Update app.py

app.py

@@ -240,7 +240,7 @@ def transfer_learning_forecasting():
 
         columns = df.columns.tolist()
         ds_col = st.selectbox("Select Date/Time column", options=columns, index=columns.index('ds') if 'ds' in columns else 0)
-        target_columns = [col for col in columns if (col != ds_col) and (col != ds_col)]
+        target_columns = [col for col in columns if (col != ds_col) and (col != 'unique_id')]
         y_col = st.selectbox("Select Target column", options=target_columns, index=0)
 
         st.session_state.ds_col = ds_col
@@ -342,7 +342,7 @@ def dynamic_forecasting():
 
         columns = df.columns.tolist()
         ds_col = st.selectbox("Select Date/Time column", options=columns, index=columns.index('ds') if 'ds' in columns else 0)
-        target_columns = [col for col in columns if (col != ds_col) and (col != ds_col)]
+        target_columns = [col for col in columns if (col != ds_col) and (col != 'unique_id')]
         y_col = st.selectbox("Select Target column", options=target_columns, index=0)
 
         st.session_state.ds_col = ds_col
@@ -398,7 +398,7 @@ def timegpt_fcst():
 
         columns = df.columns.tolist()
         ds_col = st.selectbox("Select Date/Time column", options=columns, index=columns.index('ds') if 'ds' in columns else 0)
-        target_columns = [col for col in columns if (col != ds_col) and (col != ds_col)]
+        target_columns = [col for col in columns if (col != ds_col) and (col != 'unique_id')]
         y_col = st.selectbox("Select Target column", options=target_columns, index=0)
         h = st.number_input("Forecast horizon", value=14)
 
@@ -410,19 +410,19 @@ def timegpt_fcst():
         df = df[['unique_id','ds','y']]
         
 
-        freq = determine_frequency(df)
+    freq = determine_frequency(df)
 
-        df = df.drop_duplicates(subset=['ds']).reset_index(drop=True)
-        
-        # st.write(df)
-        if st.sidebar.button("Submit"):
-            forecast_df = nixtla_client.forecast(
-                df=df,
-                h=h,
-                freq=freq,
-                level=[90]
-            )
-            st.session_state.forecast_df = forecast_df
+    df = df.drop_duplicates(subset=['ds']).reset_index(drop=True)
+    
+    # st.write(df)
+    if st.sidebar.button("Submit"):
+        forecast_df = nixtla_client.forecast(
+            df=df,
+            h=h,
+            freq=freq,
+            level=[90]
+        )
+        st.session_state.forecast_df = forecast_df
 
     if 'forecast_df' in st.session_state:
         forecast_df = st.session_state.forecast_df
@@ -465,7 +465,7 @@ def timegpt_anom():
 
         columns = df.columns.tolist()
         ds_col = st.selectbox("Select Date/Time column", options=columns, index=columns.index('ds') if 'ds' in columns else 0)
-        target_columns = [col for col in columns if (col != ds_col) and (col != ds_col)]
+        target_columns = [col for col in columns if (col != ds_col) and (col != 'unique_id')]
         y_col = st.selectbox("Select Target column", options=target_columns, index=0)
 
         df = df.rename(columns={ds_col: 'ds', y_col: 'y'})
@@ -474,16 +474,16 @@ def timegpt_anom():
         df['unique_id']=id_col
         df = df[['unique_id','ds','y']]
 
-        freq = determine_frequency(df)
+    freq = determine_frequency(df)
 
-        df = df.drop_duplicates(subset=['ds']).reset_index(drop=True)
-        if st.sidebar.button("Submit"):
-            anom_df = nixtla_client.detect_anomalies(
-                df=df,
-                freq=freq,
-                level=90
-            )
-            st.session_state.anom_df = anom_df
+    df = df.drop_duplicates(subset=['ds']).reset_index(drop=True)
+    if st.sidebar.button("Submit"):
+        anom_df = nixtla_client.detect_anomalies(
+            df=df,
+            freq=freq,
+            level=90
+        )
+        st.session_state.anom_df = anom_df
 
     if 'anom_df' in st.session_state:
         anom_df = st.session_state.anom_df