# @app.callback output error message - dash.exceptions.InvalidCallbackReturnValue:

**URL:** <https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668>\
**Category:** Dash Python\
**Created:** [July 19, 2020, 7:02pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668 "2020-07-19T19:02:16Z")\
**Posts on this page:** 16\
**Page:** 1

<div class="post-metadata">

**Author:** ![PBK](https://avatars.discourse-cdn.com/v4/letter/p/b9e5f3/32.png) [@PBK](https://community.plotly.com/u/PBK)\
**Post date:** [July 19, 2020, 7:02pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/1 "2020-07-19T19:02:16Z")

</div>

I’m outputting a pandas dataframe from an @app.callback. I keep getting the following error message:

```auto
dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `df_portfolio.children`>`
            returned a value having type `DataFrame`
            which is not JSON serializable.

            The value in question is either the only value returned,
            or is in the top level of the returned list,

            and has string representation
            ` shares share price total cost
0 8 295.55 2364.40
1 95 15.65 1486.75
2 18 27.69 498.42
3 23 10.83 249.09
4 4 52.88 211.52
5 N/A N/A 189.82`

            In general, Dash properties can only be
            dash components, strings, dictionaries, numbers, None,
            or lists of those.

```

The traceback is as follows:

```auto
Traceback (most recent call last):
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\dash.py", line 997, in add_context
    jsonResponse = json.dumps(
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\json\ __init__.py", line 234, in dumps
    return cls(
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\_plotly_utils\utils.py", line 45, in encode
    encoded_o = super(PlotlyJSONEncoder, self).encode(o)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\json\encoder.py", line 199, in encode
    chunks = self.iterencode(o, _one_shot=True)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\json\encoder.py", line 257, in iterencode
    return _iterencode(o, 0)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\_plotly_utils\utils.py", line 115, in default
    return _json.JSONEncoder.default(self, obj)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\json\encoder.py", line 179, in default
    raise TypeError(f'Object of type {o. __class__. __name__ } '
TypeError: Object of type DataFrame is not JSON serializable

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\flask\app.py", line 1950, in full_dispatch_request
    rv = self.dispatch_request()
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\flask\app.py", line 1936, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\dash.py", line 1031, in dispatch
    response.set_data(func(*args, outputs_list=outputs_list))
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\dash.py", line 1001, in add_context
    _validate.fail_callback_output(output_value, output)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\_validate.py", line 251, in fail_callback_output
    _validate_value(val, index=i)
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\_validate.py", line 241, in _validate_value
    _raise_invalid(
  File "C:\Users\pkopp\AppData\Local\Programs\Python\Python38-32\Lib\site-packages\dash\_validate.py", line 182, in _raise_invalid
    raise exceptions.InvalidCallbackReturnValue(
dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `df_portfolio.children`>`
            returned a value having type `DataFrame`
            which is not JSON serializable.

```

Why can’t I output a pandas dataframe?

Any help would be appreciated.

---

<div class="post-metadata">

**Author:** ![PBK](https://avatars.discourse-cdn.com/v4/letter/p/b9e5f3/32.png) [@PBK](https://community.plotly.com/u/PBK)\
**Post date:** [July 19, 2020, 7:19pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/2 "2020-07-19T19:19:57Z")

