euler314
/

typhoon-predict

@@ -32,6 +32,8 @@ This is a **research release**, not an operational warning service. Follow offic
 [Watch / download MP4](https://huggingface.co/euler314/typhoon-predict/resolve/87a6e366b42bb4cc0edc95d2c50c55fca21a2c93/docs/trackformer_1_2_mangkhut.mp4) · **11 September 2018, 00 UTC** · **50-member mean** · **20 seconds**
 The animation shows twenty actual six-hour forecasts through +120 h. The Western Pacific overview and close-up use the same geographic mean of the model's basin and moving-core pressure fields. Blue denotes low pressure and red high pressure, with 2 hPa isobars and 4 hPa labels. Forecast and observed routes share the same coordinates and valid times—neither is shifted to make them overlap.
 This selected example illustrates the model, not typical skill. Observations are verification only. The 50 members are distinct seeded perturbations of **past and issue-time inputs**, not separately trained networks. Member fields are registered geographically before averaging; the mean centre need not locate the minimum of the mean map. Display interpolation does not add resolution to the approximately 20 km core.
@@ -42,7 +44,7 @@ This selected example illustrates the model, not typical skill. Observations are
 Trackformer 1.1 predicts track and scalar intensity/structure outputs. **1.2 adds evolving sea-level-pressure fields and a moving pressure core**, giving the route forecast a spatial weather representation that can be inspected on a map.
-![Trackformer 1.1 versus 1.2: pressure intensity, track position and track direction on matched daily forecasts](https://huggingface.co/euler314/typhoon-predict/resolve/1e1dffa6efcc7298a4e7104a279d36cd7ea0b9eb/evaluation/released_daily/model_1_2_benchmark.png)
 | Development metric · lower is better | 1.1 | 1.2 · mean of 50 | Shared coverage |
 | --- | ---: | ---: | --- |
@@ -62,7 +64,7 @@ The cohort contains 40 recent storms and 230 storms from 1980–1999. It lies ou
 ## How Trackformer 1.2 works
-![Trackformer 1.2: causal weather history, coupled basin and moving core, pressure-based readout and autoregressive rollout](https://huggingface.co/euler314/typhoon-predict/resolve/1e1dffa6efcc7298a4e7104a279d36cd7ea0b9eb/docs/trackformer_1_2_architecture.svg)
 1. **Initialize from the past.** Nine six-hour analyses provide MSLP, 500 hPa height and winds at 850/500/200 hPa, ending at issue time. Static geography, the current observed centre, recent motion, current intensity and validity masks accompany the history. Native regional pressure detail is used when available; missing detail stays flagged. Current observations initialize the core, not future labels.
 2. **Evolve the environment.** A convolutional recurrent network combines steering-based advection with learned pressure tendencies. Multiscale spatial tokens and attention condition the environmental memory.

 [Watch / download MP4](https://huggingface.co/euler314/typhoon-predict/resolve/87a6e366b42bb4cc0edc95d2c50c55fca21a2c93/docs/trackformer_1_2_mangkhut.mp4) · **11 September 2018, 00 UTC** · **50-member mean** · **20 seconds**
+GitHub plays the lightweight animated GIF inline; click it for the full-quality MP4. Hugging Face provides a native video player. The GIF preserves all twenty forecast states, one second each, and loops without adding a frozen ending.
 The animation shows twenty actual six-hour forecasts through +120 h. The Western Pacific overview and close-up use the same geographic mean of the model's basin and moving-core pressure fields. Blue denotes low pressure and red high pressure, with 2 hPa isobars and 4 hPa labels. Forecast and observed routes share the same coordinates and valid times—neither is shifted to make them overlap.
 This selected example illustrates the model, not typical skill. Observations are verification only. The 50 members are distinct seeded perturbations of **past and issue-time inputs**, not separately trained networks. Member fields are registered geographically before averaging; the mean centre need not locate the minimum of the mean map. Display interpolation does not add resolution to the approximately 20 km core.
 Trackformer 1.1 predicts track and scalar intensity/structure outputs. **1.2 adds evolving sea-level-pressure fields and a moving pressure core**, giving the route forecast a spatial weather representation that can be inspected on a map.
+![Trackformer 1.1 versus 1.2: pressure intensity, track position and track direction on matched daily forecasts](https://huggingface.co/euler314/typhoon-predict/resolve/e0d6ce5d5de3e9ace333961745facd141341128a/evaluation/released_daily/model_1_2_benchmark.png)
 | Development metric · lower is better | 1.1 | 1.2 · mean of 50 | Shared coverage |
 | --- | ---: | ---: | --- |
 ## How Trackformer 1.2 works
+![Trackformer 1.2: causal weather history, coupled basin and moving core, pressure-based readout and autoregressive rollout](https://huggingface.co/euler314/typhoon-predict/resolve/e0d6ce5d5de3e9ace333961745facd141341128a/docs/trackformer_1_2_architecture.svg)
 1. **Initialize from the past.** Nine six-hour analyses provide MSLP, 500 hPa height and winds at 850/500/200 hPa, ending at issue time. Static geography, the current observed centre, recent motion, current intensity and validity masks accompany the history. Native regional pressure detail is used when available; missing detail stays flagged. Current observations initialize the core, not future labels.
 2. **Evolve the environment.** A convolutional recurrent network combines steering-based advection with learned pressure tendencies. Multiscale spatial tokens and attention condition the environmental memory.