good morning all We are facing an error in using global vari Kedro #questions

good morning all! We are facing an error in using ...

Ionut Barbu

10/08/2024, 9:44 AM

good morning all! We are facing an error in using global variable interpolation with the OmegaConfigLoader. The error occurs when lunching a jupyter notebook e.g with

kedro jupyter lab

the issue seems very similar/identical to the one signaled here https://kedro-org.slack.com/archives/C03RKP2LW64/p1726216824633969 the full error stack is below. The global var is located in

conf\globals.yml

The issue also occurs for the location

conf\base\globals.yml

Any help from the kedro team is very much appreciated

Copy code

Traceback (most recent call last):
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\IPython\core\shellapp.py", line 322, in init_extensions
    self.shell.extension_manager.load_extension(ext)
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\IPython\core\extensions.py", line 62, in load_extension
    return self._load_extension(module_str)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\IPython\core\extensions.py", line 79, in _load_extension
    if self._call_load_ipython_extension(mod):
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\IPython\core\extensions.py", line 129, in _call_load_ipython_extension
    mod.load_ipython_extension(self.shell)
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\ipython\__init__.py", line 62, in load_ipython_extension
    reload_kedro()
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\ipython\__init__.py", line 123, in reload_kedro
    catalog = context.catalog
              ^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\framework\context\context.py", line 187, in catalog
    return self._get_catalog()
           ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\framework\context\context.py", line 223, in _get_catalog
    conf_catalog = self.config_loader["catalog"]
                   ~~~~~~~~~~~~~~~~~~^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\config\omegaconf_config.py", line 201, in __getitem__
    base_config = self.load_and_merge_dir_config(  # type: ignore[no-untyped-call]
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\config\omegaconf_config.py", line 341, in load_and_merge_dir_config
    for k, v in OmegaConf.to_container(
                ^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\omegaconf.py", line 573, in to_container
    return BaseContainer._to_content(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\basecontainer.py", line 292, in _to_content
    value = get_node_value(key)
            ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\basecontainer.py", line 247, in get_node_value
    value = BaseContainer._to_content(
            ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\basecontainer.py", line 292, in _to_content
    value = get_node_value(key)
            ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\basecontainer.py", line 244, in get_node_value
    conf._format_and_raise(key=key, value=None, cause=e)
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 231, in _format_and_raise
    format_and_raise(
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\_utils.py", line 899, in format_and_raise
    _raise(ex, cause)
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\_utils.py", line 797, in _raise
    raise ex.with_traceback(sys.exc_info()[2])  # set env var OC_CAUSE=1 for full trace
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\basecontainer.py", line 242, in get_node_value
    node = node._dereference_node()
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 246, in _dereference_node
    node = self._dereference_node_impl(throw_on_resolution_failure=True)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 277, in _dereference_node_impl
    return parent._resolve_interpolation_from_parse_tree(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 584, in _resolve_interpolation_from_parse_tree
    resolved = self.resolve_parse_tree(
               ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 764, in resolve_parse_tree
    return visitor.visit(parse_tree)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\antlr4\tree\Tree.py", line 34, in visit
    return tree.accept(self)
           ^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar\gen\OmegaConfGrammarParser.py", line 206, in accept
    return visitor.visitConfigValue(self)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar_visitor.py", line 101, in visitConfigValue
    return self.visit(ctx.getChild(0))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\antlr4\tree\Tree.py", line 34, in visit
    return tree.accept(self)
           ^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar\gen\OmegaConfGrammarParser.py", line 342, in accept
    return visitor.visitText(self)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar_visitor.py", line 301, in visitText
    return self._unescape(list(ctx.getChildren()))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar_visitor.py", line 389, in _unescape
    text = str(self.visitInterpolation(node))
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar_visitor.py", line 125, in visitInterpolation
    return self.visit(ctx.getChild(0))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\antlr4\tree\Tree.py", line 34, in visit
    return tree.accept(self)
           ^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar\gen\OmegaConfGrammarParser.py", line 1041, in accept
    return visitor.visitInterpolationResolver(self)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\grammar_visitor.py", line 179, in visitInterpolationResolver
    return self.resolver_interpolation_callback(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 750, in resolver_interpolation_callback
    return self._evaluate_custom_resolver(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\base.py", line 694, in _evaluate_custom_resolver
    return resolver(
           ^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\omegaconf\omegaconf.py", line 445, in resolver_wrapper
    ret = resolver(*args, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\config\omegaconf_config.py", line 384, in _get_globals_value
    raise InterpolationResolutionError(
omegaconf.errors.InterpolationResolutionError: Globals key 'model_to_use' not found and no default value provided.
    full_key: performance_metrics_best_model.filepath
    object_type=dict

Nok Lam Chan

10/08/2024, 9:52 AM

Can you try open a regular notebook and run this as the top cell %load_ext kedro.ipython?

Ionut Barbu

10/08/2024, 10:00 AM

thanks Nok for the prompt reply! Opened a new notebook just with

jupyter lab

the command gives the same error.

Ionut Barbu

10/08/2024, 10:01 AM

Untitled

Nok Lam Chan

10/08/2024, 10:03 AM

alright. Did it works previously until you add some new thing, or did you pick up an old codebase? I will need these information: • How did you define your global configs & where is the config • How did you reference the global config & where it the config

Nok Lam Chan

10/08/2024, 10:17 AM

I left a comment there - but if you are using a class out of a

kedro run

context, you need to either: • make sure

bootstrap_project

are run, so that you load your project settings, and these settings have to be pass to the class correctly. (usually you don't have to handle this because kedro handles it, but you may need to to this when you write test) • Pass the argument to the certain class. For example, you may see

CONFIG_LOADER_ARGS

settings.py

, if you are creating them explicitly, you need to pass those argument to the constructor.

Ionut Barbu

10/08/2024, 11:16 AM

yes, it worked before : 1. we added global configuration stored in

conf/globals.yml

. The definition is

Copy code

## choose from LightGBMModel, SARIMAXModel
model_to_use: SARIMAXModel

No updates made to

settings.py

2. the global config is referenced in the catalog

Copy code

performance_metrics_best_model:
  type: pandas.ParquetDataset
  filepath: "data/08_reporting/performance_metrics_best_model_${globals:model_to_use}.pq"

timeseries_model:
  type: pickle.PickleDataset
  filepath: "data/06_models/forecasting_model_${globals:model_to_use}.pkl"

energy_consumption_forecast:
  type: pandas.ParquetDataset
  filepath: "data/08_reporting/forecast_${globals:model_to_use}.pq"

if you read the error message, it is the first dataset that is referenced therein. Important to mention that kedro run works perfectly and beautifully (personally very happy with this improved approach to global configs wrt kedro 0.18).

Rashida Kanchwala

10/08/2024, 11:24 AM

thank you, can you share the version of kedro you are using. You mentioned it worked before; can you clarify if it stopped working after upgrading Kedro ?

Ionut Barbu

10/08/2024, 11:25 AM

by working before, i meant before we added the global config. we have kedro 0.19.5

Nok Lam Chan

10/08/2024, 11:27 AM

Can you move the

globals.yml

to, i.e.

conf/base/globals.yml

Ionut Barbu

10/08/2024, 12:56 PM

this fixes the jupyter notebook issue however makes running the pipeline with

kedro run

fail The end of the stack trace is

Copy code

File "C:\Users\IonutBarbu\miniconda3\envs\EIT-Epsilon\Lib\site-packages\kedro\config\omegaconf_config.py", line 398, in _get_globals_value
    raise InterpolationResolutionError(
omegaconf.errors.InterpolationResolutionError: Globals key 'model_to_use' not found and no default value provided.
    full_key: model_to_use
    object_type=dict

In the code it fails on the last code line below, when trying to load the parameters programmatically in the code

Copy code

# Initialize weather data extraction class based on given settings
conf_path = str(Path.cwd() / settings.CONF_SOURCE)
conf_loader = OmegaConfigLoader(conf_source=conf_path)
parameters = conf_loader["parameters"]

Nok Lam Chan

10/08/2024, 12:58 PM

Is this piece of code coming from your own code? Why do you need to create a separate config loader ?

Ionut Barbu

10/08/2024, 12:59 PM

we need to load the params programmatically in the code. We follow the guidelines here https://docs.kedro.org/en/stable/configuration/parameters.html#how-to-load-parameters-in-code

👀 1

Nok Lam Chan

10/08/2024, 12:59 PM

https://docs.kedro.org/en/stable/configuration/advanced_configuration.html#read-configuration It may be helpful to read this section of the configuration to understand how environment work

Nok Lam Chan

10/08/2024, 1:03 PM

I can see the confusion from that section. When you are creating the class directly, you need to make sure you pass in the environment as well (assuming you want the same behavior as your kedro project). By default it assume you are not working with any environment (think just one yaml file in a folder)

Nok Lam Chan

10/08/2024, 1:07 PM

You need to pass in the right environment to the class. default

base_env=base

default_run_env=local

. > # Initialize weather data extraction class based on given settings But soon you will find new issue, because this does not taken into account your runtime parameters, i.e.

kedro run --env <your_specific_env>

. I would challenge the idea that initialising a class base on parameters settings, do you really need to do this? This create a circular dependencies in pipeline and is much harder to reason about. The architecture overview can be found here: https://docs.kedro.org/en/stable/extend_kedro/architecture_overview.html If you want to use config loader, it would be easier to get this via hook instead of creating your own separate copy https://docs.kedro.org/en/stable/hooks/introduction.html

Ionut Barbu

10/08/2024, 5:04 PM

thanks a lot Nok! specifying the environments in the call to the OmegaConfig class have solved the problem also thank you for the additional suggestions regarding our current implementation for class initialization from param settings. I will have a look at the resources you have shared

👍🏼 1

👍 1

48 Views

Open in Slack

Previous Next