import gradio as gr
from huggingface_hub import from_pretrained_keras
from huggingface_hub import KerasModelHubMixin
import transformers
from transformers import AutoTokenizer


m = from_pretrained_keras('sgonzalezsilot/FakeNews-Detection-Twitter-Thesis')
# model = from_pretrained_keras("keras-io/cct")

MODEL = "digitalepidemiologylab/covid-twitter-bert-v2"
tokenizer = AutoTokenizer.from_pretrained(MODEL)

def bert_encode(tokenizer,data,maximum_length) :
  input_ids = []
  attention_masks = []
  

  for i in range(len(data)):
      encoded = tokenizer.encode_plus(
        
        data[i],
        add_special_tokens=True,
        max_length=maximum_length,
        pad_to_max_length=True,
        truncation = True,
        return_attention_mask=True,
      )
      
      input_ids.append(encoded['input_ids'])
      attention_masks.append(encoded['attention_mask'])

  return np.array(input_ids),np.array(attention_masks)
    
# train_encodings = tokenizer(train_texts, truncation=True, padding=True)
# test_encodings = tokenizer(test_texts, truncation=True, padding=True)



    
def get_news(input_text):
    sentence_length = 110
    train_input_ids,train_attention_masks = bert_encode(tokenizer,input_text,sentence_length)    
  return m([train_input_ids,train_attention_masks])

iface = gr.Interface(fn = get_news, 
                     inputs = "text", 
                     outputs = ['text'],
                     title = 'Fake News', 
                     description="")
                     
iface.launch(inline = False)