Upload folder using huggingface_hub

checkpoint-20000/config.json

@@ -0,0 +1,113 @@
+{
+  "action_dim": 32,
+  "action_head_cfg": {
+    "_convert_": "object",
+    "_target_": "groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead",
+    "config": {
+      "_recursive_": false,
+      "_target_": "groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig",
+      "action_dim": 32,
+      "action_horizon": 24,
+      "action_loss_embodiment_ids": [
+        26,
+        17,
+        32
+      ],
+      "add_pos_embed": true,
+      "backbone_embedding_dim": 0,
+      "backbone_features_projector_cfg": null,
+      "decouple_video_action_noise": false,
+      "defer_lora_injection": true,
+      "diffusion_model_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel",
+        "diffusion_model_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P",
+        "dim": 5120,
+        "eps": 1e-06,
+        "ffn_dim": 13824,
+        "frame_seqlen": 880,
+        "freq_dim": 256,
+        "in_dim": 36,
+        "max_chunk_size": 4,
+        "model_type": "i2v",
+        "num_action_per_block": 24,
+        "num_frame_per_block": 2,
+        "num_heads": 40,
+        "num_layers": 40,
+        "num_state_per_block": 1,
+        "out_dim": 16
+      },
+      "expand_batch": null,
+      "freeze_decode_layer": false,
+      "hidden_size": 64,
+      "image_encoder_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder",
+        "image_encoder_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth"
+      },
+      "init_lora_weights": "kaiming",
+      "input_embedding_dim": 1536,
+      "load_pretrained_det_decode_layer_path": null,
+      "lora_alpha": 4,
+      "lora_rank": 4,
+      "lora_target_modules": "q,k,v,o,ffn.0,ffn.2",
+      "max_action_dim": 32,
+      "max_state_dim": 64,
+      "model_dtype": "float32",
+      "noise_beta_alpha": 1.5,
+      "noise_beta_beta": 1.0,
+      "noise_s": 0.999,
+      "num_frame_per_block": 2,
+      "num_frames": 33,
+      "num_inference_timesteps": 4,
+      "num_timestep_buckets": 1000,
+      "repa_coeff": 1.0,
+      "repa_layer": 8,
+      "skip_component_loading": true,
+      "text_encoder_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder",
+        "text_encoder_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth"
+      },
+      "tile_size_height": 34,
+      "tile_size_width": 34,
+      "tile_stride_height": 18,
+      "tile_stride_width": 16,
+      "tiled": false,
+      "train_architecture": "lora",
+      "tune_diffusion_model": true,
+      "tune_projector": true,
+      "use_gradient_checkpointing": true,
+      "use_vlln": true,
+      "vae_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE",
+        "vae_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth"
+      },
+      "video_noise_beta_alpha": 3.0,
+      "video_noise_beta_beta": 1.0,
+      "vl_self_attention_cfg": {
+        "_target_": "groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer",
+        "attention_head_dim": 64,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "num_attention_heads": 24,
+        "num_layers": 4,
+        "positional_embeddings": null
+      }
+    }
+  },
+  "action_horizon": 24,
+  "architectures": [
+    "VLA"
+  ],
+  "backbone_cfg": {
+    "_target_": "groot.vla.model.dreamzero.backbone.identity.IdentityBackbone"
+  },
+  "hidden_size": 0,
+  "model_dtype": "float32",
+  "model_type": "vla",
+  "resume_path": "/root/WAM/checkpoints/dreamzero_bimanual_ur3_lora",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3"
+}

checkpoint-20000/experiment_cfg/conf.yaml

@@ -0,0 +1,3365 @@
+model:
+  _target_: groot.vla.model.dreamzero.base_vla.VLA
+  _convert_: object
+  config:
+    _target_: groot.vla.model.dreamzero.base_vla.VLAConfig
+    _recursive_: false
+    model_dtype: float32
+    hidden_size: 0
+    action_horizon: 24
+    action_dim: 32
+    backbone_cfg:
+      _target_: groot.vla.model.dreamzero.backbone.identity.IdentityBackbone
+    action_head_cfg:
+      config:
+        backbone_features_projector_cfg: null
+        _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig
+        _recursive_: false
+        tiled: false
+        tile_size_height: 34
+        tile_size_width: 34
+        tile_stride_height: 18
+        tile_stride_width: 16
+        lora_rank: 4
+        lora_alpha: 4
+        num_frames: 33
+        num_frame_per_block: 2
+        lora_target_modules: q,k,v,o,ffn.0,ffn.2
+        init_lora_weights: kaiming
+        train_architecture: lora
+        use_gradient_checkpointing: true
+        add_pos_embed: true
+        model_dtype: float32
+        max_state_dim: 64
+        max_action_dim: 32
+        action_loss_embodiment_ids:
+        - 26
+        - 17
+        - 32
+        hidden_size: 64
+        input_embedding_dim: 1536
+        backbone_embedding_dim: 0
+        repa_layer: 8
+        repa_coeff: 1.0
+        load_pretrained_det_decode_layer_path: null
+        freeze_decode_layer: false
+        expand_batch: null
+        use_vlln: true
+        vl_self_attention_cfg:
+          _target_: groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer
+          positional_embeddings: null
+          num_layers: 4
+          num_attention_heads: 24
+          attention_head_dim: 64
+          dropout: 0.2
+          final_dropout: true
+        diffusion_model_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel
+          _convert_: object
+          diffusion_model_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+          model_type: i2v
+          frame_seqlen: 880
+          dim: 5120
+          in_dim: 36
+          ffn_dim: 13824
+          out_dim: 16
+          freq_dim: 256
+          eps: 1.0e-06
+          num_heads: 40
+          num_layers: 40
+          max_chunk_size: 4
+          num_frame_per_block: 2
+          num_action_per_block: 24
+          num_state_per_block: 1
+        text_encoder_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder
+          _convert_: object
+          text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+        image_encoder_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder
+          _convert_: object
+          image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+        vae_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE
+          _convert_: object
+          vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+        action_dim: 32
+        action_horizon: 24
+        num_inference_timesteps: 4
+        noise_beta_alpha: 1.5
+        noise_beta_beta: 1.0
+        noise_s: 0.999
+        num_timestep_buckets: 1000
+        decouple_video_action_noise: false
+        video_noise_beta_alpha: 3.0
+        video_noise_beta_beta: 1.0
+        tune_projector: true
+        tune_diffusion_model: true
+        skip_component_loading: true
+        defer_lora_injection: true
+      _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead
+      _convert_: object
+train_dataset:
+  _target_: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotMixtureDataset.from_mixture_spec
+  _convert_: object
+  mixture_spec:
+  - dataset_path:
+      bimanual_ur3:
+      - /root/WAM/datasets/bimanual-ur3-dreamzero-v2
+    dataset_weight: 1.0
+    distribute_weights: true
+  dataset_class: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotSubLangSingleActionChunkDatasetDROID
+  all_modality_configs:
+    bimanual_ur3:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.task
+    oxe_droid:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.joint_position
+        - state.gripper_position
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.joint_position
+        - action.gripper_position
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.language.language_instruction
+        - annotation.language.language_instruction_2
+        - annotation.language.language_instruction_3
+      lapa_action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - lapa_action
+    agibot:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - -3
+        - -2
+        - -1
+        - 0
+        modality_keys:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.language.action_text
+    yam:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.task
+  all_transforms:
+    bimanual_ur3:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_head
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_head
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_head
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        normalization_modes:
+          state.left_joint_pos: q99
+          state.left_gripper_pos: q99
+          state.right_joint_pos: q99
+          state.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+        normalization_modes:
+          action.left_joint_pos: q99
+          action.left_gripper_pos: q99
+          action.right_joint_pos: q99
+          action.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        state_concat_order:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        action_concat_order:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    oxe_droid:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.joint_position
+        - state.gripper_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.joint_position
+        - state.gripper_position
+        normalization_modes:
+          state.joint_position: q99
+          state.gripper_position: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.joint_position
+        - action.gripper_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.joint_position
+        - action.gripper_position
+        normalization_modes:
+          action.joint_position: q99
+          action.gripper_position: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        state_concat_order:
+        - state.joint_position
+        - state.gripper_position
+        action_concat_order:
+        - action.joint_position
+        - action.gripper_position
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    agibot:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+        normalization_modes:
+          state.left_arm_joint_position: q99
+          state.right_arm_joint_position: q99
+          state.left_effector_position: q99
+          state.right_effector_position: q99
+          state.head_position: q99
+          state.waist_position: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+        normalization_modes:
+          action.left_arm_joint_position: q99
+          action.right_arm_joint_position: q99
+          action.left_effector_position: q99
+          action.right_effector_position: q99
+          action.head_position: q99
+          action.waist_position: q99
+          action.robot_velocity: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        state_concat_order:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+        action_concat_order:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    yam:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        normalization_modes:
+          state.left_joint_pos: q99
+          state.left_gripper_pos: q99
+          state.right_joint_pos: q99
+          state.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+        normalization_modes:
+          action.left_joint_pos: q99
+          action.left_gripper_pos: q99
+          action.right_joint_pos: q99
+          action.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        state_concat_order:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        action_concat_order:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  metadata_versions:
+    bimanual_ur3: '0221'
+    oxe_droid: '0221'
+    agibot: '0221'
+    yam: '0221'
+  fps:
+    bimanual_ur3: 30
+    yam: 30
+  dataset_kwargs:
+    video_backend: decord
+    use_global_metadata: false
+    max_chunk_size: 4
+    relative_action: true
+    relative_action_keys:
+    - left_joint_pos
+    - left_gripper_pos
+    - right_joint_pos
+    - right_gripper_pos
+    relative_action_per_horizon: false
+  mixture_kwargs:
+    training: true
+    balance_dataset_weights: false
+    seed: 42
+    shard_sampling_rate: 0.1
+trainer:
+  _target_: groot.vla.experiment.VLATrainer
+  _partial_: true
+  _recursive_: false
+  callbacks: null
+  model: ???
+  train_dataset: ???
+  compute_dtype: ???
+  benchmark_time: false
+  enable_profiling: false
+  profiling_steps: 5
+  enable_prof_callback: false
+  profile_start_step: 50
+  profile_warmup_steps: 1
+  profile_active_steps: 3
+  profile_record_shapes: false
+  profile_with_stack: false
+  profile_memory: false
+wandb_project: dreamzero
+output_dir: /root/WAM/checkpoints/dreamzero_bimanual_ur3_lora
+load_from_yaml: null
+gear_credentials: null
+upload_checkpoints: false
+upload_every: 1000
+upload_last_n_checkpoints: 5
+remove_unused_columns: false
+bf16: true
+tf32: false
+global_batch_size: null
+raise_error_if_global_batch_size_not_set: false
+per_device_train_batch_size: 1
+per_device_eval_batch_size: 64
+gradient_accumulation_steps: 1
+dataloader_num_workers: 0
+dataloader_pin_memory: false
+dataloader_persistent_workers: true
+optim: adamw_torch
+learning_rate: 0.0001
+adam_beta1: 0.95
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+weight_decay: 0
+lr_scheduler_type: cosine
+warmup_ratio: 0.05
+logging_steps: 10.0
+num_train_epochs: 1000
+max_steps: 20000
+save_strategy: steps
+save_steps: 10000
+eval_strategy: 'no'
+save_total_limit: 2
+report_to: none
+seed: 42
+do_eval: false
+gradient_checkpointing: false
+ddp_find_unused_parameters: false
+ddp_bucket_cap_mb: 100
+ray_num_workers: ???
+eval_bf16: true
+torch_compile_mode: null
+pretrained_model_path: /root/WAM/checkpoints/DreamZero-AgiBot
+only_tune_projectors: false
+save_llm: false
+save_lora_only: true
+save_value_model: false
+save_q_model: false
+download_cache: false
+training_args:
+  _target_: transformers.TrainingArguments
+  output_dir: /root/WAM/checkpoints/dreamzero_bimanual_ur3_lora
+  run_name: dreamzero_bimanual_ur3_lora
+  remove_unused_columns: false
+  deepspeed: groot/vla/configs/deepspeed/zero2.json
+  gradient_checkpointing: true
+  bf16: true
+  tf32: false
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 64
+  gradient_accumulation_steps: 1
+  dataloader_num_workers: 0
+  dataloader_pin_memory: false
+  dataloader_persistent_workers: true
+  optim: adamw_torch
+  adam_beta1: 0.95
+  adam_beta2: 0.999
+  adam_epsilon: 1.0e-08
+  learning_rate: 1.0e-05
+  weight_decay: 0
+  warmup_ratio: 0.05
+  lr_scheduler_type: constant
+  logging_steps: 10.0
+  num_train_epochs: 1000
+  max_steps: 20000
+  save_strategy: steps
+  save_steps: 10000
+  save_total_limit: 2
+  report_to: none
+  seed: 42
+  do_eval: false
+  ddp_find_unused_parameters: false
+  ddp_bucket_cap_mb: 100
+  torch_compile_mode: null
+profile_dir: null
+backbone_hidden_size: 0
+backbone_cfg:
+  _target_: groot.vla.model.dreamzero.backbone.identity.IdentityBackbone
+action_head_cfg:
+  config:
+    backbone_features_projector_cfg: null
+    _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig
+    _recursive_: false
+    tiled: false
+    tile_size_height: 34
+    tile_size_width: 34
+    tile_stride_height: 18
+    tile_stride_width: 16
+    lora_rank: 4
+    lora_alpha: 4
+    num_frames: 33
+    num_frame_per_block: 2
+    lora_target_modules: q,k,v,o,ffn.0,ffn.2
+    init_lora_weights: kaiming
+    train_architecture: lora
+    use_gradient_checkpointing: true
+    add_pos_embed: true
+    model_dtype: float32
+    max_state_dim: 64
+    max_action_dim: 32
+    action_loss_embodiment_ids:
+    - 26
+    - 17
+    - 32
+    hidden_size: 64
+    input_embedding_dim: 1536
+    backbone_embedding_dim: 0
+    repa_layer: 8
+    repa_coeff: 1.0
+    load_pretrained_det_decode_layer_path: null
+    freeze_decode_layer: false
+    expand_batch: null
+    use_vlln: true
+    vl_self_attention_cfg:
+      _target_: groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer
+      positional_embeddings: null
+      num_layers: 4
+      num_attention_heads: 24
+      attention_head_dim: 64
+      dropout: 0.2
+      final_dropout: true
+    diffusion_model_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel
+      _convert_: object
+      diffusion_model_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+      model_type: i2v
+      frame_seqlen: 880
+      dim: 5120
+      in_dim: 36
+      ffn_dim: 13824
+      out_dim: 16
+      freq_dim: 256
+      eps: 1.0e-06
+      num_heads: 40
+      num_layers: 40
+      max_chunk_size: 4
+      num_frame_per_block: 2
+      num_action_per_block: 24
+      num_state_per_block: 1
+    text_encoder_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder
+      _convert_: object
+      text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+    image_encoder_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder
+      _convert_: object
+      image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+    vae_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE
+      _convert_: object
+      vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+    action_dim: 32
+    action_horizon: 24
+    num_inference_timesteps: 4
+    noise_beta_alpha: 1.5
+    noise_beta_beta: 1.0
+    noise_s: 0.999
+    num_timestep_buckets: 1000
+    decouple_video_action_noise: false
+    video_noise_beta_alpha: 3.0
+    video_noise_beta_beta: 1.0
+    tune_projector: true
+    tune_diffusion_model: true
+    skip_component_loading: true
+    defer_lora_injection: true
+  _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead
+  _convert_: object
+add_pos_embed: true
+hidden_size: 64
+attn_dropout: 0.2
+repa_layer: 8
+repa_coeff: 1.0
+load_pretrained_det_decode_layer_path: null
+expand_batch: null
+dit_version: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+train_architecture: lora
+num_frame_per_block: 2
+num_action_per_block: 24
+num_state_per_block: 1
+frame_seqlen: 880
+embodiment_tag_to_projector_index:
+  real_gr1_arms_only: 0
+  real_gr1_arms_only_annotated: 1
+  real_gr1_arms_waist: 2
+  real_gr1_arms_waist_annotated: 3
+  dexmg_gr1_arms_only_inspire: 4
+  dexmg_gr1_arms_only_fourier: 5
+  dexmg_gr1_arms_waist_fourier: 6
+  robocasa_single_arm: 7
+  onex_eve_gripper: 8
+  robocasa_gr1_arms_only_inspire_hands: 9
+  robocasa_gr1_arms_only_fourier_hands: 10
+  robocasa_gr1_fixed_lower_body_inspire_hands: 11
+  robocasa_gr1_fixed_lower_body_fourier_hands: 12
+  robocasa_panda_omron: 13
+  robocasa_bimanual_panda_parallel_gripper: 15
+  robocasa_bimanual_panda_inspire_hand: 16
+  oxe_droid: 17
+  oxe_fractal: 18
+  oxe_language_table: 19
+  oxe_bridge: 20
+  real_panda_single_arm: 21
+  hot3d_hands_only: 23
+  gr1_unified: 24
+  robocasa_gr1_arms_waist_fourier_hands: 25
+  agibot: 26
+  bimanual_ur3: 26
+  lapa: 27
+  oxe_mutex: 28
+  oxe_roboset: 29
+  oxe_plex: 30
+  dream: 31
+  yam: 32
+  xdof: 22
+  gr1_unified_segmentation: 14
+  language_table_sim: 7
+  gr1_isaac: 0
+  sim_behavior_r1_pro: 31
+  mecka_hands: 27
+  real_r1_pro_sharpa: 28
+max_length: 512
+num_views: 3
+tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+data_collator:
+  _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DefaultDataCollator
+  tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  max_length: 512
+  num_views: 3
+  embodiment_tag_mapping:
+    real_gr1_arms_only: 0
+    real_gr1_arms_only_annotated: 1
+    real_gr1_arms_waist: 2
+    real_gr1_arms_waist_annotated: 3
+    dexmg_gr1_arms_only_inspire: 4
+    dexmg_gr1_arms_only_fourier: 5
+    dexmg_gr1_arms_waist_fourier: 6
+    robocasa_single_arm: 7
+    onex_eve_gripper: 8
+    robocasa_gr1_arms_only_inspire_hands: 9
+    robocasa_gr1_arms_only_fourier_hands: 10
+    robocasa_gr1_fixed_lower_body_inspire_hands: 11
+    robocasa_gr1_fixed_lower_body_fourier_hands: 12
+    robocasa_panda_omron: 13
+    robocasa_bimanual_panda_parallel_gripper: 15
+    robocasa_bimanual_panda_inspire_hand: 16
+    oxe_droid: 17
+    oxe_fractal: 18
+    oxe_language_table: 19
+    oxe_bridge: 20
+    real_panda_single_arm: 21
+    hot3d_hands_only: 23
+    gr1_unified: 24
+    robocasa_gr1_arms_waist_fourier_hands: 25
+    agibot: 26
+    bimanual_ur3: 26
+    lapa: 27
+    oxe_mutex: 28
+    oxe_roboset: 29
+    oxe_plex: 30
+    dream: 31
+    yam: 32
+    xdof: 22
+    gr1_unified_segmentation: 14
+    language_table_sim: 7
+    gr1_isaac: 0
+    sim_behavior_r1_pro: 31
+    mecka_hands: 27
+    real_r1_pro_sharpa: 28
+num_visual_tokens_per_frame: 16
+max_state_dim: 64
+max_action_dim: 32
+language_dropout_prob: 0.0
+model_specific_transform:
+  _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+  default_instruction: Perform the default behavior.
+  language_dropout_prob: 0.0
+  always_use_default_instruction: false
+  max_state_dim: 64
+  max_action_dim: 32
+  max_length: 512
+  state_horizon: 1
+  action_horizon: 24
+  embodiment_tag_mapping:
+    real_gr1_arms_only: 0
+    real_gr1_arms_only_annotated: 1
+    real_gr1_arms_waist: 2
+    real_gr1_arms_waist_annotated: 3
+    dexmg_gr1_arms_only_inspire: 4
+    dexmg_gr1_arms_only_fourier: 5
+    dexmg_gr1_arms_waist_fourier: 6
+    robocasa_single_arm: 7
+    onex_eve_gripper: 8
+    robocasa_gr1_arms_only_inspire_hands: 9
+    robocasa_gr1_arms_only_fourier_hands: 10
+    robocasa_gr1_fixed_lower_body_inspire_hands: 11
+    robocasa_gr1_fixed_lower_body_fourier_hands: 12
+    robocasa_panda_omron: 13
+    robocasa_bimanual_panda_parallel_gripper: 15
+    robocasa_bimanual_panda_inspire_hand: 16
+    oxe_droid: 17
+    oxe_fractal: 18
+    oxe_language_table: 19
+    oxe_bridge: 20
+    real_panda_single_arm: 21
+    hot3d_hands_only: 23
+    gr1_unified: 24
+    robocasa_gr1_arms_waist_fourier_hands: 25
+    agibot: 26
+    bimanual_ur3: 26
+    lapa: 27
+    oxe_mutex: 28
+    oxe_roboset: 29
+    oxe_plex: 30
+    dream: 31
+    yam: 32
+    xdof: 22
+    gr1_unified_segmentation: 14
+    language_table_sim: 7
+    gr1_isaac: 0
+    sim_behavior_r1_pro: 31
+    mecka_hands: 27
+    real_r1_pro_sharpa: 28
+  tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+use_global_metadata: false
+num_frames: 33
+action_horizon: 24
+state_horizon: 1
+image_resolution_width: 320
+image_resolution_height: 176
+image_resolution_width_single_frame: 256
+image_resolution_height_single_frame: 256
+totensor_cfg:
+  _target_: groot.vla.data.transform.VideoToTensor
+  apply_to: ???
+crop_cfg:
+  _target_: groot.vla.data.transform.VideoCrop
+  apply_to: ???
+  scale: 0.95
+  mode: random
+resize_cfg:
+  _target_: groot.vla.data.transform.VideoResize
+  apply_to: ???
+  height: 176
+  width: 320
+  interpolation: linear
+resize_cfg_single_frame:
+  _target_: groot.vla.data.transform.VideoResize
+  apply_to: ???
+  height: 256
+  width: 256
+  interpolation: linear
+color_jitter_cfg:
+  _target_: groot.vla.data.transform.VideoColorJitter
+  apply_to: ???
+  brightness: 0.3
+  contrast: 0.4
+  saturation: 0.5
+  hue: 0.08
+random_grayscale_cfg:
+  _target_: groot.vla.data.transform.VideoRandomGrayscale
+  apply_to: ???
+  p: 0.1
+random_posterize_cfg:
+  _target_: groot.vla.data.transform.VideoRandomPosterize
+  apply_to: ???
+  bits: 4
+  p: 0.1
+normalize_cfg:
+  _target_: groot.vla.data.transform.VideoNormalize
+  apply_to: ???
+  mean:
+  - 0.5
+  - 0.5
+  - 0.5
+  std:
+  - 0.5
+  - 0.5
+  - 0.5
+to_numpy_cfg:
+  _target_: groot.vla.data.transform.VideoToNumpy
+  apply_to: ???
+modality_config_bimanual_ur3:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.task
+transform_bimanual_ur3:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_head
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_head
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_head
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    normalization_modes:
+      state.left_joint_pos: q99
+      state.left_gripper_pos: q99
+      state.right_joint_pos: q99
+      state.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+    normalization_modes:
+      action.left_joint_pos: q99
+      action.left_gripper_pos: q99
+      action.right_joint_pos: q99
+      action.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    state_concat_order:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    action_concat_order:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_oxe_droid:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.joint_position
+    - state.gripper_position
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.joint_position
+    - action.gripper_position
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.language.language_instruction
+    - annotation.language.language_instruction_2
+    - annotation.language.language_instruction_3
+  lapa_action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - lapa_action
+transform_oxe_droid:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.joint_position
+    - state.gripper_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.joint_position
+    - state.gripper_position
+    normalization_modes:
+      state.joint_position: q99
+      state.gripper_position: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.joint_position
+    - action.gripper_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.joint_position
+    - action.gripper_position
+    normalization_modes:
+      action.joint_position: q99
+      action.gripper_position: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    state_concat_order:
+    - state.joint_position
+    - state.gripper_position
+    action_concat_order:
+    - action.joint_position
+    - action.gripper_position
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_agibot:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - -3
+    - -2
+    - -1
+    - 0
+    modality_keys:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.language.action_text
+transform_agibot:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+    normalization_modes:
+      state.left_arm_joint_position: q99
+      state.right_arm_joint_position: q99
+      state.left_effector_position: q99
+      state.right_effector_position: q99
+      state.head_position: q99
+      state.waist_position: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+    normalization_modes:
+      action.left_arm_joint_position: q99
+      action.right_arm_joint_position: q99
+      action.left_effector_position: q99
+      action.right_effector_position: q99
+      action.head_position: q99
+      action.waist_position: q99
+      action.robot_velocity: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    state_concat_order:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+    action_concat_order:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_yam:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.task
+transform_yam:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    normalization_modes:
+      state.left_joint_pos: q99
+      state.left_gripper_pos: q99
+      state.right_joint_pos: q99
+      state.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+    normalization_modes:
+      action.left_joint_pos: q99
+      action.left_gripper_pos: q99
+      action.right_joint_pos: q99
+      action.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    state_concat_order:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    action_concat_order:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_configs:
+  bimanual_ur3:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.task
+  oxe_droid:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.joint_position
+      - state.gripper_position
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.joint_position
+      - action.gripper_position
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.language.language_instruction
+      - annotation.language.language_instruction_2
+      - annotation.language.language_instruction_3
+    lapa_action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - lapa_action
+  agibot:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - -3
+      - -2
+      - -1
+      - 0
+      modality_keys:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.language.action_text
+  yam:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.task
+transforms:
+  bimanual_ur3:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_head
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_head
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_head
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      normalization_modes:
+        state.left_joint_pos: q99
+        state.left_gripper_pos: q99
+        state.right_joint_pos: q99
+        state.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+      normalization_modes:
+        action.left_joint_pos: q99
+        action.left_gripper_pos: q99
+        action.right_joint_pos: q99
+        action.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      state_concat_order:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      action_concat_order:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  oxe_droid:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.joint_position
+      - state.gripper_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.joint_position
+      - state.gripper_position
+      normalization_modes:
+        state.joint_position: q99
+        state.gripper_position: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.joint_position
+      - action.gripper_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.joint_position
+      - action.gripper_position
+      normalization_modes:
+        action.joint_position: q99
+        action.gripper_position: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      state_concat_order:
+      - state.joint_position
+      - state.gripper_position
+      action_concat_order:
+      - action.joint_position
+      - action.gripper_position
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  agibot:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+      normalization_modes:
+        state.left_arm_joint_position: q99
+        state.right_arm_joint_position: q99
+        state.left_effector_position: q99
+        state.right_effector_position: q99
+        state.head_position: q99
+        state.waist_position: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+      normalization_modes:
+        action.left_arm_joint_position: q99
+        action.right_arm_joint_position: q99
+        action.left_effector_position: q99
+        action.right_effector_position: q99
+        action.head_position: q99
+        action.waist_position: q99
+        action.robot_velocity: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      state_concat_order:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+      action_concat_order:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  yam:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      normalization_modes:
+        state.left_joint_pos: q99
+        state.left_gripper_pos: q99
+        state.right_joint_pos: q99
+        state.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+      normalization_modes:
+        action.left_joint_pos: q99
+        action.left_gripper_pos: q99
+        action.right_joint_pos: q99
+        action.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      state_concat_order:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      action_concat_order:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+metadata_versions:
+  bimanual_ur3: '0221'
+  oxe_droid: '0221'
+  agibot: '0221'
+  yam: '0221'
+fps:
+  bimanual_ur3: 30
+  yam: 30
+relative_action: true
+relative_action_per_horizon: false
+relative_action_keys:
+- left_joint_pos
+- left_gripper_pos
+- right_joint_pos
+- right_gripper_pos
+max_chunk_size: 4
+dataset_shard_sampling_rate: 0.1
+mixture_dataset_cls: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotMixtureDataset.from_mixture_spec
+single_dataset_cls: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotSubLangSingleActionChunkDatasetDROID
+bimanual_ur3_data_root: /root/WAM/datasets/bimanual-ur3-dreamzero-v2
+total_training_steps: 1048576000000

checkpoint-20000/experiment_cfg/metadata.json

@@ -0,0 +1,387 @@
+{
+    "bimanual_ur3": {
+        "statistics": {
+            "state": {
+                "left_joint_pos": {
+                    "max": [
+                        0.39272642135620117,
+                        -1.5633906126022339,
+                        -0.503991425037384,
+                        3.272645950317383,
+                        -0.9481766819953918,
+                        -2.904858350753784
+                    ],
+                    "min": [
+                        -0.3771570324897766,
+                        -2.094501256942749,
+                        -2.080329179763794,
+                        1.7873648405075073,
+                        -2.147711992263794,
+                        -4.1124958992004395
+                    ],
+                    "mean": [
+                        0.10325526553666377,
+                        -1.8736811553569497,
+                        -1.0983581673333747,
+                        2.6169069817996595,
+                        -1.4701562630278233,
+                        -3.579471877655984
+                    ],
+                    "std": [
+                        0.17736679806135755,
+                        0.10570617479117343,
+                        0.39201607100081914,
+                        0.21384924497566868,
+                        0.1409157498761467,
+                        0.1395353617577876
+                    ],
+                    "q01": [
+                        -0.3711579442024231,
+                        -2.0694878101348877,
+                        -2.0616300106048584,
+                        2.2025333213806153,
+                        -1.9255082273483277,
+                        -3.9785914421081543
+                    ],
+                    "q99": [
+                        0.31588836431503253,
+                        -1.621524691581726,
+                        -0.6770879673957957,
+                        3.0869898796081543,
+                        -1.0170634984970093,
+                        -3.125347137451172
+                    ]
+                },
+                "left_gripper_pos": {
+                    "max": [
+                        155.0
+                    ],
+                    "min": [
+                        5.0
+                    ],
+                    "mean": [
+                        68.07497063362277
+                    ],
+                    "std": [
+                        71.90392175281741
+                    ],
+                    "q01": [
+                        5.0
+                    ],
+                    "q99": [
+                        155.0
+                    ]
+                },
+                "right_joint_pos": {
+                    "max": [
+                        0.6228333115577698,
+                        -0.9518941044807434,
+                        2.2040534019470215,
+                        1.1469651460647583,
+                        1.8123197555541992,
+                        -2.0412919521331787
+                    ],
+                    "min": [
+                        -0.6242511868476868,
+                        -1.8602594137191772,
+                        0.7741227149963379,
+                        -0.7902472615242004,
+                        0.896422266960144,
+                        -3.2143375873565674
+                    ],
+                    "mean": [
+                        -0.1711589763471418,
+                        -1.5248705336817323,
+                        1.4762725497740736,
+                        0.31723919273849904,
+                        1.2736636019767307,
+                        -2.712102100543118
+                    ],
+                    "std": [
+                        0.17855165358274844,
+                        0.11935789788901842,
+                        0.38049586148004083,
+                        0.3494334597427187,
+                        0.16145563581987069,
+                        0.16614453516884894
+                    ],
+                    "q01": [
+                        -0.48826812028884886,
+                        -1.7563303709030151,
+                        0.8764972686767578,
+                        -0.46909457445144653,
+                        0.9555309891700745,
+                        -3.110320014953613
+                    ],
+                    "q99": [
+                        0.3969346547126749,
+                        -1.1910033559799198,
+                        2.0898265838623047,
+                        0.9136980009078914,
+                        1.6710990762710514,
+                        -2.247980833053589
+                    ]
+                },
+                "right_gripper_pos": {
+                    "max": [
+                        155.0
+                    ],
+                    "min": [
+                        5.0
+                    ],
+                    "mean": [
+                        48.03289034249257
+                    ],
+                    "std": [
+                        65.83742744400679
+                    ],
+                    "q01": [
+                        5.0
+                    ],
+                    "q99": [
+                        155.0
+                    ]
+                }
+            },
+            "action": {
+                "left_joint_pos": {
+                    "max": [
+                        0.6409491896629333,
+                        0.4393500089645386,
+                        1.2988891005516052,
+                        0.6607646942138672,
+                        0.5198572874069214,
+                        0.661085844039917
+                    ],
+                    "min": [
+                        -0.5683311522006989,
+                        -0.26938962936401367,
+                        -1.0212045311927795,
+                        -0.9971405267715454,
+                        -0.7243021726608276,
+                        -0.6024825572967529
+                    ],
+                    "mean": [
+                        -0.0015084524820218705,
+                        0.0005983377875678123,
+                        -0.002290424559184592,
+                        0.0010431055188736083,
+                        -0.002063402976802424,
+                        -0.0029885438566056455
+                    ],
+                    "std": [
+                        0.07640505146242331,
+                        0.04757218204602487,
+                        0.1675086447372541,
+                        0.1057793520625608,
+                        0.06997368997382104,
+                        0.04410573785499074
+                    ],
+                    "q01": [
+                        -0.2917717511765659,
+                        -0.1453661823272705,
+                        -0.5652737158536911,
+                        -0.36561367511749265,
+                        -0.2500824546813965,
+                        -0.1520458436012268
+                    ],
+                    "q99": [
+                        0.290224511772394,
+                        0.23219835758209217,
+                        0.7255090385675427,
+                        0.30960572719573964,
+                        0.23530951380729673,
+                        0.18120715618133543
+                    ]
+                },
+                "left_gripper_pos": {
+                    "max": [
+                        150.0
+                    ],
+                    "min": [
+                        -150.0
+                    ],
+                    "mean": [
+                        -0.436380169468027
+                    ],
+                    "std": [
+                        38.56858874870469
+                    ],
+                    "q01": [
+                        -150.0
+                    ],
+                    "q99": [
+                        150.0
+                    ]
+                },
+                "right_joint_pos": {
+                    "max": [
+                        0.5542019307613373,
+                        0.42108917236328125,
+                        1.0793523788452148,
+                        1.1910197138786316,
+                        0.5364850759506226,
+                        0.5202045440673828
+                    ],
+                    "min": [
+                        -0.9044644236564636,
+                        -0.5416398048400879,
+                        -1.038888931274414,
+                        -0.984471507370472,
+                        -0.43937909603118896,
+                        -0.5892429351806641
+                    ],
+                    "mean": [
+                        0.000371238607641002,
+                        -0.00011677871996272672,
+                        0.0018413017534959024,
+                        -0.0020025848168666624,
+                        0.001957996548078153,
+                        0.003056476145326755
+                    ],
+                    "std": [
+                        0.07499698466064719,
+                        0.06338158643951709,
+                        0.15420383482975933,
+                        0.1301089103891752,
+                        0.05682751547035365,
+                        0.060386757551948916
+                    ],
+                    "q01": [
+                        -0.2960817728936672,
+                        -0.2291726875305176,
+                        -0.44958734035491943,
+                        -0.5362981685623527,
+                        -0.16026816844940187,
+                        -0.20857037544250487
+                    ],
+                    "q99": [
+                        0.22611151792109,
+                        0.20003667354583732,
+                        0.6424705171585082,
+                        0.3394414639472961,
+                        0.25204388022422775,
+                        0.17610403060913082
+                    ]
+                },
+                "right_gripper_pos": {
+                    "max": [
+                        150.0
+                    ],
+                    "min": [
+                        -150.0
+                    ],
+                    "mean": [
+                        -0.2119125094805862
+                    ],
+                    "std": [
+                        34.051634282158616
+                    ],
+                    "q01": [
+                        -147.0
+                    ],
+                    "q99": [
+                        142.0
+                    ]
+                }
+            }
+        },
+        "modalities": {
+            "video": {
+                "top_head": {
+                    "resolution": [
+                        1280,
+                        720
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                },
+                "hand_left": {
+                    "resolution": [
+                        640,
+                        480
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                },
+                "hand_right": {
+                    "resolution": [
+                        640,
+                        480
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                }
+            },
+            "state": {
+                "left_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "left_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                },
+                "right_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "right_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                }
+            },
+            "action": {
+                "left_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "left_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                },
+                "right_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "right_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                }
+            }
+        },
+        "embodiment_tag": "bimanual_ur3"
+    }
+}

checkpoint-20000/global_step20000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:255e08d46aa49a3a1a92b0298a9f6895ef1d40f11bd2286729784bd0284acb77
+size 1303094149

checkpoint-20000/global_step20000/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e559e2560fdf71b3858d9fabf593374fc7801450149a6c3d35620d06054cb245
+size 91558582566

checkpoint-20000/latest

@@ -0,0 +1 @@
+global_step20000

checkpoint-20000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb5aa0a5bb8b914d4cdf9ee03bf3080e95a20f1df2d1c50143186a068dfe92f0
+size 217288600

checkpoint-20000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c105e91d5e5f88b59fa69d20bc1422af072f013e832c407a0f00ed9ade2d426f
+size 14645

checkpoint-20000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0752ebe14f8f86c1c2f8f49e37343edfebdb2c5d9040760ec836a63304480ebd
+size 1465

checkpoint-20000/wandb_config.json

@@ -0,0 +1 @@
+{"project": "dreamzero", "run_id": ""}

checkpoint-20000/zero_to_fp32.py

@@ -0,0 +1,758 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:
+            shared_tensor = state_dict[converted_tensors[tensor_id]]
+            state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                state_dict[name] = tensor.contiguous()
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in shard_state_dict:
+            del state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

config.json

@@ -0,0 +1,113 @@
+{
+  "action_dim": 32,
+  "action_head_cfg": {
+    "_convert_": "object",
+    "_target_": "groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead",
+    "config": {
+      "_recursive_": false,
+      "_target_": "groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig",
+      "action_dim": 32,
+      "action_horizon": 24,
+      "action_loss_embodiment_ids": [
+        26,
+        17,
+        32
+      ],
+      "add_pos_embed": true,
+      "backbone_embedding_dim": 0,
+      "backbone_features_projector_cfg": null,
+      "decouple_video_action_noise": false,
+      "defer_lora_injection": true,
+      "diffusion_model_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel",
+        "diffusion_model_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P",
+        "dim": 5120,
+        "eps": 1e-06,
+        "ffn_dim": 13824,
+        "frame_seqlen": 880,
+        "freq_dim": 256,
+        "in_dim": 36,
+        "max_chunk_size": 4,
+        "model_type": "i2v",
+        "num_action_per_block": 24,
+        "num_frame_per_block": 2,
+        "num_heads": 40,
+        "num_layers": 40,
+        "num_state_per_block": 1,
+        "out_dim": 16
+      },
+      "expand_batch": null,
+      "freeze_decode_layer": false,
+      "hidden_size": 64,
+      "image_encoder_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder",
+        "image_encoder_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth"
+      },
+      "init_lora_weights": "kaiming",
+      "input_embedding_dim": 1536,
+      "load_pretrained_det_decode_layer_path": null,
+      "lora_alpha": 4,
+      "lora_rank": 4,
+      "lora_target_modules": "q,k,v,o,ffn.0,ffn.2",
+      "max_action_dim": 32,
+      "max_state_dim": 64,
+      "model_dtype": "float32",
+      "noise_beta_alpha": 1.5,
+      "noise_beta_beta": 1.0,
+      "noise_s": 0.999,
+      "num_frame_per_block": 2,
+      "num_frames": 33,
+      "num_inference_timesteps": 4,
+      "num_timestep_buckets": 1000,
+      "repa_coeff": 1.0,
+      "repa_layer": 8,
+      "skip_component_loading": true,
+      "text_encoder_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder",
+        "text_encoder_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth"
+      },
+      "tile_size_height": 34,
+      "tile_size_width": 34,
+      "tile_stride_height": 18,
+      "tile_stride_width": 16,
+      "tiled": false,
+      "train_architecture": "lora",
+      "tune_diffusion_model": true,
+      "tune_projector": true,
+      "use_gradient_checkpointing": true,
+      "use_vlln": true,
+      "vae_cfg": {
+        "_convert_": "object",
+        "_target_": "groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE",
+        "vae_pretrained_path": "/root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth"
+      },
+      "video_noise_beta_alpha": 3.0,
+      "video_noise_beta_beta": 1.0,
+      "vl_self_attention_cfg": {
+        "_target_": "groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer",
+        "attention_head_dim": 64,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "num_attention_heads": 24,
+        "num_layers": 4,
+        "positional_embeddings": null
+      }
+    }
+  },
+  "action_horizon": 24,
+  "architectures": [
+    "VLA"
+  ],
+  "backbone_cfg": {
+    "_target_": "groot.vla.model.dreamzero.backbone.identity.IdentityBackbone"
+  },
+  "hidden_size": 0,
+  "model_dtype": "float32",
+  "model_type": "vla",
+  "resume_path": "/root/WAM/checkpoints/dreamzero_bimanual_ur3_lora",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3"
+}

experiment_cfg/conf.yaml

@@ -0,0 +1,3365 @@
+model:
+  _target_: groot.vla.model.dreamzero.base_vla.VLA
+  _convert_: object
+  config:
+    _target_: groot.vla.model.dreamzero.base_vla.VLAConfig
+    _recursive_: false
+    model_dtype: float32
+    hidden_size: 0
+    action_horizon: 24
+    action_dim: 32
+    backbone_cfg:
+      _target_: groot.vla.model.dreamzero.backbone.identity.IdentityBackbone
+    action_head_cfg:
+      config:
+        backbone_features_projector_cfg: null
+        _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig
+        _recursive_: false
+        tiled: false
+        tile_size_height: 34
+        tile_size_width: 34
+        tile_stride_height: 18
+        tile_stride_width: 16
+        lora_rank: 4
+        lora_alpha: 4
+        num_frames: 33
+        num_frame_per_block: 2
+        lora_target_modules: q,k,v,o,ffn.0,ffn.2
+        init_lora_weights: kaiming
+        train_architecture: lora
+        use_gradient_checkpointing: true
+        add_pos_embed: true
+        model_dtype: float32
+        max_state_dim: 64
+        max_action_dim: 32
+        action_loss_embodiment_ids:
+        - 26
+        - 17
+        - 32
+        hidden_size: 64
+        input_embedding_dim: 1536
+        backbone_embedding_dim: 0
+        repa_layer: 8
+        repa_coeff: 1.0
+        load_pretrained_det_decode_layer_path: null
+        freeze_decode_layer: false
+        expand_batch: null
+        use_vlln: true
+        vl_self_attention_cfg:
+          _target_: groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer
+          positional_embeddings: null
+          num_layers: 4
+          num_attention_heads: 24
+          attention_head_dim: 64
+          dropout: 0.2
+          final_dropout: true
+        diffusion_model_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel
+          _convert_: object
+          diffusion_model_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+          model_type: i2v
+          frame_seqlen: 880
+          dim: 5120
+          in_dim: 36
+          ffn_dim: 13824
+          out_dim: 16
+          freq_dim: 256
+          eps: 1.0e-06
+          num_heads: 40
+          num_layers: 40
+          max_chunk_size: 4
+          num_frame_per_block: 2
+          num_action_per_block: 24
+          num_state_per_block: 1
+        text_encoder_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder
+          _convert_: object
+          text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+        image_encoder_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder
+          _convert_: object
+          image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+        vae_cfg:
+          _target_: groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE
+          _convert_: object
+          vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+        action_dim: 32
+        action_horizon: 24
+        num_inference_timesteps: 4
+        noise_beta_alpha: 1.5
+        noise_beta_beta: 1.0
+        noise_s: 0.999
+        num_timestep_buckets: 1000
+        decouple_video_action_noise: false
+        video_noise_beta_alpha: 3.0
+        video_noise_beta_beta: 1.0
+        tune_projector: true
+        tune_diffusion_model: true
+        skip_component_loading: true
+        defer_lora_injection: true
+      _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead
+      _convert_: object
+train_dataset:
+  _target_: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotMixtureDataset.from_mixture_spec
+  _convert_: object
+  mixture_spec:
+  - dataset_path:
+      bimanual_ur3:
+      - /root/WAM/datasets/bimanual-ur3-dreamzero-v2
+    dataset_weight: 1.0
+    distribute_weights: true
+  dataset_class: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotSubLangSingleActionChunkDatasetDROID
+  all_modality_configs:
+    bimanual_ur3:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.task
+    oxe_droid:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.joint_position
+        - state.gripper_position
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.joint_position
+        - action.gripper_position
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.language.language_instruction
+        - annotation.language.language_instruction_2
+        - annotation.language.language_instruction_3
+      lapa_action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - lapa_action
+    agibot:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - -3
+        - -2
+        - -1
+        - 0
+        modality_keys:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.language.action_text
+    yam:
+      video:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        eval_delta_indices:
+        - 0
+        modality_keys:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      state:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      action:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        modality_keys:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      language:
+        _target_: groot.vla.data.dataset.ModalityConfig
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.task
+  all_transforms:
+    bimanual_ur3:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_head
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_head
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_head
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.hand_left
+        - video.hand_right
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        normalization_modes:
+          state.left_joint_pos: q99
+          state.left_gripper_pos: q99
+          state.right_joint_pos: q99
+          state.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+        normalization_modes:
+          action.left_joint_pos: q99
+          action.left_gripper_pos: q99
+          action.right_joint_pos: q99
+          action.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        state_concat_order:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        action_concat_order:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    oxe_droid:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.joint_position
+        - state.gripper_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.joint_position
+        - state.gripper_position
+        normalization_modes:
+          state.joint_position: q99
+          state.gripper_position: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.joint_position
+        - action.gripper_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.joint_position
+        - action.gripper_position
+        normalization_modes:
+          action.joint_position: q99
+          action.gripper_position: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.exterior_image_1_left
+        - video.exterior_image_2_left
+        - video.wrist_image_left
+        state_concat_order:
+        - state.joint_position
+        - state.gripper_position
+        action_concat_order:
+        - action.joint_position
+        - action.gripper_position
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    agibot:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+        normalization_modes:
+          state.left_arm_joint_position: q99
+          state.right_arm_joint_position: q99
+          state.left_effector_position: q99
+          state.right_effector_position: q99
+          state.head_position: q99
+          state.waist_position: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+        normalization_modes:
+          action.left_arm_joint_position: q99
+          action.right_arm_joint_position: q99
+          action.left_effector_position: q99
+          action.right_effector_position: q99
+          action.head_position: q99
+          action.waist_position: q99
+          action.robot_velocity: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_head
+        - video.hand_left
+        - video.hand_right
+        state_concat_order:
+        - state.left_arm_joint_position
+        - state.right_arm_joint_position
+        - state.left_effector_position
+        - state.right_effector_position
+        - state.head_position
+        - state.waist_position
+        action_concat_order:
+        - action.left_arm_joint_position
+        - action.right_arm_joint_position
+        - action.left_effector_position
+        - action.right_effector_position
+        - action.head_position
+        - action.waist_position
+        - action.robot_velocity
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+    yam:
+      _target_: groot.vla.data.transform.ComposedModalityTransform
+      transforms:
+      - _target_: groot.vla.data.transform.VideoToTensor
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      - _target_: groot.vla.data.transform.VideoCrop
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        scale: 0.95
+        mode: random
+      - _target_: groot.vla.data.transform.VideoResize
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        height: 176
+        width: 320
+        interpolation: linear
+      - _target_: groot.vla.data.transform.VideoColorJitter
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        brightness: 0.3
+        contrast: 0.4
+        saturation: 0.5
+        hue: 0.08
+      - _target_: groot.vla.data.transform.VideoToNumpy
+        apply_to:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        normalization_modes:
+          state.left_joint_pos: q99
+          state.left_gripper_pos: q99
+          state.right_joint_pos: q99
+          state.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.StateActionToTensor
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.data.transform.StateActionTransform
+        apply_to:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+        normalization_modes:
+          action.left_joint_pos: q99
+          action.left_gripper_pos: q99
+          action.right_joint_pos: q99
+          action.right_gripper_pos: q99
+      - _target_: groot.vla.data.transform.ConcatTransform
+        video_concat_order:
+        - video.top_camera-images-rgb
+        - video.left_camera-images-rgb
+        - video.right_camera-images-rgb
+        state_concat_order:
+        - state.left_joint_pos
+        - state.left_gripper_pos
+        - state.right_joint_pos
+        - state.right_gripper_pos
+        action_concat_order:
+        - action.left_joint_pos
+        - action.left_gripper_pos
+        - action.right_joint_pos
+        - action.right_gripper_pos
+      - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+        default_instruction: Perform the default behavior.
+        language_dropout_prob: 0.0
+        always_use_default_instruction: false
+        max_state_dim: 64
+        max_action_dim: 32
+        max_length: 512
+        state_horizon: 1
+        action_horizon: 24
+        embodiment_tag_mapping:
+          real_gr1_arms_only: 0
+          real_gr1_arms_only_annotated: 1
+          real_gr1_arms_waist: 2
+          real_gr1_arms_waist_annotated: 3
+          dexmg_gr1_arms_only_inspire: 4
+          dexmg_gr1_arms_only_fourier: 5
+          dexmg_gr1_arms_waist_fourier: 6
+          robocasa_single_arm: 7
+          onex_eve_gripper: 8
+          robocasa_gr1_arms_only_inspire_hands: 9
+          robocasa_gr1_arms_only_fourier_hands: 10
+          robocasa_gr1_fixed_lower_body_inspire_hands: 11
+          robocasa_gr1_fixed_lower_body_fourier_hands: 12
+          robocasa_panda_omron: 13
+          robocasa_bimanual_panda_parallel_gripper: 15
+          robocasa_bimanual_panda_inspire_hand: 16
+          oxe_droid: 17
+          oxe_fractal: 18
+          oxe_language_table: 19
+          oxe_bridge: 20
+          real_panda_single_arm: 21
+          hot3d_hands_only: 23
+          gr1_unified: 24
+          robocasa_gr1_arms_waist_fourier_hands: 25
+          agibot: 26
+          bimanual_ur3: 26
+          lapa: 27
+          oxe_mutex: 28
+          oxe_roboset: 29
+          oxe_plex: 30
+          dream: 31
+          yam: 32
+          xdof: 22
+          gr1_unified_segmentation: 14
+          language_table_sim: 7
+          gr1_isaac: 0
+          sim_behavior_r1_pro: 31
+          mecka_hands: 27
+          real_r1_pro_sharpa: 28
+        tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  metadata_versions:
+    bimanual_ur3: '0221'
+    oxe_droid: '0221'
+    agibot: '0221'
+    yam: '0221'
+  fps:
+    bimanual_ur3: 30
+    yam: 30
+  dataset_kwargs:
+    video_backend: decord
+    use_global_metadata: false
+    max_chunk_size: 4
+    relative_action: true
+    relative_action_keys:
+    - left_joint_pos
+    - left_gripper_pos
+    - right_joint_pos
+    - right_gripper_pos
+    relative_action_per_horizon: false
+  mixture_kwargs:
+    training: true
+    balance_dataset_weights: false
+    seed: 42
+    shard_sampling_rate: 0.1
+trainer:
+  _target_: groot.vla.experiment.VLATrainer
+  _partial_: true
+  _recursive_: false
+  callbacks: null
+  model: ???
+  train_dataset: ???
+  compute_dtype: ???
+  benchmark_time: false
+  enable_profiling: false
+  profiling_steps: 5
+  enable_prof_callback: false
+  profile_start_step: 50
+  profile_warmup_steps: 1
+  profile_active_steps: 3
+  profile_record_shapes: false
+  profile_with_stack: false
+  profile_memory: false
+wandb_project: dreamzero
+output_dir: /root/WAM/checkpoints/dreamzero_bimanual_ur3_lora
+load_from_yaml: null
+gear_credentials: null
+upload_checkpoints: false
+upload_every: 1000
+upload_last_n_checkpoints: 5
+remove_unused_columns: false
+bf16: true
+tf32: false
+global_batch_size: null
+raise_error_if_global_batch_size_not_set: false
+per_device_train_batch_size: 1
+per_device_eval_batch_size: 64
+gradient_accumulation_steps: 1
+dataloader_num_workers: 0
+dataloader_pin_memory: false
+dataloader_persistent_workers: true
+optim: adamw_torch
+learning_rate: 0.0001
+adam_beta1: 0.95
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+weight_decay: 0
+lr_scheduler_type: cosine
+warmup_ratio: 0.05
+logging_steps: 10.0
+num_train_epochs: 1000
+max_steps: 20000
+save_strategy: steps
+save_steps: 10000
+eval_strategy: 'no'
+save_total_limit: 2
+report_to: none
+seed: 42
+do_eval: false
+gradient_checkpointing: false
+ddp_find_unused_parameters: false
+ddp_bucket_cap_mb: 100
+ray_num_workers: ???
+eval_bf16: true
+torch_compile_mode: null
+pretrained_model_path: /root/WAM/checkpoints/DreamZero-AgiBot
+only_tune_projectors: false
+save_llm: false
+save_lora_only: true
+save_value_model: false
+save_q_model: false
+download_cache: false
+training_args:
+  _target_: transformers.TrainingArguments
+  output_dir: /root/WAM/checkpoints/dreamzero_bimanual_ur3_lora
+  run_name: dreamzero_bimanual_ur3_lora
+  remove_unused_columns: false
+  deepspeed: groot/vla/configs/deepspeed/zero2.json
+  gradient_checkpointing: true
+  bf16: true
+  tf32: false
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 64
+  gradient_accumulation_steps: 1
+  dataloader_num_workers: 0
+  dataloader_pin_memory: false
+  dataloader_persistent_workers: true
+  optim: adamw_torch
+  adam_beta1: 0.95
+  adam_beta2: 0.999
+  adam_epsilon: 1.0e-08
+  learning_rate: 1.0e-05
+  weight_decay: 0
+  warmup_ratio: 0.05
+  lr_scheduler_type: constant
+  logging_steps: 10.0
+  num_train_epochs: 1000
+  max_steps: 20000
+  save_strategy: steps
+  save_steps: 10000
+  save_total_limit: 2
+  report_to: none
+  seed: 42
+  do_eval: false
+  ddp_find_unused_parameters: false
+  ddp_bucket_cap_mb: 100
+  torch_compile_mode: null
+profile_dir: null
+backbone_hidden_size: 0
+backbone_cfg:
+  _target_: groot.vla.model.dreamzero.backbone.identity.IdentityBackbone
+action_head_cfg:
+  config:
+    backbone_features_projector_cfg: null
+    _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHeadConfig
+    _recursive_: false
+    tiled: false
+    tile_size_height: 34
+    tile_size_width: 34
+    tile_stride_height: 18
+    tile_stride_width: 16
+    lora_rank: 4
+    lora_alpha: 4
+    num_frames: 33
+    num_frame_per_block: 2
+    lora_target_modules: q,k,v,o,ffn.0,ffn.2
+    init_lora_weights: kaiming
+    train_architecture: lora
+    use_gradient_checkpointing: true
+    add_pos_embed: true
+    model_dtype: float32
+    max_state_dim: 64
+    max_action_dim: 32
+    action_loss_embodiment_ids:
+    - 26
+    - 17
+    - 32
+    hidden_size: 64
+    input_embedding_dim: 1536
+    backbone_embedding_dim: 0
+    repa_layer: 8
+    repa_coeff: 1.0
+    load_pretrained_det_decode_layer_path: null
+    freeze_decode_layer: false
+    expand_batch: null
+    use_vlln: true
+    vl_self_attention_cfg:
+      _target_: groot.vla.model.n1_5.modules.cross_attention_dit.SelfAttentionTransformer
+      positional_embeddings: null
+      num_layers: 4
+      num_attention_heads: 24
+      attention_head_dim: 64
+      dropout: 0.2
+      final_dropout: true
+    diffusion_model_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_dit_action_casual_chunk.CausalWanModel
+      _convert_: object
+      diffusion_model_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+      model_type: i2v
+      frame_seqlen: 880
+      dim: 5120
+      in_dim: 36
+      ffn_dim: 13824
+      out_dim: 16
+      freq_dim: 256
+      eps: 1.0e-06
+      num_heads: 40
+      num_layers: 40
+      max_chunk_size: 4
+      num_frame_per_block: 2
+      num_action_per_block: 24
+      num_state_per_block: 1
+    text_encoder_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_text_encoder.WanTextEncoder
+      _convert_: object
+      text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+    image_encoder_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_image_encoder.WanImageEncoder
+      _convert_: object
+      image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+    vae_cfg:
+      _target_: groot.vla.model.dreamzero.modules.wan_video_vae.WanVideoVAE
+      _convert_: object
+      vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+    action_dim: 32
+    action_horizon: 24
+    num_inference_timesteps: 4
+    noise_beta_alpha: 1.5
+    noise_beta_beta: 1.0
+    noise_s: 0.999
+    num_timestep_buckets: 1000
+    decouple_video_action_noise: false
+    video_noise_beta_alpha: 3.0
+    video_noise_beta_beta: 1.0
+    tune_projector: true
+    tune_diffusion_model: true
+    skip_component_loading: true
+    defer_lora_injection: true
+  _target_: groot.vla.model.dreamzero.action_head.wan_flow_matching_action_tf.WANPolicyHead
+  _convert_: object
+add_pos_embed: true
+hidden_size: 64
+attn_dropout: 0.2
+repa_layer: 8
+repa_coeff: 1.0
+load_pretrained_det_decode_layer_path: null
+expand_batch: null
+dit_version: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P
+text_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth
+image_encoder_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth
+vae_pretrained_path: /root/WAM/checkpoints/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth
+train_architecture: lora
+num_frame_per_block: 2
+num_action_per_block: 24
+num_state_per_block: 1
+frame_seqlen: 880
+embodiment_tag_to_projector_index:
+  real_gr1_arms_only: 0
+  real_gr1_arms_only_annotated: 1
+  real_gr1_arms_waist: 2
+  real_gr1_arms_waist_annotated: 3
+  dexmg_gr1_arms_only_inspire: 4
+  dexmg_gr1_arms_only_fourier: 5
+  dexmg_gr1_arms_waist_fourier: 6
+  robocasa_single_arm: 7
+  onex_eve_gripper: 8
+  robocasa_gr1_arms_only_inspire_hands: 9
+  robocasa_gr1_arms_only_fourier_hands: 10
+  robocasa_gr1_fixed_lower_body_inspire_hands: 11
+  robocasa_gr1_fixed_lower_body_fourier_hands: 12
+  robocasa_panda_omron: 13
+  robocasa_bimanual_panda_parallel_gripper: 15
+  robocasa_bimanual_panda_inspire_hand: 16
+  oxe_droid: 17
+  oxe_fractal: 18
+  oxe_language_table: 19
+  oxe_bridge: 20
+  real_panda_single_arm: 21
+  hot3d_hands_only: 23
+  gr1_unified: 24
+  robocasa_gr1_arms_waist_fourier_hands: 25
+  agibot: 26
+  bimanual_ur3: 26
+  lapa: 27
+  oxe_mutex: 28
+  oxe_roboset: 29
+  oxe_plex: 30
+  dream: 31
+  yam: 32
+  xdof: 22
+  gr1_unified_segmentation: 14
+  language_table_sim: 7
+  gr1_isaac: 0
+  sim_behavior_r1_pro: 31
+  mecka_hands: 27
+  real_r1_pro_sharpa: 28
+max_length: 512
+num_views: 3
+tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+data_collator:
+  _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DefaultDataCollator
+  tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  max_length: 512
+  num_views: 3
+  embodiment_tag_mapping:
+    real_gr1_arms_only: 0
+    real_gr1_arms_only_annotated: 1
+    real_gr1_arms_waist: 2
+    real_gr1_arms_waist_annotated: 3
+    dexmg_gr1_arms_only_inspire: 4
+    dexmg_gr1_arms_only_fourier: 5
+    dexmg_gr1_arms_waist_fourier: 6
+    robocasa_single_arm: 7
+    onex_eve_gripper: 8
+    robocasa_gr1_arms_only_inspire_hands: 9
+    robocasa_gr1_arms_only_fourier_hands: 10
+    robocasa_gr1_fixed_lower_body_inspire_hands: 11
+    robocasa_gr1_fixed_lower_body_fourier_hands: 12
+    robocasa_panda_omron: 13
+    robocasa_bimanual_panda_parallel_gripper: 15
+    robocasa_bimanual_panda_inspire_hand: 16
+    oxe_droid: 17
+    oxe_fractal: 18
+    oxe_language_table: 19
+    oxe_bridge: 20
+    real_panda_single_arm: 21
+    hot3d_hands_only: 23
+    gr1_unified: 24
+    robocasa_gr1_arms_waist_fourier_hands: 25
+    agibot: 26
+    bimanual_ur3: 26
+    lapa: 27
+    oxe_mutex: 28
+    oxe_roboset: 29
+    oxe_plex: 30
+    dream: 31
+    yam: 32
+    xdof: 22
+    gr1_unified_segmentation: 14
+    language_table_sim: 7
+    gr1_isaac: 0
+    sim_behavior_r1_pro: 31
+    mecka_hands: 27
+    real_r1_pro_sharpa: 28
+num_visual_tokens_per_frame: 16
+max_state_dim: 64
+max_action_dim: 32
+language_dropout_prob: 0.0
+model_specific_transform:
+  _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+  default_instruction: Perform the default behavior.
+  language_dropout_prob: 0.0
+  always_use_default_instruction: false
+  max_state_dim: 64
+  max_action_dim: 32
+  max_length: 512
+  state_horizon: 1
+  action_horizon: 24
+  embodiment_tag_mapping:
+    real_gr1_arms_only: 0
+    real_gr1_arms_only_annotated: 1
+    real_gr1_arms_waist: 2
+    real_gr1_arms_waist_annotated: 3
+    dexmg_gr1_arms_only_inspire: 4
+    dexmg_gr1_arms_only_fourier: 5
+    dexmg_gr1_arms_waist_fourier: 6
+    robocasa_single_arm: 7
+    onex_eve_gripper: 8
+    robocasa_gr1_arms_only_inspire_hands: 9
+    robocasa_gr1_arms_only_fourier_hands: 10
+    robocasa_gr1_fixed_lower_body_inspire_hands: 11
+    robocasa_gr1_fixed_lower_body_fourier_hands: 12
+    robocasa_panda_omron: 13
+    robocasa_bimanual_panda_parallel_gripper: 15
+    robocasa_bimanual_panda_inspire_hand: 16
+    oxe_droid: 17
+    oxe_fractal: 18
+    oxe_language_table: 19
+    oxe_bridge: 20
+    real_panda_single_arm: 21
+    hot3d_hands_only: 23
+    gr1_unified: 24
+    robocasa_gr1_arms_waist_fourier_hands: 25
+    agibot: 26
+    bimanual_ur3: 26
+    lapa: 27
+    oxe_mutex: 28
+    oxe_roboset: 29
+    oxe_plex: 30
+    dream: 31
+    yam: 32
+    xdof: 22
+    gr1_unified_segmentation: 14
+    language_table_sim: 7
+    gr1_isaac: 0
+    sim_behavior_r1_pro: 31
+    mecka_hands: 27
+    real_r1_pro_sharpa: 28
+  tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+use_global_metadata: false
+num_frames: 33
+action_horizon: 24
+state_horizon: 1
+image_resolution_width: 320
+image_resolution_height: 176
+image_resolution_width_single_frame: 256
+image_resolution_height_single_frame: 256
+totensor_cfg:
+  _target_: groot.vla.data.transform.VideoToTensor
+  apply_to: ???
+crop_cfg:
+  _target_: groot.vla.data.transform.VideoCrop
+  apply_to: ???
+  scale: 0.95
+  mode: random
+resize_cfg:
+  _target_: groot.vla.data.transform.VideoResize
+  apply_to: ???
+  height: 176
+  width: 320
+  interpolation: linear
+resize_cfg_single_frame:
+  _target_: groot.vla.data.transform.VideoResize
+  apply_to: ???
+  height: 256
+  width: 256
+  interpolation: linear
+color_jitter_cfg:
+  _target_: groot.vla.data.transform.VideoColorJitter
+  apply_to: ???
+  brightness: 0.3
+  contrast: 0.4
+  saturation: 0.5
+  hue: 0.08
+random_grayscale_cfg:
+  _target_: groot.vla.data.transform.VideoRandomGrayscale
+  apply_to: ???
+  p: 0.1
+random_posterize_cfg:
+  _target_: groot.vla.data.transform.VideoRandomPosterize
+  apply_to: ???
+  bits: 4
+  p: 0.1
+normalize_cfg:
+  _target_: groot.vla.data.transform.VideoNormalize
+  apply_to: ???
+  mean:
+  - 0.5
+  - 0.5
+  - 0.5
+  std:
+  - 0.5
+  - 0.5
+  - 0.5
+to_numpy_cfg:
+  _target_: groot.vla.data.transform.VideoToNumpy
+  apply_to: ???
+modality_config_bimanual_ur3:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.task
+transform_bimanual_ur3:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_head
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_head
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_head
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.hand_left
+    - video.hand_right
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    normalization_modes:
+      state.left_joint_pos: q99
+      state.left_gripper_pos: q99
+      state.right_joint_pos: q99
+      state.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+    normalization_modes:
+      action.left_joint_pos: q99
+      action.left_gripper_pos: q99
+      action.right_joint_pos: q99
+      action.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    state_concat_order:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    action_concat_order:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_oxe_droid:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.joint_position
+    - state.gripper_position
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.joint_position
+    - action.gripper_position
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.language.language_instruction
+    - annotation.language.language_instruction_2
+    - annotation.language.language_instruction_3
+  lapa_action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - lapa_action
+transform_oxe_droid:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.joint_position
+    - state.gripper_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.joint_position
+    - state.gripper_position
+    normalization_modes:
+      state.joint_position: q99
+      state.gripper_position: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.joint_position
+    - action.gripper_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.joint_position
+    - action.gripper_position
+    normalization_modes:
+      action.joint_position: q99
+      action.gripper_position: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.exterior_image_1_left
+    - video.exterior_image_2_left
+    - video.wrist_image_left
+    state_concat_order:
+    - state.joint_position
+    - state.gripper_position
+    action_concat_order:
+    - action.joint_position
+    - action.gripper_position
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_agibot:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - -3
+    - -2
+    - -1
+    - 0
+    modality_keys:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.language.action_text
+transform_agibot:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+    normalization_modes:
+      state.left_arm_joint_position: q99
+      state.right_arm_joint_position: q99
+      state.left_effector_position: q99
+      state.right_effector_position: q99
+      state.head_position: q99
+      state.waist_position: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+    normalization_modes:
+      action.left_arm_joint_position: q99
+      action.right_arm_joint_position: q99
+      action.left_effector_position: q99
+      action.right_effector_position: q99
+      action.head_position: q99
+      action.waist_position: q99
+      action.robot_velocity: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_head
+    - video.hand_left
+    - video.hand_right
+    state_concat_order:
+    - state.left_arm_joint_position
+    - state.right_arm_joint_position
+    - state.left_effector_position
+    - state.right_effector_position
+    - state.head_position
+    - state.waist_position
+    action_concat_order:
+    - action.left_arm_joint_position
+    - action.right_arm_joint_position
+    - action.left_effector_position
+    - action.right_effector_position
+    - action.head_position
+    - action.waist_position
+    - action.robot_velocity
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_config_yam:
+  video:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    - 24
+    - 25
+    - 26
+    - 27
+    - 28
+    - 29
+    - 30
+    - 31
+    - 32
+    eval_delta_indices:
+    - 0
+    modality_keys:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  state:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  action:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    - 1
+    - 2
+    - 3
+    - 4
+    - 5
+    - 6
+    - 7
+    - 8
+    - 9
+    - 10
+    - 11
+    - 12
+    - 13
+    - 14
+    - 15
+    - 16
+    - 17
+    - 18
+    - 19
+    - 20
+    - 21
+    - 22
+    - 23
+    modality_keys:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  language:
+    _target_: groot.vla.data.dataset.ModalityConfig
+    delta_indices:
+    - 0
+    modality_keys:
+    - annotation.task
+transform_yam:
+  _target_: groot.vla.data.transform.ComposedModalityTransform
+  transforms:
+  - _target_: groot.vla.data.transform.VideoToTensor
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  - _target_: groot.vla.data.transform.VideoCrop
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    scale: 0.95
+    mode: random
+  - _target_: groot.vla.data.transform.VideoResize
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    height: 176
+    width: 320
+    interpolation: linear
+  - _target_: groot.vla.data.transform.VideoColorJitter
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  - _target_: groot.vla.data.transform.VideoToNumpy
+    apply_to:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    normalization_modes:
+      state.left_joint_pos: q99
+      state.left_gripper_pos: q99
+      state.right_joint_pos: q99
+      state.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.StateActionToTensor
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.data.transform.StateActionTransform
+    apply_to:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+    normalization_modes:
+      action.left_joint_pos: q99
+      action.left_gripper_pos: q99
+      action.right_joint_pos: q99
+      action.right_gripper_pos: q99
+  - _target_: groot.vla.data.transform.ConcatTransform
+    video_concat_order:
+    - video.top_camera-images-rgb
+    - video.left_camera-images-rgb
+    - video.right_camera-images-rgb
+    state_concat_order:
+    - state.left_joint_pos
+    - state.left_gripper_pos
+    - state.right_joint_pos
+    - state.right_gripper_pos
+    action_concat_order:
+    - action.left_joint_pos
+    - action.left_gripper_pos
+    - action.right_joint_pos
+    - action.right_gripper_pos
+  - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+    default_instruction: Perform the default behavior.
+    language_dropout_prob: 0.0
+    always_use_default_instruction: false
+    max_state_dim: 64
+    max_action_dim: 32
+    max_length: 512
+    state_horizon: 1
+    action_horizon: 24
+    embodiment_tag_mapping:
+      real_gr1_arms_only: 0
+      real_gr1_arms_only_annotated: 1
+      real_gr1_arms_waist: 2
+      real_gr1_arms_waist_annotated: 3
+      dexmg_gr1_arms_only_inspire: 4
+      dexmg_gr1_arms_only_fourier: 5
+      dexmg_gr1_arms_waist_fourier: 6
+      robocasa_single_arm: 7
+      onex_eve_gripper: 8
+      robocasa_gr1_arms_only_inspire_hands: 9
+      robocasa_gr1_arms_only_fourier_hands: 10
+      robocasa_gr1_fixed_lower_body_inspire_hands: 11
+      robocasa_gr1_fixed_lower_body_fourier_hands: 12
+      robocasa_panda_omron: 13
+      robocasa_bimanual_panda_parallel_gripper: 15
+      robocasa_bimanual_panda_inspire_hand: 16
+      oxe_droid: 17
+      oxe_fractal: 18
+      oxe_language_table: 19
+      oxe_bridge: 20
+      real_panda_single_arm: 21
+      hot3d_hands_only: 23
+      gr1_unified: 24
+      robocasa_gr1_arms_waist_fourier_hands: 25
+      agibot: 26
+      bimanual_ur3: 26
+      lapa: 27
+      oxe_mutex: 28
+      oxe_roboset: 29
+      oxe_plex: 30
+      dream: 31
+      yam: 32
+      xdof: 22
+      gr1_unified_segmentation: 14
+      language_table_sim: 7
+      gr1_isaac: 0
+      sim_behavior_r1_pro: 31
+      mecka_hands: 27
+      real_r1_pro_sharpa: 28
+    tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+modality_configs:
+  bimanual_ur3:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.task
+  oxe_droid:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.joint_position
+      - state.gripper_position
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.joint_position
+      - action.gripper_position
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.language.language_instruction
+      - annotation.language.language_instruction_2
+      - annotation.language.language_instruction_3
+    lapa_action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - lapa_action
+  agibot:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - -3
+      - -2
+      - -1
+      - 0
+      modality_keys:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.language.action_text
+  yam:
+    video:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      - 24
+      - 25
+      - 26
+      - 27
+      - 28
+      - 29
+      - 30
+      - 31
+      - 32
+      eval_delta_indices:
+      - 0
+      modality_keys:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    state:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    action:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      - 1
+      - 2
+      - 3
+      - 4
+      - 5
+      - 6
+      - 7
+      - 8
+      - 9
+      - 10
+      - 11
+      - 12
+      - 13
+      - 14
+      - 15
+      - 16
+      - 17
+      - 18
+      - 19
+      - 20
+      - 21
+      - 22
+      - 23
+      modality_keys:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    language:
+      _target_: groot.vla.data.dataset.ModalityConfig
+      delta_indices:
+      - 0
+      modality_keys:
+      - annotation.task
+transforms:
+  bimanual_ur3:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_head
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_head
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_head
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.hand_left
+      - video.hand_right
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      normalization_modes:
+        state.left_joint_pos: q99
+        state.left_gripper_pos: q99
+        state.right_joint_pos: q99
+        state.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+      normalization_modes:
+        action.left_joint_pos: q99
+        action.left_gripper_pos: q99
+        action.right_joint_pos: q99
+        action.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      state_concat_order:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      action_concat_order:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  oxe_droid:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.joint_position
+      - state.gripper_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.joint_position
+      - state.gripper_position
+      normalization_modes:
+        state.joint_position: q99
+        state.gripper_position: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.joint_position
+      - action.gripper_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.joint_position
+      - action.gripper_position
+      normalization_modes:
+        action.joint_position: q99
+        action.gripper_position: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.exterior_image_1_left
+      - video.exterior_image_2_left
+      - video.wrist_image_left
+      state_concat_order:
+      - state.joint_position
+      - state.gripper_position
+      action_concat_order:
+      - action.joint_position
+      - action.gripper_position
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  agibot:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+      normalization_modes:
+        state.left_arm_joint_position: q99
+        state.right_arm_joint_position: q99
+        state.left_effector_position: q99
+        state.right_effector_position: q99
+        state.head_position: q99
+        state.waist_position: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+      normalization_modes:
+        action.left_arm_joint_position: q99
+        action.right_arm_joint_position: q99
+        action.left_effector_position: q99
+        action.right_effector_position: q99
+        action.head_position: q99
+        action.waist_position: q99
+        action.robot_velocity: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_head
+      - video.hand_left
+      - video.hand_right
+      state_concat_order:
+      - state.left_arm_joint_position
+      - state.right_arm_joint_position
+      - state.left_effector_position
+      - state.right_effector_position
+      - state.head_position
+      - state.waist_position
+      action_concat_order:
+      - action.left_arm_joint_position
+      - action.right_arm_joint_position
+      - action.left_effector_position
+      - action.right_effector_position
+      - action.head_position
+      - action.waist_position
+      - action.robot_velocity
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+  yam:
+    _target_: groot.vla.data.transform.ComposedModalityTransform
+    transforms:
+    - _target_: groot.vla.data.transform.VideoToTensor
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    - _target_: groot.vla.data.transform.VideoCrop
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      scale: 0.95
+      mode: random
+    - _target_: groot.vla.data.transform.VideoResize
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      height: 176
+      width: 320
+      interpolation: linear
+    - _target_: groot.vla.data.transform.VideoColorJitter
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      brightness: 0.3
+      contrast: 0.4
+      saturation: 0.5
+      hue: 0.08
+    - _target_: groot.vla.data.transform.VideoToNumpy
+      apply_to:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      normalization_modes:
+        state.left_joint_pos: q99
+        state.left_gripper_pos: q99
+        state.right_joint_pos: q99
+        state.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.StateActionToTensor
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.data.transform.StateActionTransform
+      apply_to:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+      normalization_modes:
+        action.left_joint_pos: q99
+        action.left_gripper_pos: q99
+        action.right_joint_pos: q99
+        action.right_gripper_pos: q99
+    - _target_: groot.vla.data.transform.ConcatTransform
+      video_concat_order:
+      - video.top_camera-images-rgb
+      - video.left_camera-images-rgb
+      - video.right_camera-images-rgb
+      state_concat_order:
+      - state.left_joint_pos
+      - state.left_gripper_pos
+      - state.right_joint_pos
+      - state.right_gripper_pos
+      action_concat_order:
+      - action.left_joint_pos
+      - action.left_gripper_pos
+      - action.right_joint_pos
+      - action.right_gripper_pos
+    - _target_: groot.vla.model.dreamzero.transform.dreamzero_cotrain.DreamTransform
+      default_instruction: Perform the default behavior.
+      language_dropout_prob: 0.0
+      always_use_default_instruction: false
+      max_state_dim: 64
+      max_action_dim: 32
+      max_length: 512
+      state_horizon: 1
+      action_horizon: 24
+      embodiment_tag_mapping:
+        real_gr1_arms_only: 0
+        real_gr1_arms_only_annotated: 1
+        real_gr1_arms_waist: 2
+        real_gr1_arms_waist_annotated: 3
+        dexmg_gr1_arms_only_inspire: 4
+        dexmg_gr1_arms_only_fourier: 5
+        dexmg_gr1_arms_waist_fourier: 6
+        robocasa_single_arm: 7
+        onex_eve_gripper: 8
+        robocasa_gr1_arms_only_inspire_hands: 9
+        robocasa_gr1_arms_only_fourier_hands: 10
+        robocasa_gr1_fixed_lower_body_inspire_hands: 11
+        robocasa_gr1_fixed_lower_body_fourier_hands: 12
+        robocasa_panda_omron: 13
+        robocasa_bimanual_panda_parallel_gripper: 15
+        robocasa_bimanual_panda_inspire_hand: 16
+        oxe_droid: 17
+        oxe_fractal: 18
+        oxe_language_table: 19
+        oxe_bridge: 20
+        real_panda_single_arm: 21
+        hot3d_hands_only: 23
+        gr1_unified: 24
+        robocasa_gr1_arms_waist_fourier_hands: 25
+        agibot: 26
+        bimanual_ur3: 26
+        lapa: 27
+        oxe_mutex: 28
+        oxe_roboset: 29
+        oxe_plex: 30
+        dream: 31
+        yam: 32
+        xdof: 22
+        gr1_unified_segmentation: 14
+        language_table_sim: 7
+        gr1_isaac: 0
+        sim_behavior_r1_pro: 31
+        mecka_hands: 27
+        real_r1_pro_sharpa: 28
+      tokenizer_path: /root/WAM/checkpoints/umt5-xxl
+metadata_versions:
+  bimanual_ur3: '0221'
+  oxe_droid: '0221'
+  agibot: '0221'
+  yam: '0221'
+fps:
+  bimanual_ur3: 30
+  yam: 30
+relative_action: true
+relative_action_per_horizon: false
+relative_action_keys:
+- left_joint_pos
+- left_gripper_pos
+- right_joint_pos
+- right_gripper_pos
+max_chunk_size: 4
+dataset_shard_sampling_rate: 0.1
+mixture_dataset_cls: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotMixtureDataset.from_mixture_spec
+single_dataset_cls: groot.vla.data.dataset.lerobot_sharded.ShardedLeRobotSubLangSingleActionChunkDatasetDROID
+bimanual_ur3_data_root: /root/WAM/datasets/bimanual-ur3-dreamzero-v2
+total_training_steps: 1048576000000

experiment_cfg/metadata.json

@@ -0,0 +1,387 @@
+{
+    "bimanual_ur3": {
+        "statistics": {
+            "state": {
+                "left_joint_pos": {
+                    "max": [
+                        0.39272642135620117,
+                        -1.5633906126022339,
+                        -0.503991425037384,
+                        3.272645950317383,
+                        -0.9481766819953918,
+                        -2.904858350753784
+                    ],
+                    "min": [
+                        -0.3771570324897766,
+                        -2.094501256942749,
+                        -2.080329179763794,
+                        1.7873648405075073,
+                        -2.147711992263794,
+                        -4.1124958992004395
+                    ],
+                    "mean": [
+                        0.10325526553666377,
+                        -1.8736811553569497,
+                        -1.0983581673333747,
+                        2.6169069817996595,
+                        -1.4701562630278233,
+                        -3.579471877655984
+                    ],
+                    "std": [
+                        0.17736679806135755,
+                        0.10570617479117343,
+                        0.39201607100081914,
+                        0.21384924497566868,
+                        0.1409157498761467,
+                        0.1395353617577876
+                    ],
+                    "q01": [
+                        -0.3711579442024231,
+                        -2.0694878101348877,
+                        -2.0616300106048584,
+                        2.2025333213806153,
+                        -1.9255082273483277,
+                        -3.9785914421081543
+                    ],
+                    "q99": [
+                        0.31588836431503253,
+                        -1.621524691581726,
+                        -0.6770879673957957,
+                        3.0869898796081543,
+                        -1.0170634984970093,
+                        -3.125347137451172
+                    ]
+                },
+                "left_gripper_pos": {
+                    "max": [
+                        155.0
+                    ],
+                    "min": [
+                        5.0
+                    ],
+                    "mean": [
+                        68.07497063362277
+                    ],
+                    "std": [
+                        71.90392175281741
+                    ],
+                    "q01": [
+                        5.0
+                    ],
+                    "q99": [
+                        155.0
+                    ]
+                },
+                "right_joint_pos": {
+                    "max": [
+                        0.6228333115577698,
+                        -0.9518941044807434,
+                        2.2040534019470215,
+                        1.1469651460647583,
+                        1.8123197555541992,
+                        -2.0412919521331787
+                    ],
+                    "min": [
+                        -0.6242511868476868,
+                        -1.8602594137191772,
+                        0.7741227149963379,
+                        -0.7902472615242004,
+                        0.896422266960144,
+                        -3.2143375873565674
+                    ],
+                    "mean": [
+                        -0.1711589763471418,
+                        -1.5248705336817323,
+                        1.4762725497740736,
+                        0.31723919273849904,
+                        1.2736636019767307,
+                        -2.712102100543118
+                    ],
+                    "std": [
+                        0.17855165358274844,
+                        0.11935789788901842,
+                        0.38049586148004083,
+                        0.3494334597427187,
+                        0.16145563581987069,
+                        0.16614453516884894
+                    ],
+                    "q01": [
+                        -0.48826812028884886,
+                        -1.7563303709030151,
+                        0.8764972686767578,
+                        -0.46909457445144653,
+                        0.9555309891700745,
+                        -3.110320014953613
+                    ],
+                    "q99": [
+                        0.3969346547126749,
+                        -1.1910033559799198,
+                        2.0898265838623047,
+                        0.9136980009078914,
+                        1.6710990762710514,
+                        -2.247980833053589
+                    ]
+                },
+                "right_gripper_pos": {
+                    "max": [
+                        155.0
+                    ],
+                    "min": [
+                        5.0
+                    ],
+                    "mean": [
+                        48.03289034249257
+                    ],
+                    "std": [
+                        65.83742744400679
+                    ],
+                    "q01": [
+                        5.0
+                    ],
+                    "q99": [
+                        155.0
+                    ]
+                }
+            },
+            "action": {
+                "left_joint_pos": {
+                    "max": [
+                        0.6409491896629333,
+                        0.4393500089645386,
+                        1.2988891005516052,
+                        0.6607646942138672,
+                        0.5198572874069214,
+                        0.661085844039917
+                    ],
+                    "min": [
+                        -0.5683311522006989,
+                        -0.26938962936401367,
+                        -1.0212045311927795,
+                        -0.9971405267715454,
+                        -0.7243021726608276,
+                        -0.6024825572967529
+                    ],
+                    "mean": [
+                        -0.0015084524820218705,
+                        0.0005983377875678123,
+                        -0.002290424559184592,
+                        0.0010431055188736083,
+                        -0.002063402976802424,
+                        -0.0029885438566056455
+                    ],
+                    "std": [
+                        0.07640505146242331,
+                        0.04757218204602487,
+                        0.1675086447372541,
+                        0.1057793520625608,
+                        0.06997368997382104,
+                        0.04410573785499074
+                    ],
+                    "q01": [
+                        -0.2917717511765659,
+                        -0.1453661823272705,
+                        -0.5652737158536911,
+                        -0.36561367511749265,
+                        -0.2500824546813965,
+                        -0.1520458436012268
+                    ],
+                    "q99": [
+                        0.290224511772394,
+                        0.23219835758209217,
+                        0.7255090385675427,
+                        0.30960572719573964,
+                        0.23530951380729673,
+                        0.18120715618133543
+                    ]
+                },
+                "left_gripper_pos": {
+                    "max": [
+                        150.0
+                    ],
+                    "min": [
+                        -150.0
+                    ],
+                    "mean": [
+                        -0.436380169468027
+                    ],
+                    "std": [
+                        38.56858874870469
+                    ],
+                    "q01": [
+                        -150.0
+                    ],
+                    "q99": [
+                        150.0
+                    ]
+                },
+                "right_joint_pos": {
+                    "max": [
+                        0.5542019307613373,
+                        0.42108917236328125,
+                        1.0793523788452148,
+                        1.1910197138786316,
+                        0.5364850759506226,
+                        0.5202045440673828
+                    ],
+                    "min": [
+                        -0.9044644236564636,
+                        -0.5416398048400879,
+                        -1.038888931274414,
+                        -0.984471507370472,
+                        -0.43937909603118896,
+                        -0.5892429351806641
+                    ],
+                    "mean": [
+                        0.000371238607641002,
+                        -0.00011677871996272672,
+                        0.0018413017534959024,
+                        -0.0020025848168666624,
+                        0.001957996548078153,
+                        0.003056476145326755
+                    ],
+                    "std": [
+                        0.07499698466064719,
+                        0.06338158643951709,
+                        0.15420383482975933,
+                        0.1301089103891752,
+                        0.05682751547035365,
+                        0.060386757551948916
+                    ],
+                    "q01": [
+                        -0.2960817728936672,
+                        -0.2291726875305176,
+                        -0.44958734035491943,
+                        -0.5362981685623527,
+                        -0.16026816844940187,
+                        -0.20857037544250487
+                    ],
+                    "q99": [
+                        0.22611151792109,
+                        0.20003667354583732,
+                        0.6424705171585082,
+                        0.3394414639472961,
+                        0.25204388022422775,
+                        0.17610403060913082
+                    ]
+                },
+                "right_gripper_pos": {
+                    "max": [
+                        150.0
+                    ],
+                    "min": [
+                        -150.0
+                    ],
+                    "mean": [
+                        -0.2119125094805862
+                    ],
+                    "std": [
+                        34.051634282158616
+                    ],
+                    "q01": [
+                        -147.0
+                    ],
+                    "q99": [
+                        142.0
+                    ]
+                }
+            }
+        },
+        "modalities": {
+            "video": {
+                "top_head": {
+                    "resolution": [
+                        1280,
+                        720
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                },
+                "hand_left": {
+                    "resolution": [
+                        640,
+                        480
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                },
+                "hand_right": {
+                    "resolution": [
+                        640,
+                        480
+                    ],
+                    "channels": 3,
+                    "fps": 30.0
+                }
+            },
+            "state": {
+                "left_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "left_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                },
+                "right_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "right_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                }
+            },
+            "action": {
+                "left_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "left_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                },
+                "right_joint_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        6
+                    ],
+                    "continuous": true
+                },
+                "right_gripper_pos": {
+                    "absolute": true,
+                    "rotation_type": null,
+                    "shape": [
+                        1
+                    ],
+                    "continuous": true
+                }
+            }
+        },
+        "embodiment_tag": "bimanual_ur3"
+    }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb5aa0a5bb8b914d4cdf9ee03bf3080e95a20f1df2d1c50143186a068dfe92f0
+size 217288600

wandb_config.json

@@ -0,0 +1 @@
+{"project": "dreamzero", "run_id": ""}