Added label distribution

preprocess_scripts/compute_distribution_visual_scene_labels.py

@@ -0,0 +1,61 @@
+import os 
+import pandas as pd 
+import numpy as np
+from tqdm import tqdm 
+import pickle 
+from collections import Counter
+#read each file and extract the labels
+
+def read_txt_file(txt_file_path):
+
+    label_list=[]
+    print('Loading the file:%s' %(txt_file_path))
+    with open(txt_file_path, 'r') as f:
+        lines = f.readlines()
+        #lines = [x.strip() for x in lines]
+
+
+    for line in tqdm(lines):
+        label_c=line.strip().split(' ')[1:]
+        label_c=[int(l) for l in label_c]
+        label_list=label_list+label_c
+
+    return label_list
+
+
+train_file="../split_files/train_multi_label_thresh_0_4_0_1_150_labels.txt"
+val_file="../split_files/val_multi_label_thresh_0_4_0_1_150_labels.txt"
+test_file="../split_files/test_multi_label_thresh_0_4_0_1_150_labels.txt"
+label_map_file="../split_files/label_2_ind_multi_label_thresh_0_4_0_1_150_label_map.pkl"
+
+
+train_labels=read_txt_file(train_file)
+val_labels=read_txt_file(val_file)
+test_labels=read_txt_file(test_file)
+
+total_labels=train_labels+val_labels+test_labels
+
+with open(label_map_file, 'rb') as f:
+    label_map = pickle.load(f)
+
+#obtain the reverse map 
+reverse_label_map={v:k for k,v in label_map.items()}
+
+#compute the distribution of the labels
+label_names=[reverse_label_map[l] for l in total_labels]
+
+label_occurence=Counter(label_names).most_common(150)
+
+#save counter dict as dataframe
+df=pd.DataFrame(label_occurence,columns=['label','count'])
+
+df.to_csv('../split_files/label_distribution_multi_label_thresh_0_4_0_1_150.csv',index=False)
+
+
+
+
+
+
+
+
+

split_files/label_distribution_multi_label_thresh_0_4_0_1_150.csv

@@ -0,0 +1,151 @@
+label,count
+cockpit,12426
+car,11128
+locker room,8646
+ballroom,7786
+banquet,6531
+bedroom,5126
+cab,4603
+stage,4347
+funeral,4242
+shooting range,3611
+boxing ring,3422
+helicopter,3198
+club,3087
+courtroom,2962
+dining room,2879
+truck,2863
+animal shelter,2620
+control room,2458
+bathroom,2427
+war room,2418
+elevator,2371
+morgue,2144
+kitchen,2119
+bar,1849
+plane,1773
+living room,1719
+automotive repair,1651
+pool,1526
+batting cage,1498
+desert,1449
+baseball field,1322
+boat,1264
+room,1237
+concert hall,1201
+basketball court,1187
+shuttle,1168
+sea,1148
+zoo,977
+gym,977
+classroom,976
+closet,955
+corridor,862
+computer room,856
+race track,837
+arena,835
+battlefield,834
+bus,791
+lounge,664
+cave,662
+penthouse,654
+hospital,627
+salon,626
+makeup studio,612
+ship,611
+balcony,592
+stairs,568
+stadium,565
+ice rink,554
+train,546
+fair,539
+lobby,531
+deck,521
+beach,508
+casino,504
+restaurant,458
+attic,457
+foundry,423
+subway,370
+laboratory,366
+tunnel,358
+bowling alley,342
+suburban,318
+overpass,309
+theater,299
+auditorium,295
+shore,293
+retail,277
+construction site,273
+parking,256
+basement,238
+wagon,231
+press room,226
+road,221
+conference room,221
+swamp,216
+golf course,213
+hot spring,209
+graveyard,201
+clinic,197
+waterfall,194
+tent,180
+mountain,179
+forest,164
+grove,161
+river,157
+bullring,156
+bridge,154
+office,151
+playground,151
+mall,151
+skyline,150
+tennis court,140
+studio,121
+garden,120
+amusement park,110
+market,106
+tower,104
+archaeological site,103
+gas station,101
+cafe,96
+sandbank,96
+downtown,88
+agriculture field,82
+prison,80
+library,79
+art gallery,78
+factory,77
+airport,76
+bakery,70
+apartment,65
+cabin,64
+farm,63
+pond,61
+kindergarten,60
+school,59
+garage,54
+hangar,51
+booth,49
+castle,41
+cellar,40
+alley,37
+hall,35
+dorm,34
+mansion,34
+fire station,33
+station,33
+chapel,32
+yard,32
+church,30
+walkway,28
+police station,25
+loft,20
+park,15
+lake,15
+valley,14
+harbor,12
+inn,10
+plaza,10
+street,10
+grassland,8