update

.gitignore

@@ -0,0 +1 @@
+__pycache__

app.py

@@ -1,5 +1,15 @@
 import streamlit as st
 from st_pages import Page, show_pages, add_page_title
+from transformers import pipeline
+from datasets import load_dataset
+
+
+class TextClassifier:
+    bertPipe = pipeline( model="nlp-pw/test-model-3")
+
+
+class AnalysisData:
+    ds = load_dataset("nlp-pw/Disaster-Tweets-Normalized", revision="main", split='train')
 
 
 def main():
@@ -8,7 +18,7 @@ def main():
         page_title="NLP - Project",
         page_icon="🤖",
         )
-    add_page_title()
+    add_page_title("NLP - Project")
     show_pages(
         [
             Page("pages/introduction.py", "Intro", "😊"),
@@ -19,3 +29,4 @@ def main():
 
 if __name__ == "__main__":
     main()
+    

pages/data_analysis.py

@@ -2,22 +2,27 @@ import streamlit as st
 
 from matplotlib import pyplot as plt
 from wordcloud import WordCloud, STOPWORDS 
-from datasets import load_dataset
 import numpy as np
 
+from app import AnalysisData
 
-ds = load_dataset("rajteer/Natural_disaster_tweets", revision="main", split='train')
 
-text_data = ' '.join([row['tweet_text'] for row in ds])
+df = AnalysisData.ds.to_pandas(batched=False)
+disaster_types = df['disaster_type'].unique()
 
+text_data = {
+    disaster: ' '.join(df[df['disaster_type'] == disaster]['tweet_text'])
+     for disaster in disaster_types
+}
 
-wordcloud = WordCloud(width=800, height=400).generate(text_data)
+for disaster in disaster_types:
+    st.subheader(disaster + ' ' + 'Word Cloud')
+    wordcloud = WordCloud(width=800, height=400).generate(text_data[disaster])
+    fig, ax = plt.subplots(figsize=(10, 5))
+    ax.imshow(wordcloud, interpolation='bilinear')
+    ax.axis('off')
 
-fig, ax = plt.subplots(figsize=(10, 5))
-ax.imshow(wordcloud, interpolation='bilinear')
-ax.axis('off')
-
-st.pyplot(fig)
+    st.pyplot(fig)
 
 
 # DataSet links

pages/run_models.py

@@ -1,9 +1,9 @@
 import streamlit as st
 from transformers import pipeline
+from app import TextClassifier
 
-pipe = pipeline( model="nlp-pw/test-model-2")
 text = st.text_area("Enter your text here 😃")
 
 if text:
-    out = pipe(text)
+    out = TextClassifier.bertPipe(text)
     st.json(out)