File size: 1,049 Bytes
33dbef6 29e4959 33dbef6 f2a478c 33dbef6 29e4959 f2a478c 9d3ef8a 29e4959 33dbef6 29e4959 8ac658c f936c34 33dbef6 |
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 |
import streamlit as st #Web App
from transformers import pipeline
from transformers import AutoTokenizer, TFAutoModelForSequenceClassification
#title
st.title("Sentiment Analysis")
def analyze(input, model):
return "This is a sample output"
# load my fine-tuned model
fine_tuned = None
#text insert
input = st.text_area("insert text to be analyzed", value="Nice to see you today.", height=None, max_chars=None, key=None, help=None, on_change=None, args=None, kwargs=None, placeholder=None, disabled=False, label_visibility="visible")
option = st.selectbox(
'Choose a transformer model:',
('Default', 'Fine-Tuned' , 'Custom'))
if option == 'Fine-Tuned':
model = TFAutoModelForSequenceClassification.from_pretrained(fine_tuned)
tokenizer = AutoTokenizer.from_pretrained(fine_tuned)
classifier = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
else:
classifier = pipeline('sentiment-analysis')
if st.button('Analyze'):
st.write(classifier(input))
else:
st.write('Excited to analyze!')
|