"k11": 1,
"k12": 1,
"k13": 1,
"k14": 1,
"k15": 1,
}
test_data = pd.read_csv(open("data/test.csv", "r"), quotechar='"')
sub_data = pd.read_csv(open("data/sampleSubmission.csv", "r"), quotechar='"')
if not np.alltrue(test_data["id"] == sub_data["id"]):
raise Exception("IDs do not match")
yh = Yhat(username, apikey)
variabless = sub_data.columns[1:]
raw_tweets = test_data["tweet"].tolist()
for variable in variables:
model_version = best_model[variable]
model_name = "TweetClassifier_%s" % (variable,)
results_from_server = yh.raw_predict(model_name, model_version, raw_tweets)
pred = results_from_server["prediction"]["scores"]
sub_data[variable] = pred
try:
sub_data.to_csv(open(sub_file, "w"), index=False)
except IOError:
sys.stderr.write("IO error: could not write data to file")
print "Uploading to yhat"
upload_status = yh.upload(model_name,tweet_clf)
model_version = upload_status['version']
print "'%s':'%s' uploaded to yhat" % (model_name,model_version)
# Sanity check uploaded classifier by comparing remote against local scores
print "Preforming sanity check"
print "Predicting local scores"
local_sanity = tweet_clf.predict(tweet_clf.transform(sanity_raw))['scores']
local_sanity = np.array(local_sanity)
print "Getting scores from server"
results_from_server = yh.raw_predict(model_name,model_version,sanity_raw)
try:
server_sanity = results_from_server['prediction']['scores']
except:
print results_from_server
sys.exit(3)
server_sanity = np.array(server_sanity)
# Because of float point scores compare difference of scores to some level
# of tolerance rather than checking equality
score_diff = np.abs(local_sanity - server_sanity)
sanity_tolerance = 1e-3
sanity_status = np.alltrue(score_diff < sanity_tolerance)
if not sanity_status:
print "Uploading to yhat"
upload_status = yh.upload(model_name, tweet_clf)
model_version = upload_status['version']
print "'%s':'%s' uploaded to yhat" % (model_name, model_version)
# Sanity check uploaded classifier by comparing remote against local scores
print "Preforming sanity check"
print "Predicting local scores"
local_sanity = tweet_clf.predict(tweet_clf.transform(sanity_raw))['scores']
local_sanity = np.array(local_sanity)
print "Getting scores from server"
results_from_server = yh.raw_predict(model_name, model_version, sanity_raw)
try:
server_sanity = results_from_server['prediction']['scores']
except:
print results_from_server
sys.exit(3)
server_sanity = np.array(server_sanity)
# Because of float point scores compare difference of scores to some level
# of tolerance rather than checking equality
score_diff = np.abs(local_sanity - server_sanity)
sanity_tolerance = 1e-3
sanity_status = np.alltrue(score_diff < sanity_tolerance)
if not sanity_status:
's1':1, 's2':1, 's3':1, 's4':1, 's5':1,
'w1':1, 'w2':1, 'w3':1, 'w4':1,
'k1':1, 'k2':1, 'k3':1, 'k4':1, 'k5':1,
'k6':1, 'k7':1, 'k8':1, 'k9':1, 'k10':1,
'k11':1, 'k12':1, 'k13':1, 'k14':1, 'k15':1
}
test_data = pd.read_csv(open('data/test.csv', 'r'), quotechar='"')
sub_data = pd.read_csv(open('data/sampleSubmission.csv', 'r'), quotechar='"')
if not np.alltrue(test_data['id'] == sub_data['id']):
raise Exception("IDs do not match")
yh = Yhat(username, apikey)
variabless = sub_data.columns[1:]
raw_tweets = test_data['tweet'].tolist()
for variable in variables:
model_version = best_model[variable]
model_name = "TweetClassifier_%s" % (variable, )
results_from_server = yh.raw_predict(model_name, model_version, raw_tweets)
pred = results_from_server['prediction']['scores']
sub_data[variable] = pred
try:
sub_data.to_csv(open(sub_file, 'w'), index=False)
except IOError:
sys.stderr.write("IO error: could not write data to file")
# <codecell>
yh = Yhat("YOUR USERNAME", "YOUR API KEY")
# <codecell>
print yh.upload("NamedEntityFindr", clf)
# <codecell>
[model for model in yh.show_models()['models'] if model['name'] == "NamedEntityFindr"]
# <codecell>
results_from_server = yh.raw_predict("NamedEntityFindr", 1, data)
results_from_server
# <codecell>
print 'sanity check.'
print 'results all match => %s' \
% np.all(np.array(results['entities']) == np.array(results_from_server['prediction']['entities']))
# <markdowncell>
# <h2>Final Thoughts</h2>
# <ul>
# <li><a href="http://nltk.googlecode.com/svn/trunk/doc/book/ch05.html" title="Categorizing and Tagging Words - NLTK docs" target="_blank">Categorizing and Tagging Words with NLTK</a> (NLTK docs)</li>
# <li><a href="http://pixelmonkey.org/pub/nlp-training/" title="Just Enough NLP with Python" target="_blank">Just Enough NLP with Python</a> (slides)</li>
# <li><a href="http://cdn.preterhuman.net/texts/science_and_technology/artificial_intelligence/Foundations%20of%20Statistical%20Natural%20Language%20Processing%20-%20Christopher%20D.%20Manning.pdf" title="Foundations of Statistical Natural Language Processing by Christopher Manning & Hinrich Schiitze" target="_blank">Foundations of Statistical Natural Language Processing</a> by Christopher Manning & Hinrich Schiitze (PDF)</li>
