ContentChecker / app.py
abischoff's picture
Create app.py
ba5cd5a verified
Raw History Blame Contribute Delete
1.99 kB
import streamlit as st
import numpy as np
from transformers import Trainer, AutoTokenizer, XLMRobertaForSequenceClassification
import torch
import pickle
import re
def checker(content):
sentences = re.split(r"(?<=[!.;?:])\s+(?![a-zà-öø-ÿα-ωа-яא-תء-ي])", content.decode('utf-8'))
model_name = "abischoff/test-trainer"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = XLMRobertaForSequenceClassification.from_pretrained(model_name,num_labels=1)
sum = 0.0
bias = -0.0666
for dic in sentences:
inputs = tokenizer(dic.strip(), return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits.flatten()[0].numpy()+bias
logits = np.clip(logits,0,1)
print(logits)
sum += logits
return sum / len(content)
def main():
# Setze den Titel und eine moderne Überschrift
st.title("Checker")
st.markdown("""
<style>
.reportview-container {
background-color: #f4f4f9;
}
h1 {
color: #4F8BF9;
}
footer {
visibility: hidden;
}
</style>
""", unsafe_allow_html=True)
# Erstelle das Drag&Drop Upload-Feld für alle Dateiformate
uploaded_file = st.file_uploader("Drag & Drop your file here", type=None)
# Prüfen, ob eine Datei hochgeladen wurde
if uploaded_file is not None:
# Anzeige des Dateinamens
st.write(f"File uploaded: {uploaded_file.name}")
# Wenn die Datei eine Text- oder CSV-Datei ist, zeige den Inhalt an
if uploaded_file.name.endswith('.txt') or uploaded_file.name.endswith('.csv'):
content = uploaded_file.read().decode("utf-8")
st.text_area("File Content", content, height=300)
# Wenn andere Dateiformate hochgeladen werden, zeige eine Nachricht an
else:
st.write(f"Unsupported file type: {uploaded_file.name.split('.')[-1]}")
else:
st.write("Please upload a file.")