#Implement a KNN classifier in Python from scratch. # Test it using cross-validation (train:test = 4:1) on IRIS dataset # which you can obtain it from scikit-learn load iris function with K = 1, 2, 3. # Your program will be runned by ”python p5.py” from IPython import embed import numpy as np from sklearn import datasets from knn import Knn iris = datasets.load_iris() X = iris.data y = iris.target knn = Knn(X, y, ks=[1, 2, 3]) knn.classify()
knn = Knn(k, train_set, train_key2ind)
print 'Reading test dataset..'
test_set, test_key2ind, test_ind2key = dataset_reader.read_dataset(
test_filename, test_filename + '.key', False, isolate_target_sentence)
print 'Starting to classify test set:'
with open(result_filename, 'w') as o:
for ind, key_set in enumerate(test_set):
key = test_ind2key[ind]
if debug:
print 'KEY:', key
print
for instance_id, vec, text in izip(key_set.instance_ids,
key_set.context_m,
key_set.contexts_str):
if debug:
print 'QUERY:', text.strip()
result = knn.classify(key, vec, ignore_closest, debug)
if debug:
print
#brother.n 00006 501566/0.5 501573/0.4 503751/0.1
result_line = key + ' ' + instance_id
for sid, weight in result.iteritems():
result_line += ' {}/{:.4f}'.format(sid, weight)
o.write(result_line + '\n')
if debug:
print 'LABELS FOUND: ', result_line
print
knn = Knn(k, train_set, train_key2ind)
print('Reading test dataset..')
test_set, test_key2ind, test_ind2key = dataset_reader.read_dataset(test_filename, test_filename+'.key', False, isolate_target_sentence)
print('Starting to classify test set:')
with open(result_filename, 'w') as o:
for ind, key_set in enumerate(test_set):
key = test_ind2key[ind]
if debug:
print('KEY:', key)
print()
for instance_id, vec, text in zip(key_set.instance_ids, key_set.context_m, key_set.contexts_str):
if debug:
print('QUERY:', text.strip())
result = knn.classify(key, vec, ignore_closest, debug)
if debug:
print()
#brother.n 00006 501566/0.5 501573/0.4 503751/0.1
result_line = key + ' ' + instance_id
for sid, weight in result.items():
result_line += ' {}/{:.4f}'.format(sid, weight)
o.write(result_line+'\n')
if debug:
print('LABELS FOUND: ', result_line)
print()
