Added app files

app.py

@@ -0,0 +1,104 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+# from scipy.special import softmax
+# import os
+from utils import run_sentiment_analysis, preprocess
+from transformers import AutoTokenizer, AutoConfig,AutoModelForSequenceClassification
+import os
+import time
+
+# Requirements
+model_path = "bright1/fine-tuned-distilbert-base-uncased"
+tokenizer = AutoTokenizer.from_pretrained(model_path)
+config = AutoConfig.from_pretrained(model_path)
+model = AutoModelForSequenceClassification.from_pretrained(model_path)
+
+# dark_theme = set_theme()
+
+
+st.set_page_config(
+    page_title="Tweet Analyzer",
+    page_icon="🤖",
+    initial_sidebar_state="expanded",
+    menu_items={
+        'About': "# This is a header. This is an *extremely* cool app!"
+    }
+)  
+
+
+my_expander  = st.container()
+
+
+# st.sidebar.selectbox('Menu', ['About', 'Model'])
+with my_expander:
+    
+    st.markdown("""
+        <style>
+        h1 {
+            text-align: center;
+        }
+        </style>
+        """, unsafe_allow_html=True)
+    st.title(':green[Covid-19 Vaccines Tweets Analyzer]')
+    st.sidebar.markdown("""
+    ## Demo App
+
+    This app analyzes your tweets on covid vaccines and classifies them us Neutral, Negative or Positive
+    """)        
+    # my_expander.write('Container')
+    # create a three column layout
+
+    col1, col2, col3 = st.columns((1.6, 1,0.3))
+    # col2.markdown("""
+    #         <p style= font-color:red>
+    #             Results from Analyzer
+    #         </p>
+    # """,unsafe_allow_html=True)
+    st.markdown("""
+        <style>
+        p {
+            font-color: blue;
+        }
+        </style>
+        """, unsafe_allow_html=True)
+    tweet = col1.text_area('Tweets to analyze',height=200, max_chars=520, placeholder='Write your Tweets here')
+    colA, colb, colc, cold = st.columns(4)
+    clear_button = colA.button(label='Clear', type='secondary', use_container_width=True)
+    submit_button = colb.button(label='Submit', type='primary', use_container_width=True)
+    empty_container = col2.container()
+    empty_container.text("Results from Analyzer")
+    empty_container2 = col3.container()
+    empty_container2.text('Scores')
+    text = preprocess(tweet)
+    results = run_sentiment_analysis(text=text, model=model, tokenizer=tokenizer)
+    if submit_button:
+        success_message = st.success('Success', icon="✅")
+        
+        with empty_container:
+            
+            neutral = st.progress(value=results['Neutral'], text='Neutral',)
+            negative = st.progress(value=results['Negative'], text='Negative')
+            positive = st.progress(value=results['Positive'], text='Positive')
+        with empty_container2:
+            st.markdown(
+                """
+            <style>
+            [data-testid="stMetricValue"] {
+                font-size: 20px;
+            }
+            </style>
+            """,
+                unsafe_allow_html=True,
+            )
+            neutral_score = st.metric(label='Score', value=round(results['Neutral'], 4), label_visibility='collapsed')
+            negative_score = st.metric(label='Score', value=round(results['Negative'], 4), label_visibility='collapsed')
+            positive_score = st.metric(label='Score', value=round(results['Positive'], 4), label_visibility='collapsed')
+            time.sleep(5)
+            success_message.empty()
+    interpret_button = col2.button(label='Interpret',type='secondary', use_container_width=True)
+
+
+# st.help()
+    # create a date input to receive date
+

utils.py

@@ -0,0 +1,54 @@
+import numpy as np
+import pandas as pd
+from transformers import AutoTokenizer, AutoConfig,AutoModelForSequenceClassification
+from scipy.special import softmax
+import os
+
+# Requirements
+# model_path = "bright1/fine-tuned-distilbert-base-uncased"
+# tokenizer = AutoTokenizer.from_pretrained(model_path)
+# config = AutoConfig.from_pretrained(model_path)
+# model = AutoModelForSequenceClassification.from_pretrained(model_path)
+
+
+
+def check_csv(csv_file, data):
+    if os.path.isfile(csv_file):
+        data.to_csv(csv_file, mode='a', header=False, index=False, encoding='utf-8')
+    else:
+        history = data.copy()
+        history.to_csv(csv_file, index=False)
+
+#Preprocess text
+def preprocess(text):
+    new_text = []
+    for t in text.split(" "):
+        t = "@user" if t.startswith("@") and len(t) > 1 else t
+        t = "http" if t.startswith("http") else t
+        print(t)
+        new_text.append(t)
+    print(new_text)
+
+    return " ".join(new_text)
+
+#Process the input and return prediction
+def run_sentiment_analysis(text, tokenizer, model):
+    # save_text =  {'tweet': text}
+    encoded_input = tokenizer(text, return_tensors = "pt") # for PyTorch-based models
+    output = model(**encoded_input)
+    scores_ = output[0][0].detach().numpy()
+    scores_ = softmax(scores_)
+
+    # Format output dict of scores
+    labels = ["Negative", "Neutral", "Positive"]
+    scores = {l:float(s) for (l,s) in zip(labels, scores_) }
+    # save_text.update(scores)
+    # user_data = {key: [value] for key,value in save_text.items()}
+    # data = pd.DataFrame(user_data,)
+    # check_csv('history.csv', data)
+    # hist_df = pd.read_csv('history.csv')
+    return scores
+
+
+    
+