Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Web3-Humanoid-ActionNet-v1
|
| 2 |
+
|
| 3 |
+
## Overview
|
| 4 |
+
Web3-Humanoid-ActionNet-v1 is a lightweight NLP model that converts natural language commands into structured humanoid robot action intents.
|
| 5 |
+
|
| 6 |
+
Designed for robotics automation and Web3-based AI ecosystems.
|
| 7 |
+
|
| 8 |
+
## Model Details
|
| 9 |
+
- Task: Text Classification
|
| 10 |
+
- Framework: PyTorch (Transformers)
|
| 11 |
+
- Base Model: DistilBERT
|
| 12 |
+
- Language: English
|
| 13 |
+
- Output: Structured JSON intent
|
| 14 |
+
|
| 15 |
+
## Example
|
| 16 |
+
|
| 17 |
+
Input:
|
| 18 |
+
"Move the red box to the table."
|
| 19 |
+
|
| 20 |
+
Output:
|
| 21 |
+
{
|
| 22 |
+
"action": "move_object",
|
| 23 |
+
"object": "red box",
|
| 24 |
+
"destination": "table"
|
| 25 |
+
}
|
| 26 |
+
|
| 27 |
+
## Tags
|
| 28 |
+
robotics
|
| 29 |
+
web3
|
| 30 |
+
humanoid-ai
|
| 31 |
+
nlp
|
| 32 |
+
intent-classification
|