mistralai
/

Timestral-28M

Time Series Forecasting

foundation-model

Eval Results (legacy)

Model card Files Files and versions

Update README.md

#3

by richardkurle - opened 15 days ago

base: refs/heads/main

←

from: refs/pr/3

Discussion Files changed

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -73,7 +73,7 @@ This repository contains the 28M checkpoint.
 ## Model Details
 - **Size:** 28M parameters.
-- **Architecture:** patch-based transformer with time attention and variate attention, 12 layers, embedding dimension 256.
 - **Patching:** 32 time steps per patch. Each patch is embedded from its values together with their missing-value indicators.
 - **Context length:** up to 32,768 time steps (1,024 patches).
 - **Forecast horizon**: Up to 1,024 time steps (32 patches) in a single forward pass. Longer horizons are forecast block by block, with each block's median forecast appended to the context.

 ## Model Details
 - **Size:** 28M parameters.
+- **Architecture:** patch-based transformer with time attention and variate attention, 12 layers, embedding dimension 384.
 - **Patching:** 32 time steps per patch. Each patch is embedded from its values together with their missing-value indicators.
 - **Context length:** up to 32,768 time steps (1,024 patches).
 - **Forecast horizon**: Up to 1,024 time steps (32 patches) in a single forward pass. Longer horizons are forecast block by block, with each block's median forecast appended to the context.