metadata
license: apache-2.0
language:
- en
pipeline_tag: image-classification
library_name: pytorch
tags:
- robotics
- humanoid
- image-classification
- task-planning
- cpu-inference
model-index:
- name: humanoid-brain-v1
results:
- task:
type: image-classification
name: 5-task chore classification
dataset:
name: Isaac Eval (500 samples)
type: custom
metrics:
- type: accuracy
value: 97.6
name: overall_accuracy
Humanoid Brain v1
MobileNetV3-Small fine-tuned for 5 household chores:
- cleaning
- cooking
- dishwashing
- laundry
- organizing
Usage
Use with the SDK:
https://github.com/krishthesmart/humanoid-brain
Checkpoint
model.pt
Reported Eval
- Overall: 97.6% (488/500)
- Cleaning: 92%
- Cooking: 100%
- Dishwashing: 100%
- Laundry: 100%
- Organizing: 96%