Refreshed model with speed optimizations.

README.md

@@ -1,71 +1,9 @@
 ---
-license: apache-2.0
-library_name: timesfm
-pipeline_tag: time-series-forecasting
+tags:
+- model_hub_mixin
 ---
 
-# TimesFM
-
-TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
-
-**Resources and Technical Documentation**:
-* Paper: [A decoder-only foundation model for time-series forecasting](https://arxiv.org/abs/2310.10688), ICML 2024.
-* [Google Research blog](https://research.google/blog/a-decoder-only-foundation-model-for-time-series-forecasting/)
-* [GitHub repo](https://github.com/google-research/timesfm)
-
-**Authors**: Google Research
-
-This checkpoint is not an officially supported Google product. See [TimesFM in BigQuery](https://cloud.google.com/bigquery/docs/timesfm-model) for Google official support.
-
-## Checkpoint `timesfm-2.5-200m`
-
-`timesfm-2.5-200m` is the third open model checkpoint.
-
-
-### Data
-
-`timesfm-2.5-200m` is pretrained using
-
-- [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain)
-- [Wikimedia Pageviews](https://meta.wikimedia.org/wiki/Pageviews_Analysis), cutoff Nov 2023 (see [paper](https://arxiv.org/abs/2310.10688) for details).
-- [Google Trends](https://trends.google.com/trends/) top queries, cutoff EoY 2022 (see [paper](https://arxiv.org/abs/2310.10688) for details).
-- Synthetic and augmented data.
-
-### Install
-
-`pip install` from PyPI coming soon. At this point, please run
-
-```shell
-git clone https://github.com/google-research/timesfm.git
-cd timesfm
-pip install -e .
-```
-
-### Code Example
-
-```python
-import numpy as np
-import timesfm
-model = timesfm.TimesFM_2p5_200M_torch.from_pretrained("google/timesfm-2.5-200m-pytorch")
-
-model.compile(
-    timesfm.ForecastConfig(
-        max_context=1024,
-        max_horizon=256,
-        normalize_inputs=True,
-        use_continuous_quantile_head=True,
-        force_flip_invariance=True,
-        infer_is_positive=True,
-        fix_quantile_crossing=True,
-    )
-)
-point_forecast, quantile_forecast = model.forecast(
-    horizon=12,
-    inputs=[
-        np.linspace(0, 1, 100),
-        np.sin(np.linspace(0, 20, 67)),
-    ],  # Two dummy inputs
-)
-point_forecast.shape  # (2, 12)
-quantile_forecast.shape  # (2, 12, 10): mean, then 10th to 90th quantiles.
-```
+This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
+- Code: [More Information Needed]
+- Paper: [More Information Needed]
+- Docs: [More Information Needed]

config.json

@@ -6,12 +6,12 @@
   "head_dim": 80,
   "hidden_size": 1280,
   "horizon_length": 128,
-  "quantile_horizon_length": 1024,
   "intermediate_size": 1280,
   "model_type": "timesfm",
   "num_attention_heads": 16,
   "num_hidden_layers": 20,
   "patch_length": 32,
+  "quantile_horizon_length": 1024,
   "quantiles": [
     0.1,
     0.2,
@@ -23,5 +23,6 @@
     0.8,
     0.9
   ],
-  "rms_norm_eps": 1e-06
+  "rms_norm_eps": 1e-06,
+  "torch_compile": false
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0568758b5e6c310383b96e83c6582ebac80ed3256ab0bdcf2212c118c0e274ba
-size 925185264
+oid sha256:2f776efe6245e42b24bc4153ffdf61810140210e4bd3b01fb21f7aa779ab6ce8
+size 925181104