diff --git a/fastapi_backend/requirements.txt b/fastapi_backend/requirements.txt
index e790455..7937eea 100644
--- a/fastapi_backend/requirements.txt
+++ b/fastapi_backend/requirements.txt
@@ -1,17 +1,20 @@
 kobert-tokenizer @ git+https://github.com/SKTBrain/KoBERT.git#egg=kobert_tokenizer&subdirectory=kobert_hf
-sentencepiece
-streamlit
-scikit-learn
-pandas
-numpy
-pillow
-soundfile
-torch 
-torchvision
-transformers
-fastapi 
-python-multipart
-uvicorn 
-torchaudio
-librosa
-pydub
\ No newline at end of file
+sentencepiece==0.1.99
+gdown==4.7.1
+fastapi==0.104.0
+librosa==0.10.1
+openai==0.28.1
+openpyxl==3.1.2
+numpy==1.26.1
+streamlit==1.28.0
+scikit-learn==1.3.2
+soundfile==0.12.1
+torch==2.1.0
+torchvision==0.16.0
+torchaudio==2.1.0
+transformers==4.34.1
+pandas==2.1.2
+pydub==0.25.1
+python-multipart==0.0.6
+pillow==10.1.0
+uvicorn==0.23.2
\ No newline at end of file
diff --git a/fastapi_backend/server.py b/fastapi_backend/server.py
index 766eb84..8564789 100644
--- a/fastapi_backend/server.py
+++ b/fastapi_backend/server.py
@@ -403,7 +403,11 @@ async def time_train_endpoint(data_arranges:list[str],
     label_data = 'Label_Data'
 
     content = await file.read()
-    train_df = pd.read_csv(BytesIO(content), encoding='utf-8')
+    if file_ext == 'csv':
+        train_df = pd.read_csv(BytesIO(content), encoding='utf-8')
+    elif file_ext == 'xlsx':
+        train_df = pd.read_excel(BytesIO(content))
+
 
     train_df[date] = pd.to_datetime(train_df[date])
 
diff --git a/streamlit_frontend/requirements.txt b/streamlit_frontend/requirements.txt
index b94006c..981c411 100644
--- a/streamlit_frontend/requirements.txt
+++ b/streamlit_frontend/requirements.txt
@@ -1,6 +1,8 @@
-streamlit
-pandas
-pillow
-python-multipart
-streamlit-modal
-python-dotenv
\ No newline at end of file
+openpyxl==3.1.2
+openai==0.28.1
+streamlit==1.24.1
+streamlit-modal==0.1.0
+pandas==2.1.2
+pillow==9.5.0
+python-dotenv==1.0.0
+python-multipart==0.0.6
\ No newline at end of file
diff --git a/streamlit_frontend/time_series_forecasting.py b/streamlit_frontend/time_series_forecasting.py
index e1915a1..01faeba 100644
--- a/streamlit_frontend/time_series_forecasting.py
+++ b/streamlit_frontend/time_series_forecasting.py
@@ -11,6 +11,7 @@
 
 from dotenv import load_dotenv
 import os
+import openpyxl
 
 from PIL import Image
 
@@ -246,7 +247,8 @@ def main():
                 image = Image.open('timeseries_sample_data/timeseries_forecasting_sample_image.png')
                 st.image(image, caption=None)
 
-            uploaded_file = st.file_uploader(upload_train_csv, accept_multiple_files=False, type=['csv'])
+            #uploaded_file = st.file_uploader(upload_train_csv, accept_multiple_files=False, type=['csv'])
+            uploaded_file = st.file_uploader(upload_train_csv, accept_multiple_files=False, type=['csv', 'xlsx'])
 
             if uploaded_file or sample_start: 
                 
@@ -260,9 +262,12 @@ def main():
                     csv_files = {'file': csv_file_obj}
                 
                     if uploaded_file.name.split('.')[-1] == 'csv':
-                        train_df = pd.read_csv(uploaded_file, encoding='utf-8')
-                    elif uploaded_file.name.split('.')[-1] == 'excel':
-                        train_df = pd.read_excel(uploaded_file)
+                         train_df = pd.read_csv(uploaded_file, encoding='utf-8')
+                    elif uploaded_file.name.split('.')[-1] == 'xlsx':
+                         train_df = pd.read_excel(uploaded_file)
+
+                    file_ext = uploaded_file.name.split('.')[-1]
+
 
                 elif sample_start:
                     train_df = pd.read_csv('timeseries_sample_data/timeseries_forecasting_sample_data.csv', encoding='utf-8')
@@ -376,12 +381,13 @@ def main():
                             prev_window_size = int(window_size)
                             with st.spinner(training_model_spinner):
                                 response = requests.post(f"{BACKEND_URL}/time_train", files=csv_files, data={'data_arranges':list(data_arrange),
-                                                                                                                    'window_size':int(window_size),
-                                                                                                                    'horizon_factor':int(horizon_factor),
-                                                                                                                    'epoch':int(epoch),
-                                                                                                                    'learning_rate':float(learning_rate),
-                                                                                                                    'pred_col': int_col,
-                                                                                                                    'date': date})
+                                                                                                                                             'window_size':int(window_size),
+                                                                                                                                             'horizon_factor':int(horizon_factor),
+                                                                                                                                             'epoch':int(epoch),
+                                                                                                                                             'learning_rate':float(learning_rate),
+                                                                                                                                             'pred_col': int_col,
+                                                                                                                                             'date': date,
+                                                                                                                                             'file_ext':file_ext})
                                 
                                 if response.ok:
                                     res = response.json()