Added Excel import functionality

app.py

@@ -1,30 +1,9 @@
 import streamlit as st
 import pandas as pd
-from streamlit_pandas_profiling import st_profile_report
 from pathlib import Path
 from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
 
-st.set_page_config(page_title="Francesco Daimon Fernicola", page_icon=":milky_way:", layout="wide")
-
-with st.container():
-    st.subheader("Hello, and welcome to my official webpage! I am Daimon :alien:")
-    st.title("PhD Candidate in Machine Translation / Translator / Mountain enthusiast")
-    st.write("I am passionate about finding new ways to effectively use and understand Machine Translation and effectively evaluating its quality.")
-    st.write("""
-    [Github](https://github.com/FrancescoFernicola)
-    [Unibo](https://www.unibo.it/sitoweb/francesco.fernicola2)
-    [LinkedIn](https://www.linkedin.com/in/francesco-fernicola-69a0771b7/?locale=en_US)
-    [Twitter](https://twitter.com/FrancescoDaimon)
-    """)
-
-from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
-st.subheader("MBART-50 Translator")
-
-source = "In the beginning the Universe was created. This has made a lot of people very angry and been widely regarded as a bad move."
-target = ""
-model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
-tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
-
+st.set_page_config(page_title="Translation Demo", page_icon=":milky_way:", layout="wide")
 
 
 def get_translation(src_code, trg_code, src):
@@ -39,6 +18,25 @@ def get_translation(src_code, trg_code, src):
     
     return trg
 
+
+def open_input(the_file):
+    
+    if the_file.name.endswith('.tsv'):
+        parsed = pd.read_csv(the_file, sep="\t")
+    elif the_file.name.endswith('.xlsx'):
+        parsed = pd.read_excel(the_file)
+
+    return parsed
+
+
+st.subheader("MBART-50 Translator")
+
+source = "In the beginning the Universe was created. This has made a lot of people very angry and been widely regarded as a bad move."
+target = ""
+model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
+tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
+
+
 valid_languages = ['de_DE', 'en_XX', 'it_IT']
 
 
@@ -76,56 +74,53 @@ with st.form("my_form"):
             st.write("Please enter the source text, source language and target language.")
 
 
-st.subheader('Input TSV')
+st.subheader('Input Excel/TSV')
 uploaded_file = st.file_uploader("Choose a file")
 done = False
 
 
-
 if uploaded_file is not None:
     valid_languages_col = (lang for lang in valid_languages)
     valid_languages_col_trg = (lang for lang in valid_languages)
-    if uploaded_file.name.endswith('.tsv'):
-        data = pd.read_csv(uploaded_file, sep="\t")
-        st.subheader("DataFrame")
-        st.write(data)
-        st.write(data.describe())
-        columns = (col for col in data.columns)
-        src_col = st.selectbox(
-        'Select the column to translate:',
-        columns,
-        )
-        if src_col:
-            col_src_lang = st.selectbox(
-        'Source language:',
-        valid_languages_col,
+    data = open_input(uploaded_file)
+    st.subheader("DataFrame")
+    st.write(data)
+    st.write(data.describe())
+    
+    columns = (col for col in data.columns)
+    src_col = st.selectbox(
+    'Select the column to translate:',
+    columns,
+    )
+    
+    if src_col:
+        col_src_lang = st.selectbox(
+            'Source language:',
+            valid_languages_col,
         )
-            col_trg_lang = st.selectbox(
-        'Target language:',
-        valid_languages_col_trg,
+        col_trg_lang = st.selectbox(
+            'Target language:',
+            valid_languages_col_trg,
         )
-            submitted_cols = st.button("Translate column")
-            if submitted_cols:
-                translated_data = []
-                new_df = data
-                for text in data[src_col]:
-                    if len(text) > 0 and col_src_lang in valid_languages and col_trg_lang in valid_languages:
-                        with st.spinner("Translating..."):
-                            try:
-                                target_text = get_translation(col_src_lang, col_trg_lang, text)[0]
-                                translated_data.append(target_text)
+        submitted_cols = st.button("Translate column")
+        
+        if submitted_cols:
+            translated_data = []
+            new_df = data
+            for text in data[src_col]:
+                if len(text) > 0 and col_src_lang in valid_languages and col_trg_lang in valid_languages:
+                    with st.spinner("Translating..."):
+                        try:
+                            target_text = get_translation(col_src_lang, col_trg_lang, text)[0]
+                            translated_data.append(target_text)
                             except:
                                 st.subheader("Translation failed :sad:")
                                 break
+                else:
+                    st.write("Please enter the source text, source language and target language.")
                     
-                    else:
-                        st.write("Please enter the source text, source language and target language.")
-                
-                new_df[src_col] = translated_data
-                done = True
-        
-    else:
-        data = pd.read_csv(uploaded_file)
+            new_df[src_col] = translated_data
+            done = True
 
 if done:
     st.subheader("Translated DataFrame")