File size: 1,292 Bytes
de6e05c 487660b de6e05c 487660b de6e05c 487660b 08b4bcb 487660b 08b4bcb de6e05c 08b4bcb de6e05c 08b4bcb de6e05c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 | import gradio as gr
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
# Load model and tokenizer from Hugging Face
model_name = "iro-malta07/distilbert-base-german-lang-level-class" # replace with your model path
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
# Use id2label from the config
id2label = model.config.id2label
# Inference function
def classify_text(text):
if not text.endswith("."):
text += "."
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
with torch.no_grad():
outputs = model(**inputs)
probs = torch.nn.functional.softmax(outputs.logits, dim=-1)[0]
# Create a dictionary of labels and their corresponding probabilities
confidences = {id2label[i]: float(probs[i]) for i in range(len(probs))}
return confidences
# Gradio interface
demo = gr.Interface(
fn=classify_text,
inputs=gr.Textbox(lines=4, placeholder="Schreibe etwas auf Deutsch..."),
outputs=gr.Label(num_top_classes=4),
title="German Language Level Classifier",
description="Enter German text and get the predicted CEFR level (A1 to C2). 🚧 Work in progress. 🚧"
)
# Launch app
demo.launch()
|