File size: 1,121 Bytes
1ca8294 93bbccc 1ca8294 cc40b10 1ca8294 a1cb754 eb10631 871c34a 1ca8294 5228e84 1ca8294 4e1e67e |
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 |
import numpy as np
import pandas as pd
import gradio as gr
import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification
labels = ['Not_Adult', 'Adult']
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
device
model_name = 'valurank/finetuned-distilbert-adult-content-detection'
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
def get_adult_content(text):
input_tensor = tokenizer.encode(text, return_tensors='pt', truncation=True)
logits = model(input_tensor).logits
softmax = torch.nn.Softmax(dim=1)
probs = softmax(logits)[0]
probs = probs.cpu().detach().numpy()
#max_index = np.argmax(probs)
adult_content = f"{labels[0]} : {round(probs[0]*100,2)} {labels[1]} : {round(probs[1]*100,2)}"
return adult_content
demo = gr.Interface(get_adult_content, inputs = gr.inputs.Textbox(label= "Input your text here"),
outputs = gr.outputs.Textbox(label='Category'))
if __name__ == "__main__":
demo.launch(debug=True) |