initial upload

README.md

@@ -0,0 +1,21 @@
+---
+library_name: lerobot
+license: cc-by-4.0
+pipeline_tag: robotics
+tags:
+    - robotics
+    - imitation-learning
+    - active-vision
+    - humanoid
+    - pi0
+    - tavis
+datasets:
+    - tavis-benchmark/tavis-hands-gr1t2
+base_model: lerobot/pi0_base
+---
+
+π₀ multi-task checkpoint, fine-tuned on the TAVIS-Hands suite for the GR1T2 humanoid with head + bimanual wrist cameras.
+
+See the [tavis-benchmark org page](https://huggingface.co/tavis-benchmark) for benchmark description, full eval results, and usage instructions.
+
+

config.json

@@ -0,0 +1,95 @@
+{
+    "type": "pi0",
+    "n_obs_steps": 1,
+    "input_features": {
+        "observation.images.OBS_HEAD": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                480,
+                640
+            ]
+        },
+        "observation.images.OBS_WRIST_LEFT": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                480,
+                640
+            ]
+        },
+        "observation.images.OBS_WRIST_RIGHT": {
+            "type": "VISUAL",
+            "shape": [
+                3,
+                480,
+                640
+            ]
+        },
+        "observation.state": {
+            "type": "STATE",
+            "shape": [
+                44
+            ]
+        }
+    },
+    "output_features": {
+        "action": {
+            "type": "ACTION",
+            "shape": [
+                19
+            ]
+        }
+    },
+    "device": "cuda",
+    "use_amp": false,
+    "use_peft": false,
+    "push_to_hub": true,
+    "repo_id": null,
+    "private": null,
+    "tags": null,
+    "license": null,
+    "pretrained_path": null,
+    "paligemma_variant": "gemma_2b",
+    "action_expert_variant": "gemma_300m",
+    "dtype": "float32",
+    "chunk_size": 16,
+    "n_action_steps": 8,
+    "max_state_dim": 44,
+    "max_action_dim": 32,
+    "num_inference_steps": 10,
+    "time_sampling_beta_alpha": 1.5,
+    "time_sampling_beta_beta": 1.0,
+    "time_sampling_scale": 0.999,
+    "time_sampling_offset": 0.001,
+    "min_period": 0.004,
+    "max_period": 4.0,
+    "rtc_config": null,
+    "image_resolution": [
+        224,
+        224
+    ],
+    "empty_cameras": 0,
+    "normalization_mapping": {
+        "VISUAL": "IDENTITY",
+        "STATE": "MEAN_STD",
+        "ACTION": "MEAN_STD"
+    },
+    "gradient_checkpointing": true,
+    "compile_model": false,
+    "compile_mode": "max-autotune",
+    "freeze_vision_encoder": false,
+    "train_expert_only": false,
+    "optimizer_lr": 2.5e-05,
+    "optimizer_betas": [
+        0.9,
+        0.95
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 0.01,
+    "optimizer_grad_clip_norm": 1.0,
+    "scheduler_warmup_steps": 1000,
+    "scheduler_decay_steps": 30000,
+    "scheduler_decay_lr": 2.5e-06,
+    "tokenizer_max_length": 48
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bca3cedabc49c8f944c7a73e1baa98d721733d6a20fedf034706866d5989ffbe
+size 14005672408

policy_postprocessor.json

@@ -0,0 +1,32 @@
+{
+  "name": "policy_postprocessor",
+  "steps": [
+    {
+      "registry_name": "unnormalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              19
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "MEAN_STD",
+          "ACTION": "MEAN_STD"
+        }
+      },
+      "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cpu",
+        "float_dtype": null
+      }
+    }
+  ]
+}

policy_postprocessor_step_0_unnormalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89167698c8d5dcec1ceec420c9aa97552e732a27b5e2e095e8c112cbdf99ca4c
+size 15772

policy_preprocessor.json

@@ -0,0 +1,87 @@
+{
+  "name": "policy_preprocessor",
+  "steps": [
+    {
+      "registry_name": "rename_observations_processor",
+      "config": {
+        "rename_map": {}
+      }
+    },
+    {
+      "registry_name": "to_batch_processor",
+      "config": {}
+    },
+    {
+      "registry_name": "pi0_new_line_processor",
+      "config": {}
+    },
+    {
+      "registry_name": "tokenizer_processor",
+      "config": {
+        "max_length": 48,
+        "task_key": "task",
+        "padding_side": "right",
+        "padding": "max_length",
+        "truncation": true,
+        "tokenizer_name": "google/paligemma-3b-pt-224"
+      }
+    },
+    {
+      "registry_name": "device_processor",
+      "config": {
+        "device": "cuda",
+        "float_dtype": null
+      }
+    },
+    {
+      "registry_name": "normalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "observation.images.OBS_HEAD": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              480,
+              640
+            ]
+          },
+          "observation.images.OBS_WRIST_LEFT": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              480,
+              640
+            ]
+          },
+          "observation.images.OBS_WRIST_RIGHT": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              480,
+              640
+            ]
+          },
+          "observation.state": {
+            "type": "STATE",
+            "shape": [
+              44
+            ]
+          },
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              19
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "MEAN_STD",
+          "ACTION": "MEAN_STD"
+        }
+      },
+      "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
+    }
+  ]
+}

policy_preprocessor_step_5_normalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89167698c8d5dcec1ceec420c9aa97552e732a27b5e2e095e8c112cbdf99ca4c
+size 15772