darcwader

//https://api.flickr.com/services/rest/?method=flickr.interestingness.getList&api_key=aa621a9050ef8dfbd9621cc311da86aa&format=json


-(NSString*) flickrInterestingGenerateUrlFromInfo:(NSDictionary*) info {
	return [NSString stringWithFormat:@"http://farm%@.static.flickr.com/%@/%@_%@_t.jpg",
			[info objectForKey:@"farm"],
			[info objectForKey:@"server"],
			[info objectForKey:@"id"],
			[info objectForKey:@"secret"]];
}

darcwader

sudo apt-get install libgnome-keyring-dev
sudo make --directory=/usr/share/doc/git/contrib/credential/gnome-keyring
git config --global credential.helper /usr/share/doc/git/contrib/credential/gnome-keyring/git-credential-gnome-keyring

darcwader

final class DataCoordinator {
    //MARK: - singleton
    private static var coordinator: DataCoordinator?
    public class func sharedInstance() -> DataCoordinator {
        if coordinator == nil {
            coordinator = DataCoordinator()
        }
        return coordinator!
    }
    

darcwader

file_name = 'SMSSpamCollection'
with open(file_name) as f:
    corpus = f.readlines()

print("there are {} messages".format(len(corpus)))
    
corpus = [x.strip() for x in corpus] #remove trailing \n from lines

for i,message in enumerate(corpus[:10]):
    print(i, message)

darcwader

with open('SMSSpamCollection',encoding='UTF-8') as f:
    messages = pd.read_csv(f, sep='\t', quoting=csv.QUOTE_NONE, names=['label', 'message'])

messages['message'] = messages['message'].map(lambda text:text.decode(encoding='utf-8'))
messages.head()

darcwader

print("removing punctuations: "+string.punctuation)
stemmer = PorterStemmer()

def tokenize(message):
    """ removes punctuation and tokenizes the words and stems each word.
    """
    msg = "".join([ch for ch in message if ch not in string.punctuation]) # get rid of punctuations
    tokens = word_tokenize(msg) 
    #stems = [stemmer.stem(x).lower() for x in tokens] #correct way to do
    stems = [x.lower() for x in tokens] #iOS does not have porterstemmer, we are going to not use stem for now

darcwader

idf = {}
idf['idf'] = tfidf.idf_.tolist()

with open('words_idf.json', 'wb') as fp:
    json.dump(idf, fp)

print("IDF of corpus :", tfidf.idf_)

darcwader

%time spam_detector = LinearSVC().fit(messages_tfidf, messages.label)

predictions = spam_detector.predict(messages_tfidf)

print('accuracy', accuracy_score(messages['label'], predictions))
print('confusion matrix\n', confusion_matrix(messages['label'], predictions))
print('(row=expected, col=predicted)')

darcwader

plt.matshow(confusion_matrix(messages['label'], predictions), cmap=plt.cm.binary, interpolation='nearest')
plt.title('confusion matrix')
plt.colorbar()
plt.ylabel('expected label')
plt.xlabel('predicted label')

darcwader

import coremltools
coreml_model = coremltools.converters.sklearn.convert(spam_detector, "message", "spam_or_not")

#set parameters of the model
coreml_model.short_description = "Classify whether message is spam or not"
coreml_model.input_description["message"] = "TFIDF of message to be classified"
coreml_model.output_description["spam_or_not"] = "Whether message is spam or not"
#save the model
coreml_model.save("SpamMessageClassifier.mlmodel")