MCPcopy Create free account
hub / github.com/AgriQuantAI/AgriQuant-AI / normalize_data

Method normalize_data

lstm_predictor.py:183–211  ·  view source on GitHub ↗

Normalize sequences and targets Args: sequences: Input sequences targets: Target values fit: Whether to fit normalization parameters Returns: Normalized sequences and targets

(
        self,
        sequences: np.ndarray,
        targets: np.ndarray,
        fit: bool = True
    )

Source from the content-addressed store, hash-verified

181 return np.array(sequences), np.array(targets)
182
183 def normalize_data(
184 self,
185 sequences: np.ndarray,
186 targets: np.ndarray,
187 fit: bool = True
188 ) -> Tuple[np.ndarray, np.ndarray]:
189 """
190 Normalize sequences and targets
191
192 Args:
193 sequences: Input sequences
194 targets: Target values
195 fit: Whether to fit normalization parameters
196
197 Returns:
198 Normalized sequences and targets
199 """
200 if fit:
201 # Calculate normalization parameters
202 self.feature_mean = sequences.mean(axis=(0, 1))
203 self.feature_std = sequences.std(axis=(0, 1)) + 1e-8
204 self.target_mean = targets.mean()
205 self.target_std = targets.std() + 1e-8
206
207 # Normalize
208 sequences_norm = (sequences - self.feature_mean) / self.feature_std
209 targets_norm = (targets - self.target_mean) / self.target_std
210
211 return sequences_norm, targets_norm
212
213 def train(
214 self,

Callers 1

trainMethod · 0.95

Calls

no outgoing calls

Tested by

no test coverage detected