Polish dataset card

Improve readability of the dataset README (plain-text coordinates, viewer note, updated benchmark table, and richer YAML metadata).

README.md

@@ -1,5 +1,7 @@
 ---
 language: en
+pretty_name: "MUR SST ML Benchmark (Pacific, Zarr)"
+license: "other"
 tags:
   - climate
   - sea-surface-temperature
@@ -9,14 +11,14 @@ task_categories:
   - time-series-forecasting
 ---
 
-# MUR SST (Pacific) — ML Benchmark Zarr
+# MUR SST ML Benchmark (Pacific, Zarr)
 
-This repo contains a **machine-learning friendly Zarr subset** of the NASA/JPL GHRSST **MUR SST** product, packaged for easy download + ML access.
+Machine-learning friendly **Zarr subset** of NASA/JPL GHRSST **MUR SST**.
 
 - **Upstream source (public, no auth)**: `s3://mur-sst/zarr`
 - **Subset**:
-  - **Region**: Pacific, **20–50°N**, **180–240°E** (longitude stored as **0–360**)
-  - **Time**: **2018-01-01 → 2019-12-30** (729 daily frames; upstream coverage for this slice ends on 2019-12-30)
+  - **Region**: Pacific (20–50°N, 180–240°E); longitude is stored as **0–360°E**
+  - **Time**: 2018-01-01 → 2019-12-30 (729 daily frames; upstream coverage for this slice ends on 2019-12-30)
   - **Variable**: `analysed_sst` only (**float32, °C**)
 - **Chunking for ML**: `(time, lat, lon) = (7, 256, 256)` (weekly windows)
 
@@ -26,7 +28,7 @@ Why `mur-sst/zarr-v1` during extraction?
 
 ## Notes (Hub viewer)
 
-- The **Dataset Viewer is expected to be unavailable** for this repo because it contains a tar archive of a Zarr store (not a `datasets`-native format with named splits).
+- The **Dataset Viewer is expected to be unavailable** because this repo contains a tar archive of a Zarr store (not a `datasets`-native format with named splits).
 
 ## Files in this dataset repo
 
@@ -58,7 +60,7 @@ Time-contiguous splits (no leakage):
 - **Val**: 2018-12-31 → 2019-06-30
 - **Test**: 2019-07-01 → 2019-12-30
 
-## Usage example
+## Streaming code example
 
 Local:
 
@@ -68,7 +70,7 @@ ds = xr.open_zarr("pacific_sst.zarr", consolidated=True)
 print(ds)
 ```
 
-From Hugging Face (download + unpack):
+Remote (Hugging Face, after download):
 
 ```python
 import xarray as xr
@@ -85,7 +87,7 @@ Run:
 
 ```bash
 tar -xf pacific_sst.zarr.tar
-python bench/throughput_benchmark.py --local pacific_sst.zarr
+python bench/throughput_benchmark.py --local pacific_sst.zarr --s3-root mur-sst/zarr-v1
 ```
 
 Measured on this machine (see `bench/throughput_benchmark.py` for details):