File size: 11,068 Bytes
c4b87d2
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
0a58567
 
 
 
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0a58567
c4b87d2
0a58567
c4b87d2
 
 
 
 
 
0a58567
c4b87d2
 
 
0a58567
c4b87d2
 
 
0a58567
c4b87d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
"""Predictor implementation wrapping the TimeSeriesModel for GIFT-Eval."""

import logging
from collections.abc import Iterator

import numpy as np
import torch
import yaml
from gluonts.model.forecast import QuantileForecast
from gluonts.model.predictor import Predictor
from torch.nn.parallel import DistributedDataParallel as DDP

from src.data.containers import BatchTimeSeriesContainer
from src.data.frequency import parse_frequency
from src.data.scalers import RobustScaler
from src.models.model import TimeSeriesModel
from src.utils.utils import device

logger = logging.getLogger(__name__)


class TimeSeriesPredictor(Predictor):
    """Unified predictor for TimeSeriesModel supporting flexible construction."""

    def __init__(
        self,
        model: TimeSeriesModel,
        config: dict,
        ds_prediction_length: int,
        ds_freq: str,
        batch_size: int = 32,
        max_context_length: int | None = None,
        debug: bool = False,
    ) -> None:
        # Dataset-specific context (can be updated per dataset/term)
        self.ds_prediction_length = ds_prediction_length
        self.ds_freq = ds_freq
        self.batch_size = batch_size
        self.max_context_length = max_context_length
        self.debug = debug

        # Persistent model/config (unwrap DDP if needed)
        self.model = model.module if isinstance(model, DDP) else model
        self.model.eval()
        self.config = config

        # Initialize scaler (using same type as model)
        scaler_type = self.config.get("TimeSeriesModel", {}).get("scaler", "custom_robust")
        epsilon = self.config.get("TimeSeriesModel", {}).get("epsilon", 1e-3)
        if scaler_type == "custom_robust":
            self.scaler = RobustScaler(epsilon=epsilon)
        else:
            raise ValueError(f"Unsupported scaler type: {scaler_type}")

    def set_dataset_context(
        self,
        prediction_length: int | None = None,
        freq: str | None = None,
        batch_size: int | None = None,
        max_context_length: int | None = None,
    ) -> None:
        """Update lightweight dataset-specific attributes without reloading the model."""

        if prediction_length is not None:
            self.ds_prediction_length = prediction_length
        if freq is not None:
            self.ds_freq = freq
        if batch_size is not None:
            self.batch_size = batch_size
        if max_context_length is not None:
            self.max_context_length = max_context_length

    @classmethod
    def from_model(
        cls,
        model: TimeSeriesModel,
        config: dict,
        ds_prediction_length: int,
        ds_freq: str,
        batch_size: int = 32,
        max_context_length: int | None = None,
        debug: bool = False,
    ) -> "TimeSeriesPredictor":
        return cls(
            model=model,
            config=config,
            ds_prediction_length=ds_prediction_length,
            ds_freq=ds_freq,
            batch_size=batch_size,
            max_context_length=max_context_length,
            debug=debug,
        )

    @classmethod
    def from_paths(
        cls,
        model_path: str,
        config_path: str,
        ds_prediction_length: int,
        ds_freq: str,
        batch_size: int = 32,
        max_context_length: int | None = None,
        debug: bool = False,
    ) -> "TimeSeriesPredictor":
        with open(config_path) as f:
            config = yaml.safe_load(f)
        model = cls._load_model_from_path(config=config, model_path=model_path)
        return cls(
            model=model,
            config=config,
            ds_prediction_length=ds_prediction_length,
            ds_freq=ds_freq,
            batch_size=batch_size,
            max_context_length=max_context_length,
            debug=debug,
        )

    @staticmethod
    def _load_model_from_path(config: dict, model_path: str) -> TimeSeriesModel:
        try:
            model = TimeSeriesModel(**config["TimeSeriesModel"]).to(device)
            checkpoint = torch.load(model_path, map_location=device)
            model.load_state_dict(checkpoint["model_state_dict"])
            model.eval()
            logger.info(f"Successfully loaded model from {model_path}")
            return model
        except Exception as exc:  # pragma: no cover - logging path
            logger.error(f"Failed to load model from {model_path}: {exc}")
            raise

    def predict(self, test_data_input) -> Iterator[QuantileForecast]:
        """Generate forecasts for the test data."""

        if hasattr(test_data_input, "__iter__") and not isinstance(test_data_input, list):
            test_data_input = list(test_data_input)
        logger.debug(f"Processing {len(test_data_input)} time series")

