RegressorCommittee_Input',
# folder_formulations=r'/Users/calvin/Library/CloudStorage/OneDrive-Personal/Documents/2022/RegressorCommittee_Output')
#
# test.collect_csv()
# test.clean_dls_data()
# test.filter_out_D_iteration()
# test.z_scoring(threshold=1.0)
# test.collect_formulations()
# test.merge_dls_data()
# test.remove_from_unlabelled(X_val = X_val, X_val_columns_names=columns_x_val)