Update README.md
Browse files
README.md
CHANGED
|
@@ -4,29 +4,34 @@ language:
|
|
| 4 |
- en
|
| 5 |
library_name: transformers
|
| 6 |
pipeline_tag: text2text-generation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
tags:
|
| 8 |
- t5
|
| 9 |
- question-answering
|
| 10 |
- knowledge
|
| 11 |
- tiny
|
| 12 |
-
- quantized
|
| 13 |
- efficient
|
| 14 |
- edge
|
| 15 |
- mobile
|
| 16 |
- distillation
|
|
|
|
| 17 |
base_model:
|
| 18 |
- google/t5-v1_1-base
|
| 19 |
-
datasets:
|
| 20 |
-
- custom
|
| 21 |
-
model-index:
|
| 22 |
-
- name: Mosquito
|
| 23 |
-
results: []
|
| 24 |
---
|
| 25 |
|
| 26 |
# ๐ฆ Mosquito - Tiny Knowledge Model
|
| 27 |
|
| 28 |
A **7.3M parameter** T5-based model that answers general knowledge questions. Smaller than a mosquito's brain, but surprisingly capable!
|
| 29 |
|
|
|
|
|
|
|
| 30 |
## โจ Highlights
|
| 31 |
|
| 32 |
| Feature | Value |
|
|
@@ -34,8 +39,9 @@ A **7.3M parameter** T5-based model that answers general knowledge questions. Sm
|
|
| 34 |
| Parameters | 7,263,744 |
|
| 35 |
| Architecture | T5 v1.1 (Gated FFN) |
|
| 36 |
| Size (FP32) | 29 MB |
|
| 37 |
-
| Size (
|
| 38 |
| Training | Knowledge distillation |
|
|
|
|
| 39 |
|
| 40 |
## ๐ Quick Start
|
| 41 |
```python
|
|
@@ -57,6 +63,7 @@ def ask(question):
|
|
| 57 |
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
| 58 |
|
| 59 |
print(ask("Why is the sky blue?"))
|
|
|
|
| 60 |
```
|
| 61 |
|
| 62 |
## ๐ Example Outputs
|
|
|
|
| 4 |
- en
|
| 5 |
library_name: transformers
|
| 6 |
pipeline_tag: text2text-generation
|
| 7 |
+
widget:
|
| 8 |
+
- text: "question: Why is the sky blue?"
|
| 9 |
+
- text: "question: How do vaccines work?"
|
| 10 |
+
- text: "question: What is gravity?"
|
| 11 |
+
- text: "question: How do bees make honey?"
|
| 12 |
+
- text: "question: What causes earthquakes?"
|
| 13 |
+
spaces:
|
| 14 |
+
- ag14850/Mosquito-Demo
|
| 15 |
tags:
|
| 16 |
- t5
|
| 17 |
- question-answering
|
| 18 |
- knowledge
|
| 19 |
- tiny
|
|
|
|
| 20 |
- efficient
|
| 21 |
- edge
|
| 22 |
- mobile
|
| 23 |
- distillation
|
| 24 |
+
- iot
|
| 25 |
base_model:
|
| 26 |
- google/t5-v1_1-base
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
---
|
| 28 |
|
| 29 |
# ๐ฆ Mosquito - Tiny Knowledge Model
|
| 30 |
|
| 31 |
A **7.3M parameter** T5-based model that answers general knowledge questions. Smaller than a mosquito's brain, but surprisingly capable!
|
| 32 |
|
| 33 |
+
**๐ฎ [Try the Live Demo](https://huggingface.co/spaces/ag14850/Mosquito-Demo)**
|
| 34 |
+
|
| 35 |
## โจ Highlights
|
| 36 |
|
| 37 |
| Feature | Value |
|
|
|
|
| 39 |
| Parameters | 7,263,744 |
|
| 40 |
| Architecture | T5 v1.1 (Gated FFN) |
|
| 41 |
| Size (FP32) | 29 MB |
|
| 42 |
+
| Size (quantized + compressed) | ~6 MB |
|
| 43 |
| Training | Knowledge distillation |
|
| 44 |
+
| License | Apache 2.0 |
|
| 45 |
|
| 46 |
## ๐ Quick Start
|
| 47 |
```python
|
|
|
|
| 63 |
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
| 64 |
|
| 65 |
print(ask("Why is the sky blue?"))
|
| 66 |
+
# Rayleigh scattering scatters blue wavelengths, causing blue wavelengths.
|
| 67 |
```
|
| 68 |
|
| 69 |
## ๐ Example Outputs
|