Add sample usage section

README.md

@@ -1,4 +1,7 @@
 ---
+base_model: Datadog/Toto-2.0-2.5B
+license: apache-2.0
+pipeline_tag: time-series-forecasting
 tags:
 - time-series-forecasting
 - foundation-models
@@ -10,28 +13,25 @@ tags:
 - gift-eval
 - safetensors
 - pytorch_model_hub_mixin
-license: apache-2.0
-pipeline_tag: time-series-forecasting
 thumbnail: https://corp.dd-static.net/img/about/presskit/kit/press_kit.png
-base_model: Datadog/Toto-2.0-2.5B
 model-index:
 - name: Toto-2.0-2.5B-FT
   results:
-    - task:
-        type: time-series-forecasting
-      dataset:
-        name: GIFT-Eval
-        type: GIFT-Eval
-      metrics:
-        - name: CRPS
-          type: CRPS
-          value: 0.463
-        - name: MASE
-          type: MASE
-          value: 0.679
-      source:
-        name: GIFT-Eval Time Series Forecasting Leaderboard
-        url: https://huggingface.co/spaces/Salesforce/GIFT-Eval
+  - task:
+      type: time-series-forecasting
+    dataset:
+      name: GIFT-Eval
+      type: GIFT-Eval
+    metrics:
+    - type: CRPS
+      value: 0.463
+      name: CRPS
+    - type: MASE
+      value: 0.679
+      name: MASE
+    source:
+      url: https://huggingface.co/spaces/Salesforce/GIFT-Eval
+      name: GIFT-Eval Time Series Forecasting Leaderboard
 ---
 
 # Toto-2.0-2.5B-FT
@@ -51,6 +51,38 @@ A single Toto 2.0 2.5B base checkpoint finetuned on a mix that **includes the GI
 <figcaption>On the full GIFT-Eval leaderboard (foundation models + finetuned + ensemble + agentic), Toto-2.0-2.5B-FT places <b>#2 on CRPS rank, MASE rank, and #3 on raw CRPS / MASE</b>, behind only the <a href="https://huggingface.co/Datadog/Toto-2.0-Family-and-Friends">Toto 2.0 Family-and-Friends</a> ensemble.</figcaption>
 </figure>
 
+## 💻 Sample usage
+
+First, install the library:
+```bash
+pip install "toto-2 @ git+https://github.com/DataDog/toto.git#subdirectory=toto2"
+```
+
+Then you can use the model for forecasting as follows:
+
+```python
+import torch
+from toto2 import Toto2Model
+
+model = Toto2Model.from_pretrained("Datadog/Toto-2.0-2.5B-FT")
+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,
+)
+```
+
 ## 🔁 Finetuning recipe
 
 Starting from a fully-decayed [Toto-2.0-2.5B](https://huggingface.co/Datadog/Toto-2.0-2.5B) base checkpoint, we finetuned for 10,000 steps on a mix designed to expose the model to in-distribution structure without overfitting to GIFT-Eval alone: