{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np \nimport pandas as pd \nimport matplotlib.pyplot as plt\nimport matplotlib.image as mpimg\nimport os\nimport json\nimport keras","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"print(os.listdir(\"../input\"))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data = pd.read_csv('../input/train.csv')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print('Number of unique heights = {0}'.format(train_data.Height.unique().shape[0]))\nprint('Minimum height = {0}'.format(train_data.Height.min()))\nprint('Maximum height = {0}'.format(train_data.Height.max()))\nprint('Average height = {0}'.format(train_data.Height.mean()))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print('Number of unique widths = {0}'.format(train_data.Width.unique().shape[0]))\nprint('Minimum width = {0}'.format(train_data.Width.min()))\nprint('Maximum width = {0}'.format(train_data.Width.max()))\nprint('Average width = {0}'.format(train_data.Width.mean()))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.hist(train_data.groupby(['ImageId']).agg('mean')['Height'], bins=100)\nplt.xlabel('Height of images')\nplt.ylabel('Number of images')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.hist(train_data.groupby(['ImageId']).agg('mean')['Width'], bins=100)\nplt.xlabel('Width of images')\nplt.ylabel('Number of images')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.hist(train_data.groupby(['ImageId']).agg('mean')['Height']/train_data.groupby(['ImageId']).agg('mean')['Width'], bins=20)\nplt.xlabel('Ratio (height / width)')\nplt.ylabel('Number of images')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Most images have height > width.
\nThe average height = 2236
\nThe average width = 1762
\nWe may consider this when resizing the image if we will resize them to the same size.
"},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.ImageId.unique().shape[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"num_classes = train_data.groupby(by=['ImageId']).count()['ClassId'].tolist()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print('Min number of classes is {0} and maximum number is {1}'.format(min(num_classes), max(num_classes)))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.hist(num_classes, bins=74)\nplt.xlabel('Number of classes for one image')\nplt.ylabel('Number of images')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data.ClassId.unique().shape[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"with open('../input/label_descriptions.json') as json_file: \n labels_desc = json.load(json_file)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"labels_desc.keys()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"labels_desc['info']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"categories = pd.DataFrame(labels_desc['categories'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"attributes = pd.DataFrame(labels_desc['attributes'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"categories.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"attributes.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"categories","execution_count":94,"outputs":[{"output_type":"execute_result","execution_count":94,"data":{"text/plain":" id ... supercategory\n0 0 ... upperbody\n1 1 ... upperbody\n2 2 ... upperbody\n3 3 ... upperbody\n4 4 ... upperbody\n5 5 ... upperbody\n6 6 ... lowerbody\n7 7 ... lowerbody\n8 8 ... lowerbody\n9 9 ... wholebody\n10 10 ... wholebody\n11 11 ... wholebody\n12 12 ... wholebody\n13 13 ... head\n14 14 ... head\n15 15 ... head\n16 16 ... neck\n17 17 ... arms and hands\n18 18 ... arms and hands\n19 19 ... waist\n20 20 ... legs and feet\n21 21 ... legs and feet\n22 22 ... legs and feet\n23 23 ... legs and feet\n24 24 ... others\n25 25 ... others\n26 26 ... others\n27 27 ... garment parts\n28 28 ... garment parts\n29 29 ... garment parts\n30 30 ... garment parts\n31 31 ... garment parts\n32 32 ... garment parts\n33 33 ... garment parts\n34 34 ... closures\n35 35 ... closures\n36 36 ... decorations\n37 37 ... decorations\n38 38 ... decorations\n39 39 ... decorations\n40 40 ... decorations\n41 41 ... decorations\n42 42 ... decorations\n43 43 ... decorations\n44 44 ... decorations\n45 45 ... decorations\n\n[46 rows x 4 columns]","text/html":"
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
idlevelnamesupercategory
002shirt, blouseupperbody
112top, t-shirt, sweatshirtupperbody
222sweaterupperbody
332cardiganupperbody
442jacketupperbody
552vestupperbody
662pantslowerbody
772shortslowerbody
882skirtlowerbody
992coatwholebody
10102dresswholebody
11112jumpsuitwholebody
12122capewholebody
13132glasseshead
14142hathead
