Update README.md
Browse files
README.md
CHANGED
|
@@ -5,11 +5,15 @@ tags:
|
|
| 5 |
- vision
|
| 6 |
---
|
| 7 |
|
| 8 |
-
# PTA-1
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
| 11 |
|
|
|
|
| 12 |
|
|
|
|
| 13 |
|
| 14 |
## Model Details
|
| 15 |
|
|
@@ -201,15 +205,3 @@ print(parsed_answer)
|
|
| 201 |
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 202 |
|
| 203 |
[More Information Needed]
|
| 204 |
-
|
| 205 |
-
## More Information [optional]
|
| 206 |
-
|
| 207 |
-
[More Information Needed]
|
| 208 |
-
|
| 209 |
-
## Model Card Authors [optional]
|
| 210 |
-
|
| 211 |
-
[More Information Needed]
|
| 212 |
-
|
| 213 |
-
## Model Card Contact
|
| 214 |
-
|
| 215 |
-
[More Information Needed]
|
|
|
|
| 5 |
- vision
|
| 6 |
---
|
| 7 |
|
| 8 |
+
# PTA-1: Controlling Computers with Small Models
|
| 9 |
|
| 10 |
+
PTA (Prompt-to-Action) is a vision language model for computer use applications based on Florence-2.
|
| 11 |
+
With less than 300M parameters it beats larger models in GUI text and element localization.
|
| 12 |
+
This allows low latency computer automations with local execution.
|
| 13 |
|
| 14 |
+
**Model Input:** Screenshot + description_of_target_element
|
| 15 |
|
| 16 |
+
**Model Output:** BoundingBox for Target Element
|
| 17 |
|
| 18 |
## Model Details
|
| 19 |
|
|
|
|
| 205 |
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 206 |
|
| 207 |
[More Information Needed]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|