EXP_007: pi0.5 libero_plus intermediate checkpoint step=24000 (auto-saved, may be superseded)

pi05_single/libero_plus_b8_r16_step24000/config.json

@@ -0,0 +1,108 @@
+{
+  "type": "pi05",
+  "n_obs_steps": 1,
+  "input_features": {
+    "observation.images.front": {
+      "type": "VISUAL",
+      "shape": [
+        3,
+        256,
+        256
+      ]
+    },
+    "observation.images.wrist": {
+      "type": "VISUAL",
+      "shape": [
+        3,
+        256,
+        256
+      ]
+    },
+    "observation.state": {
+      "type": "STATE",
+      "shape": [
+        8
+      ]
+    },
+    "observation.images.empty_camera_0": {
+      "type": "VISUAL",
+      "shape": [
+        3,
+        224,
+        224
+      ]
+    }
+  },
+  "output_features": {
+    "action": {
+      "type": "ACTION",
+      "shape": [
+        7
+      ]
+    }
+  },
+  "device": "cuda",
+  "use_amp": false,
+  "use_peft": false,
+  "push_to_hub": false,
+  "repo_id": null,
+  "private": null,
+  "tags": null,
+  "license": null,
+  "pretrained_revision": null,
+  "paligemma_variant": "gemma_2b",
+  "action_expert_variant": "gemma_300m",
+  "dtype": "bfloat16",
+  "chunk_size": 50,
+  "n_action_steps": 10,
+  "max_state_dim": 32,
+  "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,
+  "use_relative_actions": false,
+  "relative_exclude_joints": [
+    "gripper"
+  ],
+  "action_feature_names": [
+    "action_0",
+    "action_1",
+    "action_2",
+    "action_3",
+    "action_4",
+    "action_5",
+    "action_6"
+  ],
+  "rtc_config": null,
+  "image_resolution": [
+    224,
+    224
+  ],
+  "empty_cameras": 1,
+  "tokenizer_max_length": 200,
+  "normalization_mapping": {
+    "VISUAL": "IDENTITY",
+    "STATE": "MEAN_STD",
+    "ACTION": "MEAN_STD"
+  },
+  "gradient_checkpointing": true,
+  "compile_model": false,
+  "compile_mode": "max-autotune",
+  "freeze_vision_encoder": true,
+  "train_expert_only": true,
+  "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": 15000,
+  "scheduler_decay_lr": 2.5e-06
+}

pi05_single/libero_plus_b8_r16_step24000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d00bceb5de4d55116a1e3149c7fa80a3384811e7e5344ba6075743339a9a8152
+size 9354050752

pi05_single/libero_plus_b8_r16_step24000/policy_postprocessor.json

@@ -0,0 +1,32 @@
+{
+  "name": "policy_postprocessor",
+  "steps": [
+    {
+      "registry_name": "unnormalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              7
+            ]
+          }
+        },
+        "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
+      }
+    }
+  ]
+}

pi05_single/libero_plus_b8_r16_step24000/policy_postprocessor_step_0_unnormalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77c371376b3012b8f2f17a948003817552113ef80152261d3dff9d6f5b9fc6aa
+size 3804

pi05_single/libero_plus_b8_r16_step24000/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": "normalizer_processor",
+      "config": {
+        "eps": 1e-08,
+        "features": {
+          "observation.images.front": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              256,
+              256
+            ]
+          },
+          "observation.images.wrist": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              256,
+              256
+            ]
+          },
+          "observation.state": {
+            "type": "STATE",
+            "shape": [
+              8
+            ]
+          },
+          "observation.images.empty_camera_0": {
+            "type": "VISUAL",
+            "shape": [
+              3,
+              224,
+              224
+            ]
+          },
+          "action": {
+            "type": "ACTION",
+            "shape": [
+              7
+            ]
+          }
+        },
+        "norm_map": {
+          "VISUAL": "IDENTITY",
+          "STATE": "MEAN_STD",
+          "ACTION": "MEAN_STD"
+        }
+      },
+      "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
+    },
+    {
+      "registry_name": "pi05_prepare_state_tokenizer_processor_step",
+      "config": {}
+    },
+    {
+      "registry_name": "tokenizer_processor",
+      "config": {
+        "max_length": 200,
+        "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
+      }
+    }
+  ]
+}

pi05_single/libero_plus_b8_r16_step24000/policy_preprocessor_step_2_normalizer_processor.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac5154012c9fa8ba2de1fcf61de1cb4539f78ba697787eead80399547ffa2c81
+size 3812