15152headband, head covering, hair accessoryhead
16162tieneck
17172glovearms and hands
18182watcharms and hands
19192beltwaist
20202leg warmerlegs and feet
21212tights, stockingslegs and feet
22222socklegs and feet
23232shoelegs and feet
24242bag, walletothers
25252scarfothers
26262umbrellaothers
27272hoodgarment parts
28282collargarment parts
29292lapelgarment parts
30302epaulettegarment parts
31312sleevegarment parts
32322pocketgarment parts
33332necklinegarment parts
34342buckleclosures
35352zipperclosures
36362appliquedecorations
37372beaddecorations
38382bowdecorations
39392flowerdecorations
40402fringedecorations
41412ribbondecorations
42422rivetdecorations
43432ruffledecorations
44442sequindecorations
45452tasseldecorations
\n
"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"attributes","execution_count":95,"outputs":[{"output_type":"execute_result","execution_count":95,"data":{"text/plain":" id level name supercategory\n0 0 1 above-the-hip (length) length\n1 1 1 hip (length) length\n2 2 1 micro (length) length\n3 3 1 mini (length) length\n4 4 1 above-the-knee (length) length\n5 5 1 knee (length) length\n6 6 1 below the knee (length) length\n7 7 1 midi length\n8 8 1 maxi (length) length\n9 9 1 floor (length) length\n10 10 1 single breasted opening type\n11 11 1 double breasted opening type\n12 12 1 lace up opening type\n13 13 1 wrapping opening type\n14 14 1 zip-up opening type\n15 15 1 fly (opening) opening type\n16 16 1 buckled (opening) opening type\n17 17 1 toggled (opening) opening type\n18 18 1 no opening opening type\n19 19 1 asymmetrical silhouette\n20 20 1 symmetrical silhouette\n21 21 1 peplum silhouette\n22 22 1 circle silhouette\n23 23 1 flare silhouette\n24 24 1 fit and flare silhouette\n25 25 1 trumpet silhouette\n26 26 1 mermaid silhouette\n27 27 1 balloon silhouette\n28 28 1 bell silhouette\n29 29 1 bell bottom silhouette\n.. .. ... ... ...\n62 62 1 abstract textile pattern\n63 63 1 cartoon textile pattern\n64 64 1 letters, numbers textile pattern\n65 65 1 camouflage textile pattern\n66 66 1 check textile pattern\n67 67 1 dot textile pattern\n68 68 1 fair isle textile pattern\n69 69 1 floral textile pattern\n70 70 1 geometric textile pattern\n71 71 1 paisley textile pattern\n72 72 1 stripe textile pattern\n73 73 1 houndstooth (pattern) textile pattern\n74 74 1 herringbone (pattern) textile pattern\n75 75 1 chevron textile pattern\n76 76 1 argyle textile pattern\n77 77 2 leopard animal\n78 78 2 snakeskin (pattern) animal\n79 79 2 cheetah animal\n80 80 2 peacock animal\n81 81 2 zebra animal\n82 82 2 giraffe animal\n83 83 1 toile de jouy textile pattern\n84 84 1 plant textile pattern\n85 85 1 empire waistline waistline\n86 86 1 dropped waistline waistline\n87 87 1 high waist waistline\n88 88 1 normal waist waistline\n89 89 1 low waist waistline\n90 90 1 basque (wasitline) waistline\n91 91 1 no waistline waistline\n\n[92 rows x 4 columns]","text/html":"
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
idlevelnamesupercategory
001above-the-hip (length)length
111hip (length)length
221micro (length)length
331mini (length)length
441above-the-knee (length)length
551knee (length)length
661below the knee (length)length
771midilength
881maxi (length)length
991floor (length)length
10101single breastedopening type
11111double breastedopening type
12121lace upopening type
13131wrappingopening type
14141zip-upopening type
15151fly (opening)opening type
16161buckled (opening)opening type
17171toggled (opening)opening type
18181no openingopening type
19191asymmetricalsilhouette
20201symmetricalsilhouette
21211peplumsilhouette
22221circlesilhouette
23231flaresilhouette
24241fit and flaresilhouette
25251trumpetsilhouette
26261mermaidsilhouette
27271balloonsilhouette
28281bellsilhouette
29291bell bottomsilhouette
...............
62621abstracttextile pattern
63631cartoontextile pattern
64641letters, numberstextile pattern
65651camouflagetextile pattern
66661checktextile pattern
67671dottextile pattern
68681fair isletextile pattern
69691floraltextile pattern
70701geometrictextile pattern
71711paisleytextile pattern
72721stripetextile pattern
73731houndstooth (pattern)textile pattern
74741herringbone (pattern)textile pattern
75751chevrontextile pattern
76761argyletextile pattern
77772leopardanimal
78782snakeskin (pattern)animal
79792cheetahanimal
80802peacockanimal
81812zebraanimal
82822giraffeanimal
83831toile de jouytextile pattern
84841planttextile pattern
85851empire waistlinewaistline
86861dropped waistlinewaistline
87871high waistwaistline
88881normal waistwaistline
89891low waistwaistline
90901basque (wasitline)waistline
91911no waistlinewaistline
\n

92 rows × 4 columns

\n
"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}