</div>

My code is the following:

```auto
def sample_portfolio(app, df_portfolio):
   @app.callback(
      Output('df_portfolio', 'data'),
      [Input('df_portfolio', 'value'),
      Input('my-age', 'value'),
      Input('risk-tolerance-slider', 'value'),
      Input('my-amount', 'value')]
   )

   def portfolio(df_portfolio, age, risk_tolerance, account_bal):
  
      sec_list = ["VOO", "PRULX", "VTIAX", "PFORX", "FDHY", "SPY"]
      sec_val_list = []
      # For some reason when typing in a new value for age or account_bal we get a Nonetype error.
      # This prevents the error.
      if age is None or account_bal is None:
         return [''] * len(sec_list)

      conn = http.client.HTTPSConnection("ftlightning.fasttrack.net")
      conn.request("GET", "/v1/auth/login?account=300724&pass=1243OEIL&appid=F075C6E1-759C-4009-9B47-5FE284F31F55")

      res = conn.getresponse()
      data = json.loads(res.read())

      appid = data['appid']
      token = data['token']

      headers = {'appid': appid,
                  'token': token}

      for sec in sec_list:
         conn.request("GET", "/v1/data/" + sec + "/divadjprices", headers=headers)
         res = conn.getresponse()
         data = res.read()
         dic = json.loads(data)
         sec_val_list.append(dic["prices"][-1])

      # Determine base portfolio from age
      if (age >= 18 and age <= 22):
         portolio = reference.base_portfolio[0]
      elif (age >= 23 and age <= 27):
         portfolio = reference.base_portfolio[1]
      elif (age >= 28 and age <= 32):
         portfolio = reference.base_portfolio[2]
      elif (age >= 33 and age <= 37):
         portfolio = reference.base_portfolio[3]
      elif (age >= 38 and age <= 42):
         portfolio = reference.base_portfolio[4]
      elif (age >= 43 and age <= 47):
         portfolio = reference.base_portfolio[5]
      elif (age >= 48 and age <= 52):
         portfolio = reference.base_portfolio[6]
      elif (age >= 53 and age <= 57):
         portfolio = reference.base_portfolio[7]
      elif (age >= 58 and age <= 62):
         portfolio = reference.base_portfolio[8]
      else:
         portfolio = reference.base_portfolio[9]

      # initialize adj_portfolio
      adj_portfolio = [0] * len(sec_list)
    
      # Adjust for risk tolerance.
      if risk_tolerance < 5:
         for i in range(len(adj_portfolio) - 1):
            if i <= 2:
               adj_portfolio[i] = portfolio[i] + (portfolio[i] * (-reference.adj * (5 - risk_tolerance)))
            else:
               adj_portfolio[i] = portfolio[i] + (portfolio[i] * (reference.adj * (5 - risk_tolerance)))
      elif risk_tolerance > 5:
         for i in range(len(adj_portfolio) - 1):
            if i <= 1:
               adj_portfolio[i] = portfolio[i] + (portfolio[i] * (reference.adj * (risk_tolerance - 5)))
            else:
               adj_portfolio[i] = portfolio[i] + (portfolio[i] * (-reference.adj * (risk_tolerance - 5)))
      else: # risk_tolerance = 5
         adj_portfolio = portfolio

      # Number of shares of each security.
      dom_stock = int((account_bal * adj_portfolio[0]/100) / sec_val_list[0])
      gov_bond = int((account_bal * adj_portfolio[1]/100) / sec_val_list[1])
      int_stock = int((account_bal * adj_portfolio[2]/100) / sec_val_list[2])
      int_bond = int((account_bal * adj_portfolio[3]/100) / sec_val_list[3])
      corp_bond = int(account_bal * adj_portfolio[4]/100 / sec_val_list[4])
      cash = account_bal - ((dom_stock * sec_val_list[0]) + (gov_bond * sec_val_list[1]) + 
         (int_stock * sec_val_list[2]) + (int_bond * sec_val_list[3]) + (corp_bond * sec_val_list[4]))

      test = [[dom_stock, sec_val_list[0], dom_stock * sec_val_list[0]], 
            [gov_bond, sec_val_list[1], gov_bond * sec_val_list[1]], 
            [int_stock, sec_val_list[2], int_stock * sec_val_list[2]], 
            [int_bond, sec_val_list[3], int_bond * sec_val_list[3]],
            [corp_bond, sec_val_list[4], corp_bond * sec_val_list[4]],
            ['N/A', 'N/A ', cash]]

      portfolio = pd.DataFrame(test, columns = ['shares', 'share price', 'total cost'])
      return portfolio

```

---

<div class="post-metadata">

**Author:** ![chriddyp](https://sea2.discourse-cdn.com/flex024/user_avatar/community.plotly.com/chriddyp/32/16855_2.png) [@chriddyp](https://community.plotly.com/u/chriddyp)\
**Post date:** [July 19, 2020, 11:36pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/3 "2020-07-19T23:36:25Z")

</div>

try return portfolio.to\_dict(‘records’). that is, convert the data frame into a data structure that can be serialized as json

---

<div class="post-metadata">

**Author:** ![PBK](https://avatars.discourse-cdn.com/v4/letter/p/b9e5f3/32.png) [@PBK](https://community.plotly.com/u/PBK)\
**Post date:** [July 20, 2020, 2:40am UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/4 "2020-07-20T02:40:24Z")

</div>

Thank you very much for working on a Sunday night.

---

<div class="post-metadata">

**Author:** ![wgbarron](https://avatars.discourse-cdn.com/v4/letter/w/c6cbf5/32.png) [@wgbarron](https://community.plotly.com/u/wgbarron)\
**Post date:** [October 28, 2020, 1:50am UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/5 "2020-10-28T01:50:20Z")

</div>

@chriddyp you beautiful man thank you

---

<div class="post-metadata">

**Author:** ![dasher007](https://avatars.discourse-cdn.com/v4/letter/d/c57346/32.png) [@dasher007](https://community.plotly.com/u/dasher007)\
**Post date:** [March 23, 2021, 9:45pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/6 "2021-03-23T21:45:35Z")

