[TimesFM 2.5]:`window_size` argument raises `AttributeError`

[kanenorman]

System Info

• transformers version: 5.13.0.dev0
• Platform: macOS-26.5.1-arm64-arm-64bit
• Python version: 3.11.3
• Huggingface_hub version: 1.20.1
• Safetensors version: 0.8.0
• Accelerate version: 1.14.0
• Accelerate config: not found
• DeepSpeed version: not installed
• PyTorch version (accelerator?): 2.12.1 (NA)
• Using distributed or parallel set-up in script?: No

Who can help?

The model behaves correctly when window_size=None (default). Setting window_size to any value instead results in an AttributeError.

Happy to submit a fix.

@vasqu
@kashif

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

Example script based on the documentation example

import torch

from transformers import TimesFm2_5ModelForPrediction


model = TimesFm2_5ModelForPrediction.from_pretrained(
    "google/timesfm-2.5-200m-transformers",
    device_map="auto",
)

forecast_input_tensor = [
    torch.sin(torch.linspace(0, 20, 100, dtype=torch.float32, device=model.device)),
    torch.sin(torch.linspace(0, 20, 200, dtype=torch.float32, device=model.device)),
    torch.sin(torch.linspace(0, 20, 400, dtype=torch.float32, device=model.device)),
]

with torch.no_grad():
    outputs = model(past_values=forecast_input_tensor, window_size=5)

Error

Traceback (most recent call last):
  File "/transformers/bug.py", line 18, in <module>
    outputs = model(past_values=forecast_input_tensor, window_size=5)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/transformers/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1778, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/transformers/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1789, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/transformers/src/transformers/utils/generic.py", line 907, in wrapper
    output = func(self, *args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/transformers/src/transformers/models/timesfm2_5/modeling_timesfm2_5.py", line 780, in forward
    new_inputs.extend(self._timesfm_moving_average(ts, window_size))
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/transformers/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1967, in __getattr__
    raise AttributeError(
AttributeError: 'TimesFm2_5ModelForPrediction' object has no attribute '_timesfm_moving_average'. Did you mean: '_timesfm2_5_moving_average'?

Expected behavior

Expect return TimesFm2_5OutputForPrediction