33     samples, responses = a[:,1:], a[:,0]
34     return samples, responses
45     def unroll_samples(self, samples):  argument
46         sample_n, var_n = samples.shape
48         new_samples[:,:-1] = np.repeat(samples, self.class_n, axis=0)
63     def train(self, samples, responses):  argument
64         sample_n, var_n = samples.shape
68 …      self.model.train(samples, cv2.CV_ROW_SAMPLE, responses, varType = var_types, params = params)
70     def predict(self, samples):  argument
71         return np.float32( [self.model.predict(s) for s in samples] )
78     def train(self, samples, responses):  argument
79         self.model.train(samples, responses)
81     def predict(self, samples):  argument
82         retval, results, neigh_resp, dists = self.model.find_nearest(samples, k = 10)
90     def train(self, samples, responses):  argument
91         sample_n, var_n = samples.shape
92         new_samples = self.unroll_samples(samples)
99     def predict(self, samples):  argument
100         new_samples = self.unroll_samples(samples)
110     def train(self, samples, responses):  argument
114         self.model.train(samples, responses, params = params)
116     def predict(self, samples):  argument
117         return self.model.predict_all(samples).ravel()
124     def train(self, samples, responses):  argument
125         sample_n, var_n = samples.shape
136         self.model.train(samples, np.float32(new_responses), None, params = params)
138     def predict(self, samples):  argument
139         ret, resp = self.model.predict(samples)
159     samples, responses = load_base(args['--data'])  variable
163     train_n = int(len(samples)*model.train_ratio)
170         model.train(samples[:train_n], responses[:train_n])
173     train_rate = np.mean(model.predict(samples[:train_n]) == responses[:train_n])
174     test_rate  = np.mean(model.predict(samples[train_n:]) == responses[train_n:])