Spaces:
Sleeping
Sleeping
File size: 2,704 Bytes
818a838 4b7389c e7e6566 818a838 4b7389c 818a838 4b7389c bcfdcee 818a838 6323541 0ded43b 6323541 bcfdcee 4081ce7 6323541 4081ce7 bcfdcee 4b7389c 4081ce7 4b7389c 0ded43b 6323541 4b7389c bcfdcee 818a838 0ded43b 4b7389c 818a838 |
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 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 |
import streamlit as st
import wna_googlenews as wna
import pandas as pd
from transformers import pipeline
st.set_page_config(layout="wide")
st.title("WNA Google News App")
st.subheader("Search for News and classify the headlines with sentiment analysis")
query = st.text_input("Enter Query")
models = [
"j-hartmann/emotion-english-distilroberta-base",
"SamLowe/roberta-base-go_emotions"
# "distilbert/distilbert-base-uncased-finetuned-sst-2-english"
]
settings = {
"langregion": "en/US",
"period": "1d",
"model": models[0],
"number_of_pages": 5
}
with st.sidebar:
st.title("Settings")
# add language and country parameters
st.header("Language and Country")
settings["langregion"] = st.selectbox("Select Language", ["en/US", "fr/FR"])
# input field for number of pages
st.header("Number of Pages")
settings["number_of_pages"] = st.number_input("Enter Number of Pages", min_value=1, max_value=10)
settings["region"] = settings["langregion"].split("/")[0]
settings["lang"] = settings["langregion"].split("/")[1]
# add period parameter
st.header("Period")
settings["period"] = st.selectbox("Select Period", ["1d", "7d", "30d"])
# Add models parameters
st.header("Models")
settings["model"] = st.selectbox("Select Model", models)
if st.button("Search"):
# display a loading progress
with st.spinner("Loading last news ..."):
classifier = pipeline(task="text-classification", model=settings["model"], top_k=None)
df = wna.get_news(settings, query)
with st.spinner("Processing received news ..."):
# st.dataframe(df)
# get each title colums
sentences = df["title"]
# convert into array
sentences = sentences.tolist()
# st.write(sentences)
# create new dataframe
df = pd.DataFrame(columns=["sentence", "best","second"])
# loop on each sentence and call classifier
for sentence in sentences:
cur_sentence = sentence
model_outputs = classifier(sentence)
cur_result = model_outputs[0]
#st.write(cur_result)
# get label 1
label = cur_result[0]['label']
score = cur_result[0]['score']
percentage = round(score * 100, 2)
str1 = label + " " + str(percentage)
# get label 2
label = cur_result[1]['label']
score = cur_result[1]['score']
percentage = round(score * 100, 2)
str2 = label + " " + str(percentage)
# insert cur_sentence and cur_result into dataframe
df.loc[len(df.index)] = [cur_sentence, str1, str2]
# write info on the output
st.write("Number of sentences:", len(df))
st.write("Language:", settings["lang"], "Country:", settings["region"])
st.dataframe(df)
|