app.py

import streamlit as st
import transformers
from transformers import AutoTokenizer, AutoModelForTokenClassification 
from transformers import pipeline

st.title("This is a Machine Learning app!")





#Now, we work on text summarization (TS)

#Using Facebook’s pre-trained BART Large which was fine-tuned on CNN Daily mail.

# from transformers import pipeline

model_TS= "facebook/bart-large-cnn"

#Use the pipeline API to load the summarization model you want to use.

summarizer = pipeline("summarization", model=model_TS)


# max_length=st.text_input("This is meant for text summarization tasks. What is the maximum text length you want? 130 words?")
# min_length=st.text_input("This is meant for text summarization tasks. What is the minimum text length you want? 30 words")

ARTICLE=st.text_input("This is meant for text summarization tasks: Provide your text!!")
 #Now, apply model results



results = summarizer(ARTICLE, max_length=1000, min_length=500, do_sample=False)

#Now let’s see the results. 
st.header("Your results are below")


st.subheader("This is your result for the text summary task:")

st.write(results[0]["summary_text"])



# x1 = st.slider('Select a value')
# st.write(x1, 'squared is', x1 * x1)

# x2 = st.slider('Select a second value')
# st.write(x2, 'squared is', x2 * x2)






#Unused models

#-----------------------------------
# 1.NER
#Loading pretrained tokenizers and models: Named Entity recognition task
# tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")

# modeL_NER = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")

# #Now, create a pipeline using the above

# nlp = pipeline("ner", model=modeL_NER, tokenizer=tokenizer)

# text_NER=st.text_input("This is meant for Named Entity Recognition (NER) tasks: Provide your text!!")

#     #Now, apply model results

# ner_results = nlp(text_NER) 
# for item in ner_results: 
#     st.write(item)
#----------------------