Add README

README.md

@@ -0,0 +1,87 @@
+---
+library_name: skrl
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- skrl
+model-index:
+- name: PPO
+  results:
+  - metrics:
+    - type: mean_reward
+      value: 41.69 +/- 0.06
+      name: Total reward (mean)
+    task:
+      type: reinforcement-learning
+      name: reinforcement-learning
+    dataset:
+      name: Isaac-Lift-Franka-v0
+      type: Isaac-Lift-Franka-v0
+---
+
+<!-- ---
+torch: 41.69 +/- 0.06
+jax: 42.7 +/- 0.0
+numpy: 
+--- -->
+
+# IsaacOrbit-Isaac-Lift-Franka-v0-PPO
+
+Trained agent for [NVIDIA Isaac Orbit](https://github.com/NVIDIA-Omniverse/Orbit) environments.
+
+- **Task:** Isaac-Lift-Franka-v0
+- **Agent:** [PPO](https://skrl.readthedocs.io/en/latest/api/agents/ppo.html)
+
+# Usage (with skrl)
+
+Note: Visit the skrl [Examples](https://skrl.readthedocs.io/en/latest/intro/examples.html) section to access the scripts.
+
+* PyTorch
+
+    ```python
+    from skrl.utils.huggingface import download_model_from_huggingface
+
+    # assuming that there is an agent named `agent`
+    path = download_model_from_huggingface("skrl/IsaacOrbit-Isaac-Lift-Franka-v0-PPO", filename="agent.pt")
+    agent.load(path)
+    ```
+
+* JAX
+
+    ```python
+    from skrl.utils.huggingface import download_model_from_huggingface
+
+    # assuming that there is an agent named `agent`
+    path = download_model_from_huggingface("skrl/IsaacOrbit-Isaac-Lift-Franka-v0-PPO", filename="agent.pickle")
+    agent.load(path)
+    ```
+
+# Hyperparameters
+
+```python
+# https://skrl.readthedocs.io/en/latest/api/agents/ppo.html#configuration-and-hyperparameters
+cfg = PPO_DEFAULT_CONFIG.copy()
+cfg["rollouts"] = 96  # memory_size
+cfg["learning_epochs"] = 5
+cfg["mini_batches"] = 4  # 96 * 4096 / 98304
+cfg["discount_factor"] = 0.99
+cfg["lambda"] = 0.95
+cfg["learning_rate"] = 1e-3
+cfg["learning_rate_scheduler"] = KLAdaptiveLR
+cfg["learning_rate_scheduler_kwargs"] = {"kl_threshold": 0.01, "min_lr": 1e-5}
+cfg["random_timesteps"] = 0
+cfg["learning_starts"] = 0
+cfg["grad_norm_clip"] = 1.0
+cfg["ratio_clip"] = 0.2
+cfg["value_clip"] = 0.2
+cfg["clip_predicted_values"] = True
+cfg["entropy_loss_scale"] = 0.01
+cfg["value_loss_scale"] = 1.0
+cfg["kl_threshold"] = 0
+cfg["rewards_shaper"] = None
+cfg["time_limit_bootstrap"] = True
+cfg["state_preprocessor"] = RunningStandardScaler
+cfg["state_preprocessor_kwargs"] = {"size": env.observation_space, "device": device}
+cfg["value_preprocessor"] = RunningStandardScaler
+cfg["value_preprocessor_kwargs"] = {"size": 1, "device": device}
+```