</div>

Hi Chris,

I came across a similar error. I am using the getData() to pull data from a db every 5 seconds. I am converting the dataframe to dict in my app.

```auto
import dash
import dash_core_components as dcc
import dash_html_components as html
import dash_table
import pandas as pd
import numpy as np
import mysql.connector

app = dash.Dash( __name__ )

def getData():
    conn = mysql.connector.connect(
            host=, # your host, usually localhost
            user=, # your username
            password=, # your password
            database=,# name of the data base
    )    
    c = conn.cursor()
    
    df = pd.read_sql("SELECT DATE_TIME, MES_LenSp, MES_ExzAbs, MES_DEinl, MES_DKalt FROM table WHERE DATE_TIME > '2021-03-23 09:00:00'", conn)
    #Batch data every 1000 
    bins = pd.Series(np.arange(0,160000,1000))
    
    labels = ["{0} - {1}".format(i, i + 999) for i in range(0, 160000, 1000)]
    
    df['Len_bins'] = pd.cut(df['MES_LenSp'], bins=bins, right=False)
    df['Len_bins'] = df['Len_bins'].astype(str)
    
    #Calculations
    
    
    df['Concentricity'] = 1-(df['MES_ExzAbs']/(2*df['MES_DKalt']))
    
    #Impedance calculation 
    dielectric = 1.91
    x = float(138) 
    log10 = 2.302585
    df['Impedance'] = ((x * np.log10(df['MES_DKalt']/df['MES_DEinl']))/(np.sqrt(dielectric)))
    
    #Std dev calculation 
    df['DiamCore_diff'] = df['MES_DEinl'].diff(periods=-1).abs()
    df['DiamCold_diff'] = df['MES_DKalt'].diff(periods=-1).abs()
    df['Impedance_diff'] = df['Impedance'].diff(periods=-1).abs()
    df.dropna(inplace=True)
    
    #calculate mean of the range and divide by 1.128 to get std deviation for every 1000ft 
    
    #Convert date to int64 for groupby 
    df['DATE_TIME'] = pd.to_datetime(df['DATE_TIME']).values.astype(np.int64)
    
    df_result = df.groupby(['Len_bins'], sort=False)[['DATE_TIME','DiamCore_diff',
                                          'DiamCold_diff','Impedance_diff',
                                          'Concentricity']].mean().reset_index()
    df_result.rename(columns={'Concentricity':'Concentricity_Mean'}, inplace=True)
    
    df_result['Concentricity_Mean'] = df_result['Concentricity_Mean'].round(4)
    #convert date/time from int64 to date
    df_result['DATE_TIME'] = pd.to_datetime(df_result['DATE_TIME'], unit='ns')
    
    df_result['DiamCore_STD'] = df_result['DiamCore_diff'] / 1.128
    df_result['DiamCold_STD'] = df_result['DiamCold_diff'] / 1.128
    df_result['Impedance_STD'] = df_result['Impedance_diff'] / 1.128
    
    #Round decimal places 
    df_result['DiamCore_STD'] = df_result['DiamCore_STD'].round(7)
    df_result['DiamCold_STD'] = df_result['DiamCold_STD'].round(7)
    df_result['Impedance_STD'] = df_result['Impedance_STD'].round(4)
    
    
    
    #Compare against limits and generate alarm log
    df_result['DiamCore_STD_Alarm'] = np.where(df_result['DiamCore_STD'] > 0.000012, 1, 0)
    df_result['DiamCold_STD_Alarm'] = np.where(df_result['DiamCold_STD'] > 0.000065, 1, 0)
    df_result['Impedance_STD_Alarm'] = np.where(df_result['Impedance_STD'] > 0.07, 1, 0)
    
    #Concentricity Alarm
    df_con = df[['Len_bins','Concentricity']].copy()
    df_con['Concentricity_v'] = np.where(df_con['Concentricity'] > 0.95, 1, 0)
    df_con = df_con.groupby('Len_bins')[['Concentricity_v']].sum().reset_index()
    df_con['Conc_Alarm'] = np.where(df_con['Concentricity_v'] > 500, 1, 0)
    df_con = df_con[['Len_bins','Conc_Alarm']]
    
    #Merge concentricity alarm with DiamCold, DiamCore, Impedance Alarm
    df_result = pd.merge(df_result, df_con, on='Len_bins', validate='one_to_many')
    
    df_result['Alarm'] = np.where((df_result['DiamCore_STD_Alarm'] + 
                                   df_result['DiamCold_STD_Alarm'] + 
                                   df_result['Impedance_STD_Alarm'] +
                                   df_result['Conc_Alarm']) >= 1, 1, 0)
   
    
    
    #Convert date/time to date
    df['DATE_TIME'] = pd.to_datetime(df['DATE_TIME'])
    
    
    
    #Drop rows where impedance is 0 to retain non zero rows only
    df_result = df_result[df_result['Impedance_diff'] > 0]

    
    df_summary = df_result[['DATE_TIME','Len_bins','DiamCore_STD','DiamCold_STD',
                            'Impedance_STD','Concentricity_Mean','Alarm']].copy()
    df_summary['DATE_TIME'] = pd.to_datetime(df_summary['DATE_TIME'], 
                                              format='%Y-%m-%d %H:%M:%S').dt.strftime('%Y-%m-%d %H:%M:%S')
    
    
    df_summary['Len_bins'] = df_summary['Len_bins'].astype(str)
    return df_summary

app.layout = html.Div([
      html.H4('Dashboard'),
      dcc.Interval('graph-update', interval = 5000, n_intervals = 0),
      dash_table.DataTable(
          id = 'table',
          data = getData().to_dict('records'),
          columns=[{'id': c, 'name': c} for c in getData().columns],
                        page_size = 15,
                        style_as_list_view=True,
                        style_header={
                                    'fontWeight': 'bold'
                                    },

                        
                        css=[{'selector': 'table', 'rule': 'table-layout: fixed'}],
                        style_cell={
                                    'width': '{}%'.format(len(getData().columns)),
                                    'whiteSpace': 'normal',
                                    'padding': '7px',
                                    'height': 'Auto',
                                    'fontSize': 24,
                                    'textAlign': 'center'
                                    },
                        
                        style_data_conditional=[{
                                                'if': {
                                                    'filter_query': '{Alarm} contains "1"',
                                                    'column_id': 'Alarm'
                                                    },  
                                                    'backgroundColor': '#FF4136',
                                                    'color': 'white'
                                                    },
                                                {
                                                'if': {
                                                    'filter_query': '{Alarm} contains "0"',
                                                    'column_id': 'Alarm'
                                                    },  
                                                    'backgroundColor': '#3D9970',
                                                    'color': 'white'
                                                    },
                                            
                                                ] 
         )])

@app.callback(
        dash.dependencies.Output('table','data'),
        [dash.dependencies.Input('graph-update', 'n_intervals')])
def updateTable(n):
     return getData()

if __name__ == ' __main__':
     app.run_server(port = 8087,debug=False)

