Update app.py
Browse files
app.py
CHANGED
|
@@ -1,27 +1,23 @@
|
|
|
|
|
| 1 |
from huggingface_hub import from_pretrained_fastai
|
| 2 |
import gradio as gr
|
| 3 |
-
from fastai.text.all import *
|
| 4 |
|
| 5 |
-
#
|
| 6 |
repo_id = "luis56125/news2"
|
|
|
|
|
|
|
| 7 |
learner = from_pretrained_fastai(repo_id)
|
| 8 |
|
| 9 |
-
#
|
| 10 |
labels = ['Mundo', 'Deportes', 'Negocios', 'Ciencia/Tecnolog铆a']
|
| 11 |
|
| 12 |
-
#
|
| 13 |
def predict(text):
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
|
|
|
| 17 |
|
| 18 |
-
#
|
| 19 |
-
|
| 20 |
-
"Global stock markets have fallen sharply as investors worry about the potential impact of rising interest rates.",
|
| 21 |
-
"Scientists have discovered a new species of dinosaur that sheds light on the evolutionary history of reptiles.",
|
| 22 |
-
"The local sports team won their championship game after a stunning comeback in the second half.",
|
| 23 |
-
"New advancements in artificial intelligence are revolutionizing how we interact with technology."
|
| 24 |
-
]
|
| 25 |
|
| 26 |
-
# Crea y lanza la interfaz de Gradio
|
| 27 |
-
gr.Interface(fn=predict, inputs="text", outputs=gr.outputs.Label(num_top_classes=4), examples=examples).launch(share=True)
|
|
|
|
| 1 |
+
# Importaci贸n de las bibliotecas necesarias
|
| 2 |
from huggingface_hub import from_pretrained_fastai
|
| 3 |
import gradio as gr
|
|
|
|
| 4 |
|
| 5 |
+
# Identificador del repositorio en Hugging Face donde est谩 almacenado el modelo
|
| 6 |
repo_id = "luis56125/news2"
|
| 7 |
+
|
| 8 |
+
# Cargar el modelo preentrenado desde Hugging Face
|
| 9 |
learner = from_pretrained_fastai(repo_id)
|
| 10 |
|
| 11 |
+
# Definir las etiquetas de clasificaci贸n disponibles
|
| 12 |
labels = ['Mundo', 'Deportes', 'Negocios', 'Ciencia/Tecnolog铆a']
|
| 13 |
|
| 14 |
+
# Definir una funci贸n para predecir la categor铆a de un texto dado
|
| 15 |
def predict(text):
|
| 16 |
+
# Obtener las probabilidades de las etiquetas desde el modelo
|
| 17 |
+
probs = learner.predict(text)
|
| 18 |
+
# Devolver un diccionario que mapea cada etiqueta a su probabilidad correspondiente
|
| 19 |
+
return {labels[i]: float(probs[i]) for i in range(len(labels))}
|
| 20 |
|
| 21 |
+
# Crear una interfaz de usuario para el modelo utilizando Gradio
|
| 22 |
+
gr.Interface(fn=predict, inputs="text", outputs=gr.components.Label(num_top_classes=5)).launch(share=False)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
|
|
|
|
|
|