Using Dataiku
Sort by:
21 - 30 of
96
- Hi, We have created a generic Dataiku Project which need the variable to be updated in each run. Instead of opening Dataiku project and changing the variable parameter or creating the multiple scenari…
Last answer by MattscoHi,
You can use the following code to edit the variable from a python recipe:
import dataiku
client = dataiku.api_client()
project = client.get_project("PROJECTKEY")
variables = project.get_variables()
print(variables)
variables["standard"]["foo"] = "bar"
project.set_variables(variables)
print(variables)Last answer by MattscoHi,
You can use the following code to edit the variable from a python recipe:
import dataiku
client = dataiku.api_client()
project = client.get_project("PROJECTKEY")
variables = project.get_variables()
print(variables)
variables["standard"]["foo"] = "bar"
project.set_variables(variables)
print(variables) - In a jupyter notebook I can use the Dataset class from the dataiku package to read a dataset if I know it's name. However, I can't find a way to list all the datasets programmatically, in a notebook. …Solution bySolution by Mattsco
Hi David,
Yes, you can do :
import dataiku
client = dataiku.api_client()
project = client.get_project('TEST_PROJECT')
datasets = project.list_datasets() - Hello, I'm trying to generate a new Dataset from a custom python recipe (inside a Plugin) but I'm getting always Error Disk (DSS can't find the Dataset - i can found it in the flow). I think that I'm …Solution by
Solution by Alan_FustéWith Sync Recipe:
client = dataiku.api_client()
project = client.get_project(project_name)
middle_dataset_name = 'pre_' + output_dataset_name
builder = SyncRecipeCreator("sync_" + middle_dataset_name, project)
builder = builder.with_input(input_dataset_name)
builder = builder.with_new_output(middle_dataset_name, "filesystem_managed")
recipe = builder.build()
recipe_def = recipe.get_definition_and_payload()
recipe_payload = {}
recipe_def.data['recipe']['params']['schemaMode'] = 'FREE_SCHEMA_NAME_BASED'
recipe_def.set_json_payload(recipe_payload)
recipe.set_definition_and_payload(recipe_def)
middle_dataset = dataiku.Dataset(middle_dataset_name, project_key=project_name, ignore_flow=True) - Hi all, I have created a plugin custom recipe, it works well and I want to use it inside a python code (in a scenario) like I'm doing, for example, with the SyncRecipe (defining input, output and buil…Solution bySolution by Alan_FustéIt's not possible
- Hello, Can I create a dataset (for example, with project.create_dataset()) directly from pandas (without saving before the creation)? I am getting a CSV file from an url with pandas.read_csv() Thanks,…Solution by
Solution by Clément_StenacHi,
While it is possible, the project.create_dataset() API is currently fairly difficult to use. We do not yet have documentation nor samples for it.
We recommend that you create a dataset from the UI first. - What is the correct way to delete a scenario via python api? Project has a create_scenario method, but no delete_scenario, and the scenario object does not have delete method as well.Solution bySolution by Clément_StenacScenario deletion is not part of the public API
Model's predicted values are not the same when using Dataiku API to retrieve the model and use it in
I built an XGBoost model in DSS, and retrieved it using the Dataiku API inside python recipe. However, when I use the predict() function on the retrieved model object, the predicted values are not the…Last answer by
Last answer by UserBirdHi,
The original model apply a set of preprocessing on data before doing the prediction. If you want to reproduce the result, you can use directly `my_predictor.predict()`.
Another thing to be aware is that in the scoring recipe, for reproducebility, you should check the Force backend engine option
Cheers,
21 - 30 of
963