```

The error message:

```auto
Traceback (most recent call last):
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 2447, in wsgi_app
    response = self.full_dispatch_request()
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1952, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1821, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\_compat.py", line 39, in reraise
    raise value
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1950, in full_dispatch_request
    rv = self.dispatch_request()
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1936, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\dash.py", line 1078, in dispatch
    response.set_data(func(*args, outputs_list=outputs_list))
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\dash.py", line 1044, in add_context
    _validate.fail_callback_output(output_value, output)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 261, in fail_callback_output
    _validate_value(val, index=i)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 251, in _validate_value
    _raise_invalid(
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 190, in _raise_invalid
    raise exceptions.InvalidCallbackReturnValue(
dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `table.data`>`
returned a value having type `DataFrame`
which is not JSON serializable.

The value in question is either the only value returned,
or is in the top level of the returned list,

and has string representation

```

Does it matter if i convert the dataframe to dict in the app as opposed to in the getData()? Doing this because its easier for me to format column headers

Any help is appreciated. 🙂

---

<div class="post-metadata">

**Author:** ![AnnMarieW](https://avatars.discourse-cdn.com/v4/letter/a/ec9cab/32.png) [@AnnMarieW](https://community.plotly.com/u/AnnMarieW)\
**Post date:** [March 23, 2021, 11:33pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/7 "2021-03-23T23:33:12Z")

</div>

Hi @dasher007

Try changing your callback to:

```auto
return get_data().to_dict('records')

```

---

<div class="post-metadata">

**Author:** ![dasher007](https://avatars.discourse-cdn.com/v4/letter/d/c57346/32.png) [@dasher007](https://community.plotly.com/u/dasher007)\
**Post date:** [March 24, 2021, 12:56pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/8 "2021-03-24T12:56:26Z")

</div>

Hi @AnnMarieW, thanks for the sugestion. I changed the getData function to return as dict

```auto
return getData().to_dict('records')

```

but that didn’t help, still throws out the same error.

---

<div class="post-metadata">

**Author:** ![dasher007](https://avatars.discourse-cdn.com/v4/letter/d/c57346/32.png) [@dasher007](https://community.plotly.com/u/dasher007)\
**Post date:** [March 24, 2021, 1:35pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/9 "2021-03-24T13:35:37Z")

</div>

FYI, the changes i made

```auto
def getData():
.
.
.
    return getData().to_dict('records')

app.layout = html.Div([
      html.H4('Dashboard'),
      dcc.Interval('graph-update', interval = 5000, n_intervals = 0),
      dash_table.DataTable(
          id = 'table',
          data = getData().to_dict('records'),
          columns=[{'id': c, 'name': c} for c in getData().columns],
                        page_size = 15,
                        style_as_list_view=True,
                        style_header={
                                    'fontWeight': 'bold'
                                    },

                        
                        css=[{'selector': 'table', 'rule': 'table-layout: fixed'}],
                        style_cell={
                                    'width': '{}%'.format(len(getData().columns)),
                                    'whiteSpace': 'normal',
                                    'padding': '7px',
                                    'height': 'Auto',
                                    'fontSize': 24,
                                    'textAlign': 'center'
                                    },
                        
                        style_data_conditional=[{
                                                'if': {
                                                    'filter_query': '{Alarm} contains "1"',
                                                    'column_id': 'Alarm'
                                                    },  
                                                    'backgroundColor': '#FF4136',
                                                    'color': 'white'
                                                    },
                                                {
                                                'if': {
                                                    'filter_query': '{Alarm} contains "0"',
                                                    'column_id': 'Alarm'
                                                    },  
                                                    'backgroundColor': '#3D9970',
                                                    'color': 'white'
                                                    },
                                            
                                                ] 
         )])

@app.callback(
        dash.dependencies.Output('table','data'),
        [dash.dependencies.Input('graph-update', 'n_intervals')])
def updateTable(n):
     return getData().to_dict('records')

if __name__ == ' __main__':
     app.run_server(port = 8087,debug=False)

