Skip to content

run quick start but error occur: annotate.py: error: unrecognized arguments dataset.path=... dataset.name=... GPU_IDS=... #3

@hanbit-o

Description

@hanbit-o

Hi, thank you for your impressive work.

I try to follow your instructions from installation to quick start for my video.

I completed the installation by following: Installation.

and... I try to do quick start with my video (adroit hammering task).

python script/annotate.py \
    dataset.path="/home/hanbit-o/code/NILS/video" \
    dataset.name="test" \
    GPU_IDS=[0] \
    PROC_PER_GPU=1 \
    embodiment_prompt_grounding_dino="the yellow robotic hand" \
    embodiment_prompt_clipseg="the yellow robotic hand"

but... an error occurs as below:

...
2025-03-31 19:09:29,116 - INFO - Retrieving objects in scene
2025-03-31 19:09:41,309 - INFO - High overlap classes: [{'name': 'hammer', 'color': 'bronze'}]
2025-03-31 19:09:41,309 - INFO - Detected the following objects in the scene: ['table', 'hammer', 'metal bar', 'wooden block', 'hand']
2025-03-31 19:09:43,824 - INFO - Labeling frames
2025-03-31 19:09:48,373 - INFO - Computing image embeddings for the provided images...
2025-03-31 19:09:48,497 - INFO - Image embeddings computed.
usage: annotate.py [-h] [--model MODEL] [--output OUTPUT] [--save_all] [--multirun MULTIRUN] [--key_dim KEY_DIM] [--value_dim VALUE_DIM] [--pix_feat_dim PIX_FEAT_DIM] [--disable_long_term] [--max_mid_term_frames MAX_MID_TERM_FRAMES] [--min_mid_term_frames MIN_MID_TERM_FRAMES] [--max_long_term_elements MAX_LONG_TERM_ELEMENTS] [--num_prototypes NUM_PROTOTYPES]
                   [--top_k TOP_K] [--mem_every MEM_EVERY] [--chunk_size CHUNK_SIZE] [--size SIZE] [--GROUNDING_DINO_CONFIG_PATH GROUNDING_DINO_CONFIG_PATH] [--GROUNDING_DINO_CHECKPOINT_PATH GROUNDING_DINO_CHECKPOINT_PATH] [--DINO_THRESHOLD DINO_THRESHOLD] [--DINO_NMS_THRESHOLD DINO_NMS_THRESHOLD] [--SAM_ENCODER_VERSION SAM_ENCODER_VERSION]
                   [--SAM_CHECKPOINT_PATH SAM_CHECKPOINT_PATH] [--MOBILE_SAM_CHECKPOINT_PATH MOBILE_SAM_CHECKPOINT_PATH] [--SAM_NUM_POINTS_PER_SIDE SAM_NUM_POINTS_PER_SIDE] [--resegment_whole_image RESEGMENT_WHOLE_IMAGE] [--SAM_NUM_POINTS_PER_BATCH SAM_NUM_POINTS_PER_BATCH] [--SAM_PRED_IOU_THRESHOLD SAM_PRED_IOU_THRESHOLD]
                   [--SAM_OVERLAP_THRESHOLD SAM_OVERLAP_THRESHOLD] [--img_path IMG_PATH] [--detection_every DETECTION_EVERY] [--num_voting_frames NUM_VOTING_FRAMES] [--rp_check_interval RP_CHECK_INTERVAL] [--temporal_setting TEMPORAL_SETTING] [--max_missed_detection_count MAX_MISSED_DETECTION_COUNT] [--max_num_objects MAX_NUM_OBJECTS] [--prompt PROMPT]
                   [--sam_variant SAM_VARIANT]
annotate.py: error: unrecognized arguments: dataset.path=/home/hanbit-o/code/NILS/video dataset.name=test GPU_IDS=[0] PROC_PER_GPU=1 embodiment_prompt_grounding_dino=the yellow robotic hand embodiment_prompt_clipseg=the yellow robotic hand
Process: 0 - Labeling Demonstrations:   0%|       

Do you have any advice to address this?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions