Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,22 @@
+# zero2robot/policies
+
+ONNX policies that drive the [zero2robot](https://github.com/kaushikb11/zero2robot) course's live browser demos. Each file is the ONNX export of a from-scratch chapter policy; the site fetches these at build time (sha256-verified against `checkpoints/models.yaml`) and runs them in `onnxruntime-web`. No training required to read the course — these are the provisioned demo weights.
+
+| model | chapter | contract | sha256 |
+|---|---|---|---|
+| [`bc_policy.onnx`](bc_policy.onnx) | `ch1.1-bc` | `v1` | `dbfbb95a79cb…` |
+| [`dagger.onnx`](dagger.onnx) | `ch4.2-corrections` | `v1` | `e898ee4d2828…` |
+| [`ppo_agent.onnx`](ppo_agent.onnx) | `ch2.1-ppo` | `v1` | `abbf577b80ee…` |
+| [`diffusion_denoiser.onnx`](diffusion_denoiser.onnx) | `ch1.4-diffusion` | `v2` | `259cc0ec2f43…` |
+| [`flow_velocity.onnx`](flow_velocity.onnx) | `ch1.5-flow` | `v2` | `7443b65df656…` |
+| [`sac_actor.onnx`](sac_actor.onnx) | `ch2.2-sac` | `v1` | `3a6ebefe1a76…` |
+| [`walk_actor.onnx`](walk_actor.onnx) | `ch2.5-walk` | `v1` | `8dd6524fb76c…` |
+| [`shaped_walk.onnx`](shaped_walk.onnx) | `ch2.4-rewards` | `v1` | `2c3e5981f6eb…` |
+| [`height_hack.onnx`](height_hack.onnx) | `ch2.4-rewards` | `v1` | `442c3fe5e43e…` |
+| [`dr_narrow.onnx`](dr_narrow.onnx) | `ch2.7-dr` | `v1` | `ceb4cf4f2811…` |
+| [`dr_randomized.onnx`](dr_randomized.onnx) | `ch2.7-dr` | `v1` | `beda8abccae9…` |
+| [`serl_actor.onnx`](serl_actor.onnx) | `ch4.3-serl` | `v1` | `df33c8668ed0…` |
+| [`scratch_actor.onnx`](scratch_actor.onnx) | `ch4.3-serl` | `v1` | `e37483838052…` |
+| [`offline_policy.onnx`](offline_policy.onnx) | `ch4-offline-primer` | `v1` | `c4d347b61484…` |
+| [`offline_bc.onnx`](offline_bc.onnx) | `ch4-offline-primer` | `v1` | `31bf4b2d9af3…` |
+| [`act_policy.onnx`](act_policy.onnx) | `ch1.3-act` | `v2` | `f8c7168576f3…` |