Rajat Sen 2f1625c208 Fix model loading issues and forecast_naive slicing bug in TimesFM 2.5 ...

- Allow model wrapper constructors (__init__) to accept and ignore extra keyword arguments (e.g. proxies) passed by huggingface_hub during from_pretrained.
- Implement load_checkpoint for TimesFM_2p5_200M_torch and TimesFM_2p5_200M_flax to restore weights from local paths.
- Fix slicing bug in PyTorch's forecast_naive to correctly slice the time/horizon dimension ([:, :horizon, :]) instead of quantiles.
- Add unit tests in tests/test_model_loading.py covering local checkpoint loading, hub compatibility, and prediction shape correctness.

2026-06-07 17:28:32 +00:00

..

__init__.py

test: add unit tests for configs, torch layers, utils, and base utils

2026-04-08 14:13:37 -04:00

test_base_utils.py

test: add unit tests for configs, torch layers, utils, and base utils

2026-04-08 14:13:37 -04:00

test_configs.py

test: add unit tests for configs, torch layers, utils, and base utils

2026-04-08 14:13:37 -04:00

test_model_loading.py

Fix model loading issues and forecast_naive slicing bug in TimesFM 2.5

2026-06-07 17:28:32 +00:00

test_torch_layers.py

test: add unit tests for configs, torch layers, utils, and base utils

2026-04-08 14:13:37 -04:00

test_torch_utils.py

test: add unit tests for configs, torch layers, utils, and base utils

2026-04-08 14:13:37 -04:00