Lisibonny commited on
Commit
6558791
1 Parent(s): 90e8f7a

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +5 -4
app.py CHANGED
@@ -3,7 +3,7 @@ import streamlit as st
3
  import streamlit.components.v1 as components
4
  from transformers import *
5
  from carga_articulos import cargar_articulos
6
- from preprocesamiento_articulos import limpieza_articulos
7
  from entrenamiento_modelo import term_document_matrix, tf_idf_score
8
  from resultados_consulta import resultados_consulta, detalles_resultados
9
  import tensorflow as tf
@@ -126,9 +126,10 @@ def main():
126
 
127
  for i in range(len(df_answer)):
128
 
129
- text=df_answer.loc[i, "resumen"]
130
- inputs = tokenizer(query, text, return_tensors='tf')
131
- st.write(f'Aqui {inputs}' )
 
132
  outputs = qa_model(input_ids=inputs['input_ids'], attention_mask=inputs['attention_mask'])
133
  answer_start_index = int(tf.math.argmax(outputs.start_logits, axis=-1)[0])
134
  answer_end_index = int(tf.math.argmax(outputs.end_logits, axis=-1)[0])
 
3
  import streamlit.components.v1 as components
4
  from transformers import *
5
  from carga_articulos import cargar_articulos
6
+ from preprocesamiento_articulos import limpieza_articulos, remove_URL, remove_html_markup
7
  from entrenamiento_modelo import term_document_matrix, tf_idf_score
8
  from resultados_consulta import resultados_consulta, detalles_resultados
9
  import tensorflow as tf
 
126
 
127
  for i in range(len(df_answer)):
128
 
129
+ text=remove_html_markup(df_answer.loc[i, "resumen"])
130
+ text=remove_URL(text)
131
+ st.write(f'Aqui {text}' )
132
+ inputs = tokenizer(query, text[0:512], return_tensors='tf')
133
  outputs = qa_model(input_ids=inputs['input_ids'], attention_mask=inputs['attention_mask'])
134
  answer_start_index = int(tf.math.argmax(outputs.start_logits, axis=-1)[0])
135
  answer_end_index = int(tf.math.argmax(outputs.end_logits, axis=-1)[0])