In [18]:

    

import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.cross_validation import cross_val_score
import numpy as np
from skimage.io import imread, imshow, imsave
from skimage.feature import blob_doh
from skimage.color import rgb2gray
from skimage.transform import resize
from glob import glob
import os
from sklearn.feature_extraction.text import CountVectorizer
import matplotlib.pyplot as plt


    

In [2]:

    

%matplotlib inline


    

In [3]:

    

trainlabel = pd.read_csv('../Data/train.csv')


    

In [4]:

    

whaleIDset = set(trainlabel['whaleID'])


    

In [5]:

    

print(len(whaleIDset), len(trainlabel))


    

    




(447, 4544)

In [29]:

    

im1 = imread('../../BigData/kaggle-right-whale/imgs_subset/w_300.jpg')


    

In [30]:

    

imshow(im1)


    

    
Out[30]:





<matplotlib.image.AxesImage at 0x10cadbc50>






    

In [17]:

    

image_gray = rgb2gray(im1)


    

In [25]:

    

blobs_doh = blob_doh(image_gray, max_sigma=300, threshold=0.1, min_sigma=2)


    

In [26]:

    

import matplotlib.pyplot as plt
fig, ax = plt.subplots(1, 1)
ax.imshow(im1, interpolation='nearest')
for blob in blobs_doh:
    y, x, r = blob
    c = plt.Circle((x, y), r, color='red', linewidth=2, fill=False)
    ax.add_patch(c)


    

In [40]:

    

im2 = resize(im1,(200,300,3))


    

In [41]:

    

imshow(im2)


    

    
Out[41]:





<matplotlib.image.AxesImage at 0x118c4b090>






    

In [43]:

    

files = glob('../../BigData/kaggle-right-whale/imgs/*jpg')


    

In [48]:

    

for file in files:
    im1 = imread(file)
    imsmall = resize(im1, (200,300,3))
    splitfile = os.path.splitext(file)
    fsmall = splitfile[0] + '_small' + splitfile[1]
    imsave(fsmall, imsmall)


    

In [7]:

    

dloc = '../../BigData/kaggle-right-whale/imgs/'


    

In [8]:

    

#features = []
#for image in trainlabel['Image']:
#    dfile = dloc + image
#    splitfile = os.path.splitext(file)
#    fsmall = splitfile[0] + '_small' + splitfile[1]
#    imdata = imread(fsmall)
#    features.append(imdata.flatten())
features = []
foundlist = []
files = glob(dloc + '*_small.jpg')
for ifile in files:
    si = str(i)
    tailloc = 'w_' + si + '_small.jpg'
    dfile = dloc + tailloc
    if 'w_' + si + '.jpg' in trainlabel['Image'].values and dfile in files:
        imdata = imread(dfile)
        features.append(imdata.flatten())
        trainlabelindex = trainlabel['Image'].values == 'w_' + si + '.jpg'
        trainlabelvalue = trainlabel['whaleID'].values[trainlabelindex][0]
        foundlist.append(trainlabelvalue)


    

In [9]:

    

features = np.array(features)


    

In [10]:

    

features.shape


    

    
Out[10]:





(4543, 180000)

In [11]:

    

vectorizer = CountVectorizer(min_df=1)


    

In [12]:

    

X = features
y = vectorizer.fit_transform(foundlist).toarray()


    

In [16]:

    

ysum = np.sum(y, axis=0)


    

In [25]:

    

plt.plot(ysum, '.')


    

    
Out[25]:





[<matplotlib.lines.Line2D at 0x10c332d90>]






    

In [61]:

    

clf = RandomForestClassifier()


    

In [ ]:

    

scores = cross_val_score(clf, X, y, cv=5)
print(scores, scores.mean(), scores.std())


    

In [11]:

    

clf.fit(X, y)


    

    
Out[11]:





RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',
            max_depth=None, max_features='auto', max_leaf_nodes=None,
            min_samples_leaf=1, min_samples_split=2,
            min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,
            oob_score=False, random_state=None, verbose=0,
            warm_start=False)

In [12]:

    

test = pd.read_csv(dloc + 'test.csv')


    

In [14]:

    

Xtest = test


    

In [15]:

    

ypredict = clf.predict(Xtest)


    

In [16]:

    

ypredict


    

    
Out[16]:





array([2, 0, 9, ..., 3, 9, 2])

In [17]:

    

dfpredict = pd.DataFrame(ypredict)


    

In [22]:

    

dfpredict.columns = ['Label']


    

In [26]:

    

dfpredict['ImageId'] = np.arange(28000) + 1


    

In [27]:

    

dfpredict.to_csv(dloc + 'predict_RFbenchmark.csv', index=False)


    

In [ ]: