| method: | |
| name: ManiGaussian_BC | |
| use_fabric: true | |
| eval_neural_renderer: false | |
| use_depth: true | |
| use_neural_rendering: true | |
| num_view_for_nerf: 20 | |
| language_model: CLIP | |
| language_model_dim: 512 | |
| image_crop_size: 64 | |
| bounds_offset: | |
| - 0.15 | |
| voxel_sizes: | |
| - 100 | |
| include_prev_layer: false | |
| npoints: 4096 | |
| num_latents: 2048 | |
| latent_dim: 512 | |
| transformer_depth: 6 | |
| transformer_iterations: 1 | |
| cross_heads: 1 | |
| cross_dim_head: 64 | |
| latent_heads: 8 | |
| latent_dim_head: 64 | |
| pos_encoding_with_lang: true | |
| conv_downsample: true | |
| lang_fusion_type: seq | |
| voxel_patch_size: 5 | |
| voxel_patch_stride: 5 | |
| final_dim: 128 | |
| input_dropout: 0.1 | |
| attn_dropout: 0.1 | |
| decoder_dropout: 0.0 | |
| lr: 0.0005 | |
| lr_scheduler: true | |
| num_warmup_steps: 3000 | |
| optimizer: lamb | |
| lambda_weight_l2: 1.0e-06 | |
| trans_loss_weight: 1.0 | |
| rot_loss_weight: 1.0 | |
| grip_loss_weight: 1.0 | |
| collision_loss_weight: 1.0 | |
| rotation_resolution: 5 | |
| activation: lrelu | |
| norm: None | |
| crop_augmentation: true | |
| transform_augmentation: | |
| apply_se3: true | |
| aug_xyz: | |
| - 0.125 | |
| - 0.125 | |
| - 0.125 | |
| aug_rpy: | |
| - 0.0 | |
| - 0.0 | |
| - 45.0 | |
| aug_rot_resolution: ${method.rotation_resolution} | |
| demo_augmentation: true | |
| demo_augmentation_every_n: 10 | |
| no_skip_connection: false | |
| no_perceiver: false | |
| no_language: false | |
| keypoint_method: heuristic | |
| use_wandb: true | |
| lambda_bc: 1.0 | |
| point_cloud_encoder: | |
| NAME: PointNextEncoder | |
| blocks: | |
| - 1 | |
| - 1 | |
| - 1 | |
| - 1 | |
| - 1 | |
| strides: | |
| - 1 | |
| - 2 | |
| - 2 | |
| - 2 | |
| - 2 | |
| width: 128 | |
| in_channels: 6 | |
| sa_layers: 3 | |
| sa_use_res: true | |
| radius: 0.05 | |
| radius_scaling: 2.5 | |
| nsample: 32 | |
| expansion: 4 | |
| aggr_args: | |
| feature_type: dp_fj | |
| reduction: max | |
| group_args: | |
| NAME: ballquery | |
| normalize_dp: true | |
| conv_args: | |
| order: conv-norm-act | |
| act_args: | |
| act: relu | |
| inplace: true | |
| norm_args: | |
| norm: bn | |
| point_cloud_decoder: | |
| layers: 2 | |
| neural_renderer: | |
| visdom: false | |
| render_freq: 2000 | |
| use_clip: false | |
| use_dynamic_field: true | |
| lambda_nerf: 0.01 | |
| lambda_embed: 0.01 | |
| lambda_rgb: 1.0 | |
| lambda_l1: 0.8 | |
| lambda_ssim: 0.2 | |
| lambda_dyna: 0.1 | |
| lambda_reg: 0.0 | |
| dataset: | |
| use_processed_data: true | |
| bg_color: | |
| - 0 | |
| - 0 | |
| - 0 | |
| zfar: 4.0 | |
| znear: 0.1 | |
| trans: | |
| - 0.0 | |
| - 0.0 | |
| - 0.0 | |
| scale: 1.0 | |
| mask_gt_rgb: false | |
| raft: | |
| encoder_dims: | |
| - 32 | |
| - 48 | |
| - 96 | |
| gsnet: | |
| encoder_dims: | |
| - 32 | |
| - 48 | |
| - 96 | |
| decoder_dims: | |
| - 48 | |
| - 64 | |
| - 96 | |
| parm_head_dim: 32 | |
| foundation_model_name: diffusion | |
| d_embed: 3 | |
| loss_embed_fn: cosine | |
| d_latent: 128 | |
| use_depth_supervision: false | |
| d_lang: 128 | |
| voxel_shape: 100 | |
| image_width: 128 | |
| image_height: 128 | |
| z_near: 0.1 | |
| z_far: 4.0 | |
| coordinate_bounds: | |
| - -0.3 | |
| - -0.5 | |
| - 0.6 | |
| - 0.7 | |
| - 0.5 | |
| - 1.6 | |
| use_code: true | |
| use_code_viewdirs: false | |
| use_xyz: true | |
| use_viewdirs: true | |
| mlp: | |
| n_blocks: 5 | |
| d_hidden: 512 | |
| combine_layer: 3 | |
| combine_type: average | |
| beta: 0.0 | |
| use_spade: false | |
| opacity_scale: 1.0 | |
| opacity_bias: -2.0 | |
| scale_bias: 0.02 | |
| scale_scale: 0.003 | |
| xyz_scale: 0.1 | |
| xyz_bias: 0.0 | |
| max_sh_degree: 1 | |
| next_mlp: | |
| d_in: 3 | |
| d_lang: 128 | |
| d_out: 3 | |
| n_blocks: 5 | |
| d_hidden: 512 | |
| combine_layer: 3 | |
| combine_type: average | |
| beta: 0.0 | |
| use_spade: false | |
| warm_up: 3000 | |
| use_action: true | |
| lr: 0.0005 | |
| lambda_weight_l2: 0.0001 | |
| code: | |
| num_freqs: 6 | |
| freq_factor: 1.5 | |
| include_input: true | |
| ddp: | |
| master_addr: localhost | |
| master_port: 12435 | |
| num_devices: 2 | |
| device_id: 0 | |
| rlbench: | |
| task_name: gs_rgb_emb_001_dyna_new_01_0305 | |
| tasks: | |
| - close_jar | |
| - open_drawer | |
| - sweep_to_dustpan_of_size | |
| - meat_off_grill | |
| - turn_tap | |
| - slide_block_to_color_target | |
| - put_item_in_drawer | |
| - reach_and_drag | |
| - push_buttons | |
| - stack_blocks | |
| demos: 20 | |
| num_view_for_nerf: 21 | |
| demo_path: /mnt/disk_1/guanxing/GNFactor/data/train_data | |
| episode_length: 15 | |
| cameras: | |
| - front | |
| camera_resolution: | |
| - 128 | |
| - 128 | |
| scene_bounds: | |
| - -0.3 | |
| - -0.5 | |
| - 0.6 | |
| - 0.7 | |
| - 0.5 | |
| - 1.6 | |
| include_lang_goal_in_obs: true | |
| headless: true | |
| replay: | |
| batch_size: 1 | |
| timesteps: 1 | |
| prioritisation: false | |
| task_uniform: true | |
| use_disk: true | |
| path: /mnt/disk_1/guanxing/GNFactor/replay/gs_rgb_emb_001_dyna_new_01_0305 | |
| max_parallel_processes: 8 | |
| evaluation: | |
| eval_freq: 10000 | |
| eval_episodes: 25 | |
| time_in_state: true | |
| record_every_n: 10000 | |
| episode_length: 15 | |
| gpu: 0 | |
| framework: | |
| use_online_evaluation: false | |
| log_freq: 100 | |
| save_freq: 10000 | |
| train_envs: 1 | |
| replay_ratio: ${replay.batch_size} | |
| transitions_before_train: 200 | |
| tensorboard_logging: false | |
| csv_logging: true | |
| training_iterations: 100010 | |
| gpu: 0 | |
| env_gpu: 0 | |
| logdir: logs/ | |
| logging_level: 20 | |
| seeds: 1 | |
| start_seed: 0 | |
| load_existing_weights: false | |
| num_weights_to_keep: 60 | |
| num_workers: 0 | |
| record_every_n: 5 | |
| use_wandb: true | |
| wandb_project: gnfactor | |
| wandb_group: gs_rgb_emb_001_dyna_new_01_0305 | |
| seed: 0 | |
| wandb_name: gs_rgb_emb_001_dyna_new_01_0305 | |
| tqdm_mininterval: 10 | |