The Dataiku Frontrunner Awards have just launched to recognize your achievements! Submit Your Entry

New plugin

devesh
Level 1
New plugin

Hello Everyone,

I'm creating a Plugin for model Monitoring,

Which has 4 Input datasets and 5 Output datasets.

in Recipe.py : I have input these table as - 

input_model_dataset = get_input_names_for_role('input1')
input_model_dataset = get_input_names_for_role('input2')
input_model_dataset = get_input_names_for_role('input3')
input_model_dataset = get_input_names_for_role('input4')

 

for reading the datasets : 

 

input1_df = input_datasets[0]
input2_df = input_datasets[1]
input3_df = input_datasets[2]
input4_df = input_datasets[3]

new_input1 =  = input1_df.get_dataframe()
new_input2 = input2_df.get_dataframe()
new_input3 = input3_df.get_dataframe()
new_input4 = input4_df.get_dataframe()

but i'm getting error as 

Error in python process: At line 47: <type 'exceptions.IndexError'>: list index out of range

 

Can anybody help me with it.

0 Kudos
4 Replies
tgb417
Neuron
Neuron

@devesh ,

It's super cool that you are working on building your own model monitoring toolset.

I'm not clear that I have an answer for your error message.  And this may be a little bit off-topic.

However, yesterday I learned about an existing plugin that has already been created that I was not aware of called the Model Drift Monitoring Plugin.   

https://www.dataiku.com/product/plugins/model-drift/

Considering the work you are doing you might find this either interesting or even useful.

 

--Tom
0 Kudos
NN
Level 3

Just trying to see if i can help you out. hoping i have understood your request correctly.
 i usally call my input datasets this way 

 

input1=get_input_names_for_role('input1')[0]
input1_ds=dataiku.Dataset(input1)
input1_df=input1_ds.get_dataframe()

 

devesh
Level 1
Author
input1=get_input_names_for_role('input1')[0]

This lne is giving me error, List is out index.

I don't know what causing this.

0 Kudos
NN
Level 3

is your input role same as below ?

"inputRoles": [
        {
            "name": "input1",
            "label": "First Dataset",
            "arity": "UNARY",
            "required": true,
            "acceptsDataset": true
        }
]

 

You can also check out this example
https://knowledge.dataiku.com/latest/kb/customization/plugins/recipe.html#edit-definitions-in-recipe...

0 Kudos
A banner prompting to get Dataiku DSS