TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Python
fix(timesfm3/mlx): accept a 2D ndarray of contexts in predict_batch
rachittshah55f2d8d746e2
diff --git a/src/timesfm3/mlx/timesfm3_forecaster.py b/src/timesfm3/mlx/timesfm3_forecaster.pyindex 8815014..98933bf 100644--- a/src/timesfm3/mlx/timesfm3_forecaster.py+++ b/src/timesfm3/mlx/timesfm3_forecaster.py@@ -175,7 +175,7 @@ class TimesFM3Forecaster:raise NotImplementedError(f"padding_mode={padding_mode!r} is not yet supported by the MLX backend.")- if not contexts:+ if len(contexts) == 0:returnn = len(contexts)diff --git a/src/timesfm3/mlx/timesfm3_forecaster_test.py b/src/timesfm3/mlx/timesfm3_forecaster_test.pyindex 9403d59..6ae8495 100644--- a/src/timesfm3/mlx/timesfm3_forecaster_test.py+++ b/src/timesfm3/mlx/timesfm3_forecaster_test.py@@ -156,6 +156,19 @@ class TimesFM3MlxRealWeightsTest(unittest.TestCase):self.assertEqual(np.asarray(out.forecast).shape, (64,))self.assertEqual(np.asarray(out.quantiles).shape, (64, 9))+ def test_predict_batch_accepts_2d_ndarray_contexts(self):+ # A 2D ndarray passed straight to predict_batch is a batch of univariate+ # rows. `if not contexts` raised "truth value ambiguous" on it; the guard+ # must be a length check instead (matches the torch backend / #483).+ forecaster = timesfm3_forecaster.TimesFM3Forecaster.from_pretrained(_CHECKPOINT)+ contexts = np.stack(+ [np.sin(np.linspace(0, 40, 512)), np.cos(np.linspace(0, 30, 512))]+ ).astype(np.float32) # (2, 512)+ outs = list(forecaster.predict_batch(contexts, horizon=32, return_quantiles=True))+ self.assertEqual(len(outs), 2)+ self.assertEqual(np.asarray(outs[0].forecast).shape, (32,))+ self.assertEqual(np.asarray(outs[1].quantiles).shape, (32, 9))+def test_global_context_default_matches_torch(self):# The MLX backend must expose the same 15,360-step context cap as the torch# backend so the two are backend-equivalent for very long contexts.