add basic streamlit mock

to do = move input to pd.df

streamlit/requirements.txt

@@ -0,0 +1,11 @@
+Cython>=0.28.5
+numpy>=1.22.0
+scikit-learn==0.22.2.post1
+pandas>=1.3.0
+joblib
+tqdm
+nltk
+tensorflow>=2.7.2,<3
+setuptools>=65.5.1
+pranaam
+streamlit

streamlit/streamlit_app.py

@@ -0,0 +1,40 @@
+import streamlit as st
+import pandas as pd
+from pranaam import pranaam
+import base64
+
+def download_file(df):
+    csv = df.to_csv(index=False)
+    b64 = base64.b64encode(csv.encode()).decode()
+    href = f'<a href="data:file/csv;base64,{b64}" download="results.csv">Download results</a>'
+    st.markdown(href, unsafe_allow_html=True)
+
+def app():
+    # Set app title
+    st.title("pranaam: predict religion based on name")
+
+    # Generic info.
+    st.write('Pranaam uses the Bihar Land Records data, plot-level land records, to build machine learning models that predict religion and caste from the name.')
+    st.write('[Github](https://github.com/appeler/pranaam)')
+
+    # Upload CSV file
+    uploaded_file = st.file_uploader("Choose a CSV file", type=["csv"])
+
+    # Load data
+    if uploaded_file is not None:
+        df = pd.read_csv(uploaded_file)
+        st.write("Data loaded successfully!")
+    else:
+        st.stop()
+
+    lname_col = st.selectbox("Select column with last name", df.columns)
+    lang = st.selectbox("Select the language", ["eng", "hin"])
+    function = sidebar_options[selected_function]
+    if st.button('Run'):
+        transformed_df = pranaam.pred_rel(df, namecol=lname_col, lang = lang)
+        st.dataframe(transformed_df)
+        download_file(transformed_df)
+
+# Run the app
+if __name__ == "__main__":
+    app()