Streamlit application files added

LSTM_predictions-final.csv

@@ -0,0 +1,13 @@
+Date,Prediction
+2022-01-31,321870080.0
+2022-02-28,299647800.0
+2022-03-31,339766940.0
+2022-04-30,334998460.0
+2022-05-31,350785380.0
+2022-06-30,342564500.0
+2022-07-31,356114430.0
+2022-08-31,357524860.0
+2022-09-30,346851840.0
+2022-10-31,358977100.0
+2022-11-30,347742500.0
+2022-12-31,359557760.0

Prophet_predictions.csv

@@ -0,0 +1,13 @@
+Date,Prediction
+2022-01-31,317881610.0590188
+2022-02-28,293104833.38518673
+2022-03-31,331135132.92652655
+2022-04-30,327193033.51301265
+2022-05-31,344923058.9539489
+2022-06-30,340493712.42480046
+2022-07-31,358751759.0707788
+2022-08-31,365728493.44860756
+2022-09-30,360607364.1991499
+2022-10-31,379558692.5370447
+2022-11-30,373912496.74845016
+2022-12-31,393332601.9413679

RNN_predictions-final.csv

@@ -0,0 +1,13 @@
+Date,Prediction
+2022-01-31,323736300.0
+2022-02-28,293960030.0
+2022-03-31,326672160.0
+2022-04-30,317307230.0
+2022-05-31,327673340.0
+2022-06-30,317215900.0
+2022-07-31,327947260.0
+2022-08-31,327757340.0
+2022-09-30,317391870.0
+2022-10-31,327808030.0
+2022-11-30,317342240.0
+2022-12-31,327862300.0

app.py

@@ -0,0 +1,56 @@
+import streamlit as st
+import pandas as pd
+import plotly.express as px
+
+# Function to load predictions
+@st.cache_data
+def load_predictions(filename):
+    data = pd.read_csv(filename)
+    data['Date'] = pd.to_datetime(data['Date']).dt.strftime('%Y-%m')
+    return data
+
+# Title and Introduction with icons
+st.title('💹 2022 Receipt Count Predictions')
+st.markdown("## Explore the Receipt Count Predictions for the year 2022 📈")
+
+# Sidebar for model selection
+st.sidebar.title("🔍 Model Selection")
+model_option = st.sidebar.selectbox(
+    'Choose a model:',
+    ('LSTM', 'RNN', 'Prophet')
+)
+
+# Button to load predictions
+if st.sidebar.button('Get Predictions'):
+    filename_map = {'LSTM': 'results/LSTM_predictions-final.csv', 
+                    'RNN': 'results/rnn_predictions-final.csv', 
+                    'Prophet': 'results/Prophet_predictions.csv'}
+
+    predictions = load_predictions(filename_map[model_option])
+
+    # Display Data Table
+    st.markdown("### Predicted Receipt Counts")
+    st.dataframe(predictions)
+
+    # Plotting Predictions and fixing the error
+    st.markdown("### Interactive Prediction Graph")
+    fig = px.line(predictions, x='Date', y='Prediction', 
+                  title=f'Monthly Predicted Receipt Counts for 2022 ({model_option} Model)', 
+                  markers=True)
+    st.plotly_chart(fig, use_container_width=True)
+
+    # Download links
+    st.markdown('## Download Results')
+    st.download_button(
+        label="Download Data as CSV",
+        data=predictions.to_csv(index=False).encode('utf-8'),
+        file_name='predictions.csv',
+        mime='text/csv',
+    )
+
+    st.download_button(
+        label="Download Plot as Image",
+        data=fig.to_image(format="png"),
+        file_name='predictions_plot.png',
+        mime='image/png',
+    )

requirements.txt

@@ -0,0 +1,8 @@
+pandas
+numpy
+matplotlib
+plotly
+streamlit
+scikit-learn
+prophet
+kaleido