```

The error:

```auto
In general, Dash properties can only be
dash components, strings, dictionaries, numbers, None,
or lists of those.
127.0.0.1 - - [24/Mar/2021 08:24:36] "POST /_dash-update-component HTTP/1.1" 500 -
[2021-03-24 08:24:39,206] ERROR in app: Exception on /_dash-update-component [POST]
Traceback (most recent call last):
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\dash.py", line 1040, in add_context
    jsonResponse = json.dumps(
  File "C:\Users\savanna\Anaconda3\lib\json\ __init__.py", line 234, in dumps
    return cls(
  File "C:\Users\savanna\Anaconda3\lib\site-packages\_plotly_utils\utils.py", line 59, in encode
    encoded_o = super(PlotlyJSONEncoder, self).encode(o)
  File "C:\Users\savanna\Anaconda3\lib\json\encoder.py", line 199, in encode
    chunks = self.iterencode(o, _one_shot=True)
  File "C:\Users\savanna\Anaconda3\lib\json\encoder.py", line 257, in iterencode
    return _iterencode(o, 0)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\_plotly_utils\utils.py", line 134, in default
    return _json.JSONEncoder.default(self, obj)
  File "C:\Users\savanna\Anaconda3\lib\json\encoder.py", line 179, in default
    raise TypeError(f'Object of type {o. __class__. __name__ } '
TypeError: Object of type DataFrame is not JSON serializable

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 2447, in wsgi_app
    response = self.full_dispatch_request()
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1952, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1821, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\_compat.py", line 39, in reraise
    raise value
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1950, in full_dispatch_request
    rv = self.dispatch_request()
  File "C:\Users\savanna\Anaconda3\lib\site-packages\flask\app.py", line 1936, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\dash.py", line 1078, in dispatch
    response.set_data(func(*args, outputs_list=outputs_list))
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\dash.py", line 1044, in add_context
    _validate.fail_callback_output(output_value, output)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 261, in fail_callback_output
    _validate_value(val, index=i)
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 251, in _validate_value
    _raise_invalid(
  File "C:\Users\savanna\Anaconda3\lib\site-packages\dash\_validate.py", line 190, in _raise_invalid
    raise exceptions.InvalidCallbackReturnValue(
dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `table.data`>`
returned a value having type `DataFrame`
which is not JSON serializable.

```

Another thing I noticed is that when i print getData() i see the data getting updated however the data doesn’t update in the app.

Any help is appreciated!

---

<div class="post-metadata">

**Author:** ![Ipan98](https://avatars.discourse-cdn.com/v4/letter/i/3ec8ea/32.png) [@Ipan98](https://community.plotly.com/u/Ipan98)\
**Post date:** [June 10, 2021, 3:08pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/10 "2021-06-10T15:08:42Z")

</div>

Hi guys I am having the same errors:

```auto
def findActivity(choice):
  if choice>=1 and choice<=100:
    return 'Recovery'
  elif choice >100 and choice <=200:
    return 'Preparative'
  elif choice >=180 and choice <=325:
    return 'Organizational'
  elif choice >=300 and choice <=450:
    return 'Maintenance'
  elif choice >450 and choice <=640:
    return 'Thresold'
  elif choice >=640 and choice <=900:
    return 'Fitness'
  elif choice >=810 and choice <=1200:
    return 'Game'
  else:
    return "Training"
def idFunction(year,ind, indexList, i, days, weekDay):
  if ind==0 and days[weekDay][ind] not in [1,2,3,4,5,6]:
    if i==1:
      m_id=date(year-1, 12, days[weekDay][ind])
    else:
      m_id=date(year, i-1, days[weekDay][ind])
  elif ind==indexList[-1] and days[weekDay][ind] not in [24,25,26,27,28,29, 30,31]:
    if i==12:
      m_id=date(year+1, 1, days[weekDay][ind])
    else:
      m_id=date(year, i+1, days[weekDay][ind])
  else:
    m_id=date(year, i, days[weekDay][ind])

  return m_id
def activity(d):
  year=int(d[0:4])
  month=int(d[5:7])
  day=int(d[8:])
  
  c = conn.cursor()
  c.execute('SELECT activity FROM planner WHERE date=%s and teamName =%s', (datetime.date(year,month,day),"UCF"))
  result = c.fetchone()#fetchall
  c.execute('SELECT first, last, sum(load) AS load FROM "UCF_Women_Soccer_DEV".ff90_training_load WHERE date=%s GROUP BY first, last', (datetime.date(year,month,day),))
  result2 = c.fetchall()
  if result:
    return result[0]#result[0]
  elif result2:
    load=[value for first, last, value in result2]
    mode=statistics.mode(load)
    return findActivity(mode)
  else:
    return "OFF"
my_calendar = html.Div([
    dbc.Row(
        dbc.Col(
            dcc.Dropdown(
                id="month",
                options=[{
                    "label": x,
                    "value": x
                } for x in calendar.month_name],
                value=calendar.month_name[date.today().month],
                clearable=False,
            ))),
    html.Br(),
])
years = html.Div([
    dbc.Row(
        dbc.Col(
            dcc.Dropdown(
                id="years",
                options=[{
                    "label": x,
                    "value": x
                } for x in range(2000,2100)],
                value=2021,
                clearable=False,
            ))),
    html.Br(),
])
table_header = [
    html.Thead(html.Tr([html.Th("KEY"), html.Th(dbc.Badge("RECOVERY", color="primary")), html.Th(dbc.Badge("PREPARATIVE", color="success")), html.Th(dbc.Badge("ORGANIZATIONAL", color="warning")), html.Th(dbc.Badge("MAINTENANCE", color="info")), html.Th(dbc.Badge("THRESOLD", color="light")), html.Th(dbc.Badge("FITNESS", color="dark")), html.Th(dbc.Badge("GAME", color="danger"))]))
]

row1 = html.Tr([html.Td("RPE"), html.Td("<3"),html.Td("3-4"), html.Td("4-5"), html.Td("5-6"), html.Td("7-8"), html.Td("8-9"), html.Td("9-10")])
row2 = html.Tr([html.Td("VOL"), html.Td("<35min"), html.Td("35-45min"), html.Td("45-60min"), html.Td("45-75min"), html.Td("75-80min"), html.Td("80-100min"), html.Td("90min")])
row3 = html.Tr([html.Td("TL"), html.Td("<100"), html.Td("105-200"), html.Td("180-325"), html.Td("300-450"), html.Td("490-640"), html.Td("640-900"), html.Td(">810")])
row4 = html.Tr([html.Td("Readiness"), html.Td("<65"), html.Td(">77.5"), html.Td("66-75"), html.Td("70-80"), html.Td("75-85"), html.Td("82.5-95"), html.Td("87.5")])
row5 = html.Tr([html.Td("Recovery"), html.Td("n/a"), html.Td("12hr"), html.Td("<24hr"), html.Td("24hr"), html.Td("48hr"), html.Td("72hr"), html.Td(">72hr")])
table_body = [html.Tbody([row1, row2, row3, row4, row5])]

table = dbc.Table(table_header + table_body, bordered=True,
    hover=True,
    responsive=True,
    striped=True,)
DATABASE_URL = 'postgres://panjfsdgbbpqjp:e82392fdcc0e809c9ee02a9459d9da3e0e414094bdbc8f88acb69d3f7686da5a@ec2-52-44-31-100.compute-1.amazonaws.com:5432/d2itj4h57p7v0h'
conn = psycopg2.connect(DATABASE_URL, sslmode='require')
options6=["Highest Training Load", "Lowest Training Load"]
colors = {
    'background': '#ffffed',
    'text': '#7FDBFF'
}#style={'backgroundColor': colors['background']},
def layout():
  return html.Div(children=[
                    #html.Div(my_calendar, id='monthChoice'),
                    dbc.Row(html.H1("Planner"), className="container_title",),
                    html.Br(),
                    html.Div(dbc.Row([dbc.Col(html.H4("Team average", style={
        'textAlign': 'left',
    })), dbc.Col(html.Div([dbc.Row(html.H4("Last 7 days")),dcc.Dropdown(
                                    id="LastSevenDays",
                                    options=[{
                                        "label": x,
                                        "value": x
                                    } for x in options6],
                                    value=options6[0],
                                    clearable=False,
                                ), html.Div(id="LastSevenDayOutput")], className="container_title"))])),
                    html.Br(),html.Br(),
                    html.Div([dbc.Row([dbc.Col(my_calendar), dbc.Col(years)], no_gutters=True),
                    dbc.Row([
                        dbc.Col(
                            [html.Div("Monday"),
                             ])
                    ], no_gutters=True)], style={
                        "width": "98%",
                        "margin-right": "0", 
                    }),html.Br(),html.Br(),html.Div(table),html.Br(),
                    dbc.Row([dbc.Col(html.Div(id="day-activity", style={'overflowY':'scroll', 'height':400, 'width':500})),dbc.Col(dcc.Graph(id='forecast', style={'display': 'none'}))]),
                    html.Div(id="output2-container", children=[], style={'display': 'none'}),
                
                    html.Div(id='dropdownChoice'),
                    html.Div(id='Output6', style={'display': 'none'}),
])
@app.callback(Output("LastSevenDayOutput", "children"), Input("LastSevenDays", "value"))
def lastSevenDaysOutput(choice):
  c = conn.cursor()
  if choice=="Highest Training Load":
    c.execute('select first || \' \'|| last as player_name, sum(load) as Training_load \
                    from "UCF_Women_Soccer_DEV".ff90_training_load \
                    where date >\'' + "2021-02-28" + '\' and date <= \'' + "2021-03-7"+ '\' \
                    group by ROLLUP (player_name) order by Training_load DESC LIMIT 6')
  else:
    c.execute('select first || \' \'|| last as player_name, sum(load) as Training_load \
                    from "UCF_Women_Soccer_DEV".ff90_training_load \
                    where date >\'' + "2021-02-28" + '\' and date <= \'' + "2021-03-7"+ '\' \
                    group by ROLLUP (player_name) order by Training_load LIMIT 5')
  result=c.fetchall()
  row=[]
  names=[]
  load=[]
  if result:
    for i in result:
      if i[0] ==None:
        continue 
      j=i[1]/7
      j=round(j,1)
      names.append(i[0])
      load.append(j)
      #row.append(dbc.Row([dbc.Col(i[0]), dbc.Col(j)]))
    list_of_tuples = list(zip(names, load))  
    df = pd.DataFrame(list_of_tuples,
                  columns = ['Name', '7 Days average'])
    return dash_table.DataTable(
    id='table',
    columns=[{"name": i, "id": i} for i in df.columns],
    data=df.to_dict('records'),
) 
  return row

@app.callback([Output("monday", "children")], [Input("month", "value"), Input("years", "value")])
def displayCalendar(value, yearss):
    df = [
        "January", "February", "March", "April", "May", "June", "July",
        "August", "September", "October", "November", "December"
    ]
    data1 = [
        'Preparative', 'Organizational', 'Maintenance', 'Thresold', 'Fitness',
        'Game', 'Recovery'
    ]
    for i, month in enumerate(df):
        if month == value:
            break
    i = i + 1
    print(i)
    days = calendar.monthcalendar(yearss, i)
    if i==1:
      previousMonth=calendar.monthcalendar(yearss-1, 12)
    else:
      previousMonth=calendar.monthcalendar(yearss, i-1)
    if i==12:
      nextMonth=calendar.monthcalendar(yearss+1, 1)
    else:
      nextMonth=calendar.monthcalendar(yearss, i+1)
    a=0
    for j in days[0]:
      if j ==0:
        days[0][a]=previousMonth[-1][a]
      a+=1
    a=0
    for j in days[-1]:
      if j ==0:
        days[-1][a]=nextMonth[0][a]
      a+=1
    monday = []
    data1 = [
        'Preparative', 'Organizational', 'Maintenance', 'Thresold', 'Fitness',
        'Game', 'Recovery', 'OFF'
    ]
    j = days
    days = pd.DataFrame(days)
    days.columns = [
        'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday',
        'Sunday'
    ]
    for ind in days.index:
        m_id=str(idFunction(yearss,ind, days.index, i, days, 'Monday'))
        m_act=activity(m_id)
        if days['Monday'][ind] != 0:
            monday.append(
                dbc.Card(
                    dbc.CardBody([
                        html.Div(
                            dbc.Button(children=days['Monday'][ind],
                                       id={
                                           'type': 'date-button',
                                           'index': m_id
                                       },
                                       color="link")),
                       dcc.Dropdown(
                                    id={
                'type': 'dropdown2',
                'index': m_id },
                                    options=[{
                                        "label": x,
                                        "value": x
                                    } for x in data1],
                                    value=m_act,
                                    clearable=False,
                                )
                                
                                       
                    ])))
      
    monday.append(dcc.Store(id='permanent-date'))
    monday.append(dcc.Store(id='permanent-object'))
    monday.append(dcc.ConfirmDialog(
        id='confirm',
        message='Doing this change might override training specification, are you sure you want to continue?',
    ))
    
    return monday