        # Group series by their effective length (after optional truncation),
        # then process each uniform-length group in sub-batches up to batch_size.
        def _effective_length(entry) -> int:
            target = entry["target"]
            if target.ndim == 1:
                seq_len = len(target)
            else:
                # target shape is [num_channels, seq_len]
                seq_len = target.shape[1]
            if self.max_context_length is not None:
                seq_len = min(seq_len, self.max_context_length)
            return seq_len

        length_to_items: dict[int, list[tuple[int, object]]] = {}
        for idx, entry in enumerate(test_data_input):
            seq_len = _effective_length(entry)
            length_to_items.setdefault(seq_len, []).append((idx, entry))

        total = len(test_data_input)
        ordered_results: list[QuantileForecast | None] = [None] * total

        for _, items in length_to_items.items():
            for i in range(0, len(items), self.batch_size):
                chunk = items[i : i + self.batch_size]
                entries = [entry for (_orig_idx, entry) in chunk]
                batch_forecasts = self._predict_batch(entries)
                for forecast_idx, (orig_idx, _entry) in enumerate(chunk):
                    ordered_results[orig_idx] = batch_forecasts[forecast_idx]

        return ordered_results  # type: ignore[return-value]

    def _predict_batch(self, test_data_batch: list) -> list[QuantileForecast]:
        """Generate predictions for a batch of time series."""

        logger.debug(f"Processing batch of size: {len(test_data_batch)}")

        try:
            batch_container = self._convert_to_batch_container(test_data_batch)

            if isinstance(device, torch.device):
                device_type = device.type
            else:
                device_type = "cuda" if "cuda" in str(device).lower() else "cpu"
            enable_autocast = device_type == "cuda"

            with torch.autocast(
                device_type=device_type,
                dtype=torch.bfloat16,
                enabled=enable_autocast,
            ):
                with torch.no_grad():
                    model_output = self.model(batch_container, drop_enc_allow=False)

            forecasts = self._convert_to_forecasts(model_output, test_data_batch, batch_container)

            logger.debug(f"Generated {len(forecasts)} forecasts")
            return forecasts
        except Exception as exc:  # pragma: no cover - logging path
            logger.error(f"Error in batch prediction: {exc}")
            raise

    def _convert_to_batch_container(self, test_data_batch: list) -> BatchTimeSeriesContainer:
        """Convert gluonts test data to BatchTimeSeriesContainer."""

        batch_size = len(test_data_batch)
        history_values_list = []
        start_dates = []
        frequencies = []

        for entry in test_data_batch:
            target = entry["target"]

            if target.ndim == 1:
                target = target.reshape(-1, 1)
            else:
                target = target.T

            if self.max_context_length is not None and len(target) > self.max_context_length:
                target = target[-self.max_context_length :]

            history_values_list.append(target)
            start_dates.append(entry["start"].to_timestamp().to_datetime64())
            frequencies.append(parse_frequency(entry["freq"]))

        history_values_np = np.stack(history_values_list, axis=0)
        num_channels = history_values_np.shape[2]

        history_values = torch.tensor(history_values_np, dtype=torch.float32, device=device)

        future_values = torch.zeros(
            (batch_size, self.ds_prediction_length, num_channels),
            dtype=torch.float32,
            device=device,
        )

        return BatchTimeSeriesContainer(
            history_values=history_values,
            future_values=future_values,
            start=start_dates,
            frequency=frequencies,
        )

    def _convert_to_forecasts(
        self,
        model_output: dict,
        test_data_batch: list,
        batch_container: BatchTimeSeriesContainer,
    ) -> list[QuantileForecast]:
        """Convert model predictions to QuantileForecast objects."""

        predictions = model_output["result"]
        scale_statistics = model_output["scale_statistics"]

        if predictions.ndim == 4:
            predictions_unscaled = self.scaler.inverse_scale(predictions, scale_statistics)
            is_quantile = True
            quantile_levels = self.model.quantiles
        else:
            predictions_unscaled = self.scaler.inverse_scale(predictions, scale_statistics)
            is_quantile = False
            quantile_levels = [0.5]

        forecasts: list[QuantileForecast] = []
        for idx, entry in enumerate(test_data_batch):
            history_length = int(batch_container.history_values.shape[1])
            start_date = entry["start"]
            forecast_start = start_date + history_length

            if is_quantile:
                pred_array = predictions_unscaled[idx].cpu().numpy()

                if pred_array.shape[1] == 1:
                    pred_array = pred_array.squeeze(1)
                    forecast_arrays = pred_array.T
                else:
                    forecast_arrays = pred_array.transpose(2, 0, 1)

                forecast = QuantileForecast(
                    forecast_arrays=forecast_arrays,
                    forecast_keys=[str(q) for q in quantile_levels],
                    start_date=forecast_start,
                )
            else:
                pred_array = predictions_unscaled[idx].cpu().numpy()

                if pred_array.shape[1] == 1:
                    pred_array = pred_array.squeeze(1)
                    forecast_arrays = pred_array.reshape(1, -1)
                else:
                    forecast_arrays = pred_array.reshape(1, *pred_array.shape)

                forecast = QuantileForecast(
                    forecast_arrays=forecast_arrays,
                    forecast_keys=["0.5"],
                    start_date=forecast_start,
                )

            forecasts.append(forecast)

        return forecasts


__all__ = ["TimeSeriesPredictor"]