Upload policy weights, train config and readme

README.md

@@ -1,29 +1,21 @@
 ---
-datasets: rh5_sample_dataset
+datasets: usr256864/lr_ds
 library_name: lerobot
 license: apache-2.0
-model_name: pi0
+model_name: groot
 pipeline_tag: robotics
 tags:
-- pi0
-- lerobot
 - robotics
+- lerobot
+- groot
 ---
 
-# Model Card for pi0
+# Model Card for groot
 
 <!-- Provide a quick summary of what the model is/does. -->
 
 
-**π₀ (Pi0)**
-
-π₀ is a Vision-Language-Action model for general robot control, from Physical Intelligence. The LeRobot implementation is adapted from their open source OpenPI repository.
-
-**Model Overview**
-
-π₀ represents a breakthrough in robotics as the first general-purpose robot foundation model developed by Physical Intelligence. Unlike traditional robots that are narrow specialists programmed for repetitive motions, π₀ is designed to be a generalist policy that can understand visual inputs, interpret natural language instructions, and control a variety of different robots across diverse tasks.
-
-For more details, see the [Physical Intelligence π₀ blog post](https://www.physicalintelligence.company/blog/pi0).
+_Model type not recognized — please update this template._
 
 
 This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).

config.json

@@ -1,5 +1,5 @@
 {
-    "type": "pi0",
+    "type": "groot",
     "n_obs_steps": 1,
     "input_features": {
         "observation.image": {
@@ -27,49 +27,55 @@
     },
     "device": "cuda",
     "use_amp": false,
-    "push_to_hub": false,
+    "use_peft": false,
+    "push_to_hub": true,
     "repo_id": "usr256864/pi0",
     "private": null,
     "tags": null,
     "license": null,
-    "pretrained_path": "outputs/train/2025-11-14/01-13-16_pi0_training/checkpoints/080000/pretrained_model",
-    "paligemma_variant": "gemma_2b",
-    "action_expert_variant": "gemma_300m",
-    "dtype": "bfloat16",
+    "pretrained_path": null,
     "chunk_size": 50,
     "n_action_steps": 50,
-    "max_state_dim": 32,
+    "max_state_dim": 64,
     "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,
-    "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",
-    "optimizer_lr": 2.5e-05,
+    "image_size": [
+        224,
+        224
+    ],
+    "base_model_path": "nvidia/GR00T-N1.5-3B",
+    "tokenizer_assets_repo": "lerobot/eagle2hg-processor-groot-n1p5",
+    "embodiment_tag": "new_embodiment",
+    "tune_llm": false,
+    "tune_visual": false,
+    "tune_projector": true,
+    "tune_diffusion_model": true,
+    "lora_rank": 0,
+    "lora_alpha": 16,
+    "lora_dropout": 0.1,
+    "lora_full_model": false,
+    "optimizer_lr": 0.0001,
     "optimizer_betas": [
-        0.9,
-        0.95
+        0.95,
+        0.999
     ],
     "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
+    "optimizer_weight_decay": 1e-05,
+    "warmup_ratio": 0.05,
+    "use_bf16": true,
+    "video_backend": "decord",
+    "balance_dataset_weights": true,
+    "balance_trajectory_weights": true,
+    "dataset_paths": null,
+    "output_dir": "./tmp/gr00t",
+    "save_steps": 1000,
+    "max_steps": 10000,
+    "batch_size": 32,
+    "dataloader_num_workers": 8,
+    "report_to": "wandb",
+    "resume": false
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7bd1aaac8507f32f420e53e36701f0a3b23fe63a53ef25405384355c1a35e9d
-size 7011548640
+oid sha256:411b556159f0e7518b7ee80cffdc85dc6f5e54c8c5722ae86b30308c349e5b48
+size 6964802952

