Duplicate from JulianHame/toxicity-classifier-app

.gitattributes

@@ -0,0 +1,3 @@
+toxicity_model.h5 filter=lfs diff=lfs merge=lfs -text
+train\[1\].csv filter=lfs diff=lfs merge=lfs -text
+train.csv filter=lfs diff=lfs merge=lfs -text

.github/workflows/sync_to_huggingface.yml

@@ -0,0 +1,20 @@
+name: Sync to Hugging Face hub
+on:
+  push:
+    branches: [main]
+
+  # to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+jobs:
+  sync-to-hub:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Push to hub
+        env:
+          HF_TOKEN: ${{ secrets.HF_TOKEN }}
+        run: git push --force https://JulianHame:$HF_TOKEN@huggingface.co/spaces/JulianHame/streamlit-application main

README.md

@@ -0,0 +1,14 @@
+---
+title: Toxicity Classifier App
+emoji: 🔥
+colorFrom: red
+colorTo: indigo
+sdk: streamlit
+sdk_version: 1.17.0
+app_file: app.py
+pinned: false
+duplicated_from: JulianHame/toxicity-classifier-app
+---
+
+# CS482-project-streamlit-application
+huggingface spaces deployment of a streamlit python application

app.py

@@ -0,0 +1,77 @@
+import streamlit as st
+from transformers import pipeline
+import tensorflow as tf
+import numpy as np
+import pandas as pd
+from tensorflow.keras.layers import TextVectorization
+from tensorflow import keras
+
+model = tf.keras.models.load_model('toxicity_model.h5')
+
+dataset = pd.read_csv('train.csv')
+comments = dataset['comment_text']
+
+vectorizer = TextVectorization(max_tokens = 2500000,
+                               output_sequence_length=1800,
+                               output_mode='int')
+
+vectorizer.adapt(comments.values)
+
+st.title('Toxicity Classifier')
+
+st.header('Write a message here:')
+text = st.text_area('The toxicity of the message will be evaluated.',
+                   value = "You're fucking ugly.")
+
+input_str = vectorizer(text)
+res = model.predict(np.expand_dims(input_str,0))
+classification = res[0].tolist()
+
+toxicity = classification[0]
+toxicity_severe = classification[1]
+obscene = classification[2]
+threat = classification[3]
+insult = classification[4]
+identity_hate = classification[5]
+
+highest_class = "Severe toxicity"
+highest_class_rating = toxicity_severe
+if(obscene > highest_class_rating):
+    highest_class = "Obscenity"
+    highest_class_rating = obscene
+if(threat > highest_class_rating):
+    highest_class = "Threat"
+    highest_class_rating = threat
+if(insult > highest_class_rating):
+    highest_class = "Insult"
+    highest_class_rating = insult
+if(identity_hate > highest_class_rating):
+    highest_class = "Identity hate"
+    highest_class_rating = identity_hate
+
+st.write("---")
+st.write("Overall toxicity rating: " +str(toxicity))
+st.write("---")
+st.write("Classifications:")
+if(toxicity_severe > 0.5):
+    st.write("Severely toxic - " +str(toxicity_severe))
+if(obscene > 0.5):
+    st.write("Obscene - " +str(obscene))
+if(threat > 0.5):
+    st.write("Threat - " +str(threat))
+if(insult > 0.5):
+    st.write("Insult - " +str(insult))
+if(identity_hate > 0.5):
+    st.write("Identity hate - " +str(identity_hate))
+st.write("---")
+st.write("Invalid classifications:")
+if(toxicity_severe <= 0.5):
+    st.write("Severely toxic - " +str(toxicity_severe))
+if(obscene <= 0.5):
+    st.write("Obscene - " +str(obscene))
+if(threat <= 0.5):
+    st.write("Threat - " +str(threat))
+if(insult <= 0.5):
+    st.write("Insult - " +str(insult))
+if(identity_hate <= 0.5):
+    st.write("Identity hate - " +str(identity_hate))

requirements.txt

@@ -0,0 +1,9 @@
+streamlit
+opencv-python-headless
+numpy
+easyocr
+Pillow
+torch
+transformers
+tensorflow
+pandas

toxicity_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b957463ca9befba2c69a5f49b3a6baa6e3a3974b6d40cc57fb6638cbb19d8fc
+size 77969544

train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd4084611bd27c939ba98e5e63bc3e5a2c1a4e99477dcba46c829e4c986c429d
+size 68802655