```

I have a multi page daash app, and I get the error when I come at the page with this code

---

<div class="post-metadata">

**Author:** ![ckeoirm](https://avatars.discourse-cdn.com/v4/letter/c/b3f665/32.png) [@ckeoirm](https://community.plotly.com/u/ckeoirm)\
**Post date:** [January 20, 2022, 6:52am UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/11 "2022-01-20T06:52:42Z")

</div>

same issue:

i am using ayncio and aoodbc and i get the following bug:

_dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `my-output.children`>`_  
_returned a value having type `coroutine`_  
_which is not JSON serializable._

_The value in question is either the only value returned,_  
_or is in the top level of the returned list,_

_and has string representation_  
_`<coroutine object datasets_refresh at 0x000001DF45383940>`_

_In general, Dash properties can only be_  
_dash components, strings, dictionaries, numbers, None,_  
_or lists of those._

---

<div class="post-metadata">

**Author:** ![jlfsjunior](https://sea2.discourse-cdn.com/flex024/user_avatar/community.plotly.com/jlfsjunior/32/16061_2.png) [@jlfsjunior](https://community.plotly.com/u/jlfsjunior)\
**Post date:** [January 20, 2022, 10:31am UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/12 "2022-01-20T10:31:39Z")

</div>

Could you share your callback with us? All the errors reported here is due to a callback function returning an object that is not JSON serializable to a `DataFrame` “data” props. In your case, it seems that you are missing an assignment in your `await` statement, if the coroutine returns the data.

`data` should be a list of dictionaries in the same format as in `pd.DataFrame.to_dict("records")`.

---

<div class="post-metadata">

**Author:** ![Robin66014](https://avatars.discourse-cdn.com/v4/letter/r/dbc845/32.png) [@Robin66014](https://community.plotly.com/u/Robin66014)\
**Post date:** [April 26, 2023, 2:07pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/13 "2023-04-26T14:07:55Z")

</div>

I have a somewhat similar problem.  
I have this working for a lot of pandas dataframes, but now I have a dictionary ({‘Data Type’: {‘Id’: dtype(‘int64’), ‘SepalLengthCm’: dtype(‘float64’), ‘SepalWidthCm’: dtype(‘float64’), ‘PetalLengthCm’: dtype(‘float64’), ‘PetalWidthCm’: dtype(‘float64’), ‘Species’: dtype(‘int32’)}, ‘Missing Values%’: { … }}

I have converted this to a pandas dataframe both using pd.DataFrame() and pd.DataFrame.from\_dict() which works fine as you can see in this image:

 ![image](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/a/2/a2b544780503d663762be98829017b6bc5974b37.png)

However, when I now try to convert this dataframe using .to\_dict(‘records’) (dash\_table.DataTable(df.to\_dict(‘records’)), )

```auto
dash.exceptions.InvalidCallbackReturnValue: The callback for `<Output `container-checks-button-pressed.children`>`
                returned a tree with one value having type `DataTable`
                which is not JSON serializable.

The value in question is located at
[0] Div 
[3] DataTable,

                and has string representation
                `DataTable(data=[{'Data Type': dtype('int64'), 'Missing Values%': 0.0, 'Unique Values%': 100, 'Minimum Value': 1.0, 'Maximum Value': 150.0, 'DQ Issue': 'Possible ID colum: drop before modeling process., Id has a correlation >= 0.8 with Species. Possible data leakage. Double check this variable.'}, {'Data Type': dtype('float64'), 'Missing Values%': 0.0, 'Unique Values%': 'NA', 'Minimum Value': 4.3, 'Maximum Value': 7.9, 'DQ Issue': 'No issue'}, {'Data Type': dtype('float64'), 'Missing Values%': 0.0, 'Unique Values%': 'NA', 'Minimum Value': 2.0, 'Maximum Value': 4.4, 'DQ Issue': 'has 4 outliers greater than upper bound (4.05) or lower than lower bound(2.05). Cap them or remove them.'}, {'Data Type': dtype('float64'), 'Missing Values%': 0.0, 'Unique Values%': 'NA', 'Minimum Value': 1.0, 'Maximum Value': 6.9, 'DQ Issue': "has a high correlation with ['Id', 'SepalLengthCm']. Consider dropping one of them., PetalLengthCm has a correlation >= 0.8 with Species. Possible data leakage. Double check this variable."}, {'Data Type': dtype('float64'), 'Missing Values%': 0.0, 'Unique Values%': 'NA', 'Minimum Value': 0.1, 'Maximum Value': 2.5, 'DQ Issue': "has a high correlation with ['Id', 'SepalLengthCm', 'PetalLengthCm']. Consider dropping one of them., PetalWidthCm has a correlation >= 0.8 with Species. Possible data leakage. Double check this variable."}, {'Data Type': dtype('int32'), 'Missing Values%': 0.0, 'Unique Values%': 2, 'Minimum Value': 0.0, 'Maximum Value': 2.0, 'DQ Issue': "has a high correlation with ['Id', 'PetalLengthCm', 'PetalWidthCm']. Consider dropping one of them."}])`

                In general, Dash properties can only be
                dash components, strings, dictionaries, numbers, None,
                or lists of those.

```

Any idea whats going wrong? as up until the pandas dataframe it all goes well, and using the .to\_dict(‘records’) I have converted more dataframes to datatables, but somehow for this one it doesn’t work

---

<div class="post-metadata">

**Author:** ![edmoman](https://sea2.discourse-cdn.com/flex024/user_avatar/community.plotly.com/edmoman/32/20693_2.png) [@edmoman](https://community.plotly.com/u/edmoman)\
**Post date:** [April 26, 2023, 2:24pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/14 "2023-04-26T14:24:21Z")

</div>

That df has multiple indexes. Perhaps you need to flatten them.

---

<div class="post-metadata">

**Author:** ![AlexisRangelC](https://avatars.discourse-cdn.com/v4/letter/a/e68b1a/32.png) [@AlexisRangelC](https://community.plotly.com/u/AlexisRangelC)\
**Post date:** [May 23, 2023, 6:42pm UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/15 "2023-05-23T18:42:36Z")

</div>

Hi there, check again your dataframe structure, i was struggling with this situation and the problem was the headers (maybe DQ issue in your case) … try to make less complex the table in order to debug the situation

---

<div class="post-metadata">

**Author:** ![Robin66014](https://avatars.discourse-cdn.com/v4/letter/r/dbc845/32.png) [@Robin66014](https://community.plotly.com/u/Robin66014)\
**Post date:** [May 30, 2023, 8:51am UTC](https://community.plotly.com/t/app-callback-output-error-message-dash-exceptions-invalidcallbackreturnvalue/42668/16 "2023-05-30T08:51:31Z")

</div>

It was solved by resetting the indexes indeed so that only one indexing existed! Resetting the index and removing curly brackets often solve the problem, time has learned me 🙂