Download README.md from Zipeng365/TSDecompose-Benchmark-Leaderboard: direct link, hf CLI and curl.
- Browser
- Download file 1.12 kB
-
https://huggingface.co/spaces/Zipeng365/TSDecompose-Benchmark-Leaderboard/resolve/main/README.md
- Command line
-
hf download hf://spaces/Zipeng365/TSDecompose-Benchmark-Leaderboard/README.md
-
curl -L -o README.md https://huggingface.co/spaces/Zipeng365/TSDecompose-Benchmark-Leaderboard/resolve/main/README.md
A newer version of the Gradio SDK is available: 6.30.0
title: TSDecompose Benchmark Leaderboard
emoji: 📈
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false
license: mit
datasets:
- Zipeng365/TSDecompose-Benchmark
TSDecompose Benchmark Leaderboard
Interactive leaderboard for the TSDecompose benchmark release.
The Space reads leaderboard tables from the dataset repository:
Zipeng365/TSDecompose-Benchmark
The dataset repository remains the source of record. This Space is only the web presentation layer.
The default view is paper-first: it anchors the web leaderboard to the camera-ready Table 2 / Figure 3 six-family summary from the 6-scenario, 50-draw core protocol. Table 2 uses tier-balanced aggregation over valid metric values: stationary columns average Tier 1 and Tier 2 means with equal weight, and non-stationary columns use Tier 3 means. The 22-method synthetic and semi-synthetic views are benchmark-only living extensions that use the released five-metric ranking files from the dataset repository; they are not replacements for the primary Table 2 view and are not camera-ready paper claims.