Update README.md

c5b7356 verified 4 days ago

4.06 kB

tags:
  - time-series-forecasting
  - foundation-models
  - pretrained-models
  - time-series
  - timeseries
  - forecasting
  - observability
  - safetensors
  - pytorch_model_hub_mixin
license: apache-2.0
pipeline_tag: time-series-forecasting
thumbnail: null
results:
  - task: time-series-forecasting
    dataset:
      name: GIFT-Eval
    metrics:
      - name: MASE
        type: mase
        value: 0.757
      - name: CRPS
        type: brier_score
        value: 0.524
    source:
      name: GIFT-Eval Time Series Forecasting Leaderboard
      url: https://huggingface.co/spaces/Salesforce/GIFT-Eval
  - task: time-series-forecasting
    dataset:
      name: BOOM
    metrics:
      - name: MASE
        type: mase
        value: 0.624
      - name: CRPS
        type: brier_score
        value: 0.717
    source:
      name: BOOM 💥 Observability Time-Series Forecasting Leaderboard
      url: https://huggingface.co/spaces/Datadog/BOOM

Toto-2.0-4m

Toto (Time Series Optimized Transformer for Observability) is a family of time series foundation models for multivariate forecasting developed by Datadog. Toto 2.0 is the current generation, featuring u-μP-scaled transformers ranging from 4M to 2.5B parameters.

✨ Key Features

Zero-Shot Forecasting: Forecast without fine-tuning on your specific time series.
Multi-Variate Support: Efficiently process multiple variables using alternating time/variate attention.
Probabilistic Predictions: Generate point forecasts and uncertainty estimates via a quantile output head.
Decoder-Only Architecture: Support for variable prediction horizons and context lengths.
u-μP Scaling: Stable training transfer across all model sizes.

Overview of the Toto 2.0 architecture.

⚡ Quick Start

Inference code is available on GitHub.

Installation

pip install "toto-2 @ git+https://github.com/DataDog/toto.git#subdirectory=toto2"

Inference Example

import torch
from toto2 import Toto2Model

model = Toto2Model.from_pretrained("Datadog/Toto-2.0-22m")
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = model.to(device).eval()

# (batch, n_variates, time_steps)
target = torch.randn(1, 1, 512, device=device)
target_mask = torch.ones_like(target, dtype=torch.bool)
series_ids = torch.zeros(1, 1, dtype=torch.long, device=device)

# Returns quantiles of shape (9, batch, n_variates, horizon)
# Quantile levels: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
quantiles = model.forecast(
    {"target": target, "target_mask": target_mask, "series_ids": series_ids},
    horizon=96,
    decode_block_size=768,
    has_missing_values=False,
)

For more examples, see the Quick Start notebook and GluonTS integration notebook.

💾 Available Checkpoints

Checkpoint	Parameters
Toto-2.0-4m	4M
Toto-2.0-22m	22M
Toto-2.0-313m	313M
Toto-2.0-1B	1B
Toto-2.0-2.5B	2.5B

🔗 Additional Resources

📖 Citation

(citation coming soon)