diff --git a/.github/workflows/eval-model.yml b/.github/workflows/eval-model.yml index 51f611cd..28b4b2b6 100644 --- a/.github/workflows/eval-model.yml +++ b/.github/workflows/eval-model.yml @@ -36,13 +36,13 @@ jobs: run: pip install ".[dev]" -f https://download.pytorch.org/whl/cpu/torch_stable.html - name: Run Eval Script for Chronos-2 - run: python scripts/evaluation/evaluate.py chronos-2 ci/evaluate/backtest_config.yaml $CHRONOS_2_RESULTS_CSV --model-id=s3://autogluon/chronos-2 --device=cpu --torch-dtype=float32 + run: python scripts/evaluation/evaluate.py chronos-2 ci/evaluate/backtest_config.yaml $CHRONOS_2_RESULTS_CSV --model-id=amazon/chronos-2 --device=cpu --torch-dtype=float32 - name: Print Chronos-2 CSV run: cat $CHRONOS_2_RESULTS_CSV - name: Run Eval Script for Chronos-Bolt run: python scripts/evaluation/evaluate.py chronos-bolt ci/evaluate/backtest_config.yaml $CHRONOS_BOLT_RESULTS_CSV --model-id=amazon/chronos-bolt-small --device=cpu --torch-dtype=float32 - + - name: Print Chronos-Bolt CSV run: cat $CHRONOS_BOLT_RESULTS_CSV diff --git a/README.md b/README.md index 61a1615a..c5138f05 100644 --- a/README.md +++ b/README.md @@ -19,7 +19,7 @@ ## 🚀 News -- **20 Oct 2025**: 🚀 [Chronos-2](https://arxiv.org/abs/2510.15821) released. It offers _zero-shot_ support for univariate, multivariate, and covariate-informed forecasting tasks. Chronos-2 achieves the best performance on fev-bench, GIFT-Eval and Chronos Benchmark II amongst pretrained models. Check out [this notebook](notebooks/chronos-2-quickstart.ipynb) to get started with Chronos-2. +- **20 Oct 2025**: 🚀 [Chronos-2](https://huggingface.co/amazon/chronos-2) released. It offers _zero-shot_ support for univariate, multivariate, and covariate-informed forecasting tasks. Chronos-2 achieves the best performance on fev-bench, GIFT-Eval and Chronos Benchmark II amongst pretrained models. Check out [this notebook](notebooks/chronos-2-quickstart.ipynb) to get started with Chronos-2. - **14 Feb 2025**: 🚀 Chronos-Bolt is now available on Amazon SageMaker JumpStart! Check out the [tutorial notebook](notebooks/deploy-chronos-bolt-to-amazon-sagemaker.ipynb) to learn how to deploy Chronos endpoints for production use in 3 lines of code. - **12 Dec 2024**: 📊 We released [`fev`](https://github.com/autogluon/fev), a lightweight package for benchmarking time series forecasting models based on the [Hugging Face `datasets`](https://huggingface.co/docs/datasets/en/index) library. - **26 Nov 2024**: ⚡️ Chronos-Bolt models released [on HuggingFace](https://huggingface.co/collections/amazon/chronos-models-65f1791d630a8d57cb718444). Chronos-Bolt models are more accurate (5% lower error), up to 250x faster and 20x more memory efficient than the original Chronos models of the same size! @@ -39,7 +39,7 @@ This package provides an interface to the Chronos family of **pretrained time se | Model ID | Parameters | | ---------------------------------------------------------------------- | ---------- | -| [`s3://autogluon/chronos-2`](https://arxiv.org/abs/2510.15821) | 120M | +| [`amazon/chronos-2`](https://huggingface.co/amazon/chronos-2) | 120M | | [`amazon/chronos-bolt-tiny`](https://huggingface.co/amazon/chronos-bolt-tiny) | 9M | | [`amazon/chronos-bolt-mini`](https://huggingface.co/amazon/chronos-bolt-mini) | 21M | | [`amazon/chronos-bolt-small`](https://huggingface.co/amazon/chronos-bolt-small) | 48M | @@ -48,7 +48,7 @@ This package provides an interface to the Chronos family of **pretrained time se | [`amazon/chronos-t5-mini`](https://huggingface.co/amazon/chronos-t5-mini) | 20M | | [`amazon/chronos-t5-small`](https://huggingface.co/amazon/chronos-t5-small) | 46M | | [`amazon/chronos-t5-base`](https://huggingface.co/amazon/chronos-t5-base) | 200M | -| [`amazon/chronos-t5-large`](https://huggingface.co/amazon/chronos-t5-large) | 710M | +| [`amazon/chronos-t5-large`](https://huggingface.co/amazon/chronos-t5-large) | 710M | @@ -68,7 +68,7 @@ A minimal example showing how to perform forecasting using Chronos-2: import pandas as pd # requires: pip install 'pandas[pyarrow]' from chronos import Chronos2Pipeline -pipeline = Chronos2Pipeline.from_pretrained("s3://autogluon/chronos-2", device_map="cuda") +pipeline = Chronos2Pipeline.from_pretrained("amazon/chronos-2", device_map="cuda") # Load historical target values and past values of covariates context_df = pd.read_parquet("https://autogluon.s3.amazonaws.com/datasets/timeseries/electricity_price/train.parquet") diff --git a/notebooks/chronos-2-quickstart.ipynb b/notebooks/chronos-2-quickstart.ipynb index e17a9951..32a6111c 100644 --- a/notebooks/chronos-2-quickstart.ipynb +++ b/notebooks/chronos-2-quickstart.ipynb @@ -36,7 +36,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "id": "fcc7e496", "metadata": {}, "outputs": [], @@ -53,7 +53,7 @@ "\n", "# Load the Chronos-2 pipeline\n", "# GPU recommended for faster inference, but CPU is also supported\n", - "pipeline: Chronos2Pipeline = BaseChronosPipeline.from_pretrained(\"s3://autogluon/chronos-2/\", device_map=\"cuda\")" + "pipeline: Chronos2Pipeline = BaseChronosPipeline.from_pretrained(\"amazon/chronos-2\", device_map=\"cuda\")" ] }, { diff --git a/test/test_chronos2.py b/test/test_chronos2.py index 3c4281cb..0efde91d 100644 --- a/test/test_chronos2.py +++ b/test/test_chronos2.py @@ -13,10 +13,9 @@ import torch from chronos import BaseChronosPipeline, Chronos2Pipeline -from chronos.chronos2.dataset import convert_df_input_to_list_of_dicts_input from chronos.chronos2.config import Chronos2CoreConfig +from chronos.chronos2.dataset import convert_df_input_to_list_of_dicts_input from chronos.chronos2.layers import MHA - from test.util import validate_tensor DUMMY_MODEL_PATH = Path(__file__).parent / "dummy-chronos2-model" @@ -35,6 +34,10 @@ def test_base_chronos2_pipeline_loads_from_s3(): BaseChronosPipeline.from_pretrained("s3://autogluon/chronos-2", device_map="cpu") +def test_base_chronos2_pipeline_loads_from_hf(): + BaseChronosPipeline.from_pretrained("amazon/chronos-2", device_map="cpu") + + @pytest.mark.parametrize( "inputs, prediction_length, expected_output_shapes", [