search

google-research / timesfm

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
https://research.google/blog/a-decoder-only-foundation-model-for-time-series-forecasting/
Apache License 2.0
3.02k stars 228 forks source link

Please publish train_state_unpadded_shape_dtype_struct #41

Open sdmorrey opened 1 month ago

sdmorrey commented 1 month ago

We really need this file very soon the model will not run at all

Restored checkpoint in 6.29 seconds. Jitting decoding. ERROR:absl:For checkpoint version > 1.0, we require users to provide train_state_unpadded_shape_dtype_structduring checkpoint saving/restoring, to avoid potential silent bugs when loading checkpoints to incompatible unpadded shapes of TrainState.

rajatsen91 commented 1 month ago

I just checked in a fresh install. Model loading and inference still works despite of that error message. Please verify whether you can run inference. If so, for now please treat this ERROR:absl as a warning not an error.

sf19881108 commented 1 month ago

I have downloaded the model locally and encountered an error while running the code： Constructed model weights in 1.72 seconds. Restoring checkpoint from model/timesfm-1.0-200m/checkpoints/. WARNING:absl:No registered CheckpointArgs found for handler type: <class 'paxml.checkpoints.FlaxCheckpointHandler'> WARNING:absl:Configured CheckpointManager using deprecated legacy API. Please follow the instructions at https://orbax.readthedocs.io/en/latest/api_refactor.html to migrate by May 1st, 2024. WARNING:absl:train_state_unpadded_shape_dtype_struct is not provided. We assume train_state is unpadded. ERROR:absl:For checkpoint version > 1.0, we require users to provide train_state_unpadded_shape_dtype_struct during checkpoint saving/restoring, to avoid potential silent bugs when loading checkpoints to incompatible unpadded shapes of TrainState. Restored checkpoint in 0.95 seconds. Jitting decoding. Does anyone knows how to deal it?