Fix app

app.py

@@ -18,7 +18,7 @@ st.set_page_config(
 )
 
 # Preprocessing
-@st.cache(show_spinner=False)
+@st.cache_data
 def merge(B, C, A):
   i = j = k = 0
 
@@ -53,6 +53,7 @@ def merge(B, C, A):
 
   return A
 
+@st.cache_data
 def merge_sort(dataframe):
   if len(dataframe) > 1:
       center = len(dataframe) // 2
@@ -66,6 +67,7 @@ def merge_sort(dataframe):
   else:
       return dataframe
 
+@st.cache_data
 def drop (dataframe):
   def get_columns_containing(dataframe, substrings):
     return [col for col in dataframe.columns if any(substring.lower() in col.lower() for substring in substrings)]
@@ -76,6 +78,7 @@ def drop (dataframe):
     
   return dataframe
 
+@st.cache_data
 def date_format(dataframe):
   for i, d, s in dataframe.itertuples():
     dataframe['Date'][i] = dataframe['Date'][i].strip()
@@ -86,6 +89,7 @@ def date_format(dataframe):
 
   return dataframe
 
+@st.cache_data
 def group_to_three(dataframe):
   dataframe['Date'] = pd.to_datetime(dataframe['Date'])
   dataframe = dataframe.groupby([pd.Grouper(key='Date', freq='3D')])['Sales'].mean().round(2)
@@ -94,7 +98,7 @@ def group_to_three(dataframe):
   return dataframe
 
 # SARIMAX Model
-@st.cache(show_spinner=False)
+@st.cache_data
 def train_test(dataframe):
   n = round(len(dataframe) * 0.2)
   training_y = dataframe.iloc[:-n,0]
@@ -105,7 +109,7 @@ def train_test(dataframe):
   future_X = dataframe.iloc[0:,1:]
   return (training_y, test_y, test_y_series, training_X, test_X, future_X)
 
-@st.cache(show_spinner=False)
+@st.cache_data
 def model_fitting(dataframe, Exo):
     futureModel = pm.auto_arima(dataframe['Sales'], X=Exo, start_p=1, start_q=1,
                          test='adf',min_p=1,min_q=1,
@@ -119,7 +123,7 @@ def model_fitting(dataframe, Exo):
     model = futureModel
     return model
 
-@st.cache(show_spinner=False)
+@st.cache_data
 def test_fitting(dataframe, Exo, trainY):
     trainTestModel = auto_arima(X = Exo, y = trainY, start_p=1, start_q=1,
                            test='adf',min_p=1,min_q=1,
@@ -133,7 +137,7 @@ def test_fitting(dataframe, Exo, trainY):
     model = trainTestModel
     return model
 
-@st.cache(show_spinner=False)
+@st.cache_data
 def forecast_accuracy(forecast, actual):
     mape = np.mean(np.abs(forecast - actual)/np.abs(actual)).round(4)  # MAPE
     rmse = (np.mean((forecast - actual)**2)**.5).round(2)  # RMSE
@@ -145,7 +149,7 @@ def forecast_accuracy(forecast, actual):
     minmax = 1 - np.mean(mins/maxs)             # minmax
     return({'mape':mape, 'rmse':rmse, 'corr':corr, 'min-max':minmax})
 
-@st.cache(show_spinner=False)
+@st.cache_data
 def sales_growth(dataframe, fittedValues):
     sales_growth = fittedValues.to_frame()
     sales_growth = sales_growth.reset_index()
@@ -170,6 +174,7 @@ model_name = "google/tapas-large-finetuned-wtq"
 tokenizer = TapasTokenizer.from_pretrained(model_name)
 model = TapasForQuestionAnswering.from_pretrained(model_name, local_files_only=False)
 
+@st.cache_resource
 def load_tapas_model(model, tokenizer):
   pipe = pipeline("table-question-answering", model=model, tokenizer=tokenizer)
   return pipe