train_config.json

@@ -1,7 +1,7 @@
 {
     "dataset": {
-        "repo_id": "rh5_sample_dataset",
-        "root": "../datasets/RH5_dataset/",
+        "repo_id": "usr256864/lr_ds",
+        "root": "../datasets/RH5_dataset_qdot/",
         "episodes": null,
         "image_transforms": {
             "enable": false,
@@ -79,9 +79,48 @@
         "video_backend": "torchcodec",
         "streaming": false
     },
-    "env": null,
+    "env": {
+        "type": "rh5",
+        "task": "walk",
+        "fps": 30,
+        "features": {
+            "action": {
+                "type": "ACTION",
+                "shape": [
+                    32
+                ]
+            },
+            "agent_pos": {
+                "type": "STATE",
+                "shape": [
+                    63
+                ]
+            },
+            "pixels/top": {
+                "type": "VISUAL",
+                "shape": [
+                    480,
+                    640,
+                    3
+                ]
+            }
+        },
+        "features_map": {
+            "action": "action",
+            "agent_pos": "observation.state",
+            "top": "observation.image",
+            "pixels/top": "observation.image"
+        },
+        "max_parallel_tasks": 1,
+        "disable_env_checker": true,
+        "config": null,
+        "episode_length": 1000,
+        "obs_type": "pixels_agent_pos",
+        "render_mode": "rgb_array",
+        "use_velocity": false
+    },
     "policy": {
-        "type": "pi0",
+        "type": "groot",
         "n_obs_steps": 1,
         "input_features": {
             "observation.image": {
@@ -109,96 +148,114 @@
         },
         "device": "cuda",
         "use_amp": false,
-        "push_to_hub": false,
+        "use_peft": false,
+        "push_to_hub": true,
         "repo_id": "usr256864/pi0",
         "private": null,
         "tags": null,
         "license": null,
-        "pretrained_path": "outputs/train/2025-11-14/01-13-16_pi0_training/checkpoints/080000/pretrained_model",
-        "paligemma_variant": "gemma_2b",
-        "action_expert_variant": "gemma_300m",
-        "dtype": "bfloat16",
+        "pretrained_path": null,
         "chunk_size": 50,
         "n_action_steps": 50,
-        "max_state_dim": 32,
+        "max_state_dim": 64,
         "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,
-        "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",
-        "optimizer_lr": 2.5e-05,
+        "image_size": [
+            224,
+            224
+        ],
+        "base_model_path": "nvidia/GR00T-N1.5-3B",
+        "tokenizer_assets_repo": "lerobot/eagle2hg-processor-groot-n1p5",
+        "embodiment_tag": "new_embodiment",
+        "tune_llm": false,
+        "tune_visual": false,
+        "tune_projector": true,
+        "tune_diffusion_model": true,
+        "lora_rank": 0,
+        "lora_alpha": 16,
+        "lora_dropout": 0.1,
+        "lora_full_model": false,
+        "optimizer_lr": 0.0001,
         "optimizer_betas": [
-            0.9,
-            0.95
+            0.95,
+            0.999
         ],
         "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
+        "optimizer_weight_decay": 1e-05,
+        "warmup_ratio": 0.05,
+        "use_bf16": true,
+        "video_backend": "decord",
+        "balance_dataset_weights": true,
+        "balance_trajectory_weights": true,
+        "dataset_paths": null,
+        "output_dir": "./tmp/gr00t",
+        "save_steps": 1000,
+        "max_steps": 10000,
+        "batch_size": 32,
+        "dataloader_num_workers": 8,
+        "report_to": "wandb",
+        "resume": false
     },
-    "output_dir": "outputs/train/2025-11-14/01-13-16_pi0_training",
-    "job_name": "pi0_training",
-    "resume": true,
+    "output_dir": "outputs/train/2026-02-11/13-18-22_debug_train",
+    "job_name": "debug_train",
+    "resume": false,
     "seed": 1000,
     "num_workers": 4,
-    "batch_size": 32,
-    "steps": 200000,
-    "eval_freq": 20000,
-    "log_freq": 200,
+    "batch_size": 1,
+    "steps": 2,
+    "eval_freq": 0,
+    "log_freq": 2,
+    "tolerance_s": 0.0001,
     "save_checkpoint": true,
-    "save_freq": 20000,
+    "save_freq": 2,
     "use_policy_training_preset": true,
     "optimizer": {
         "type": "adamw",
-        "lr": 2.5e-05,
-        "weight_decay": 0.01,
-        "grad_clip_norm": 1.0,
+        "lr": 0.0001,
+        "weight_decay": 1e-05,
+        "grad_clip_norm": 10.0,
         "betas": [
-            0.9,
-            0.95
+            0.95,
+            0.999
         ],
         "eps": 1e-08
     },
     "scheduler": {
         "type": "cosine_decay_with_warmup",
-        "num_warmup_steps": 1000,
-        "num_decay_steps": 30000,
-        "peak_lr": 2.5e-05,
-        "decay_lr": 2.5e-06
+        "num_warmup_steps": 500,
+        "num_decay_steps": 10000,
+        "peak_lr": 0.0001,
+        "decay_lr": 1e-05
     },
     "eval": {
-        "n_episodes": 50,
-        "batch_size": 50,
+        "n_episodes": 1,
+        "batch_size": 1,
         "use_async_envs": false
     },
     "wandb": {
-        "enable": true,
+        "enable": false,
         "disable_artifact": true,
         "project": "lerobot",
         "entity": null,
         "notes": null,
-        "run_id": "k9pdbqh0",
+        "run_id": null,
         "mode": null
     },
-    "checkpoint_path": "outputs/train/2025-11-14/01-13-16_pi0_training/checkpoints/080000",
-    "rename_map": {}
+    "peft": null,
+    "use_rabc": false,
+    "rabc_progress_path": null,
+    "rabc_kappa": 0.01,
+    "rabc_epsilon": 1e-06,
+    "rabc_head_mode": "sparse",
+    "rename_map": {},
+    "checkpoint_path": null,
+    "eval_ds": "../datasets/RH5_dataset_qdot/",
+    "episode_idx": 0,
+    "absolute": false,
+    "num_eval_samples": 20,
+    "use_velocity": true
 }