GraphSchema(
    nodes={
        'schema_validator': SchemaNode(needs={'importer': '__importer__'}, uses=<class 'rasa.graph_components.validators.default_recipe_validator.DefaultV1RecipeValidator'>, constructor_name='create', fn='validate', config={}, eager=False, is_target=False, is_input=True, resource=None),
        'finetuning_validator': SchemaNode(needs={'importer': 'schema_validator'}, uses=<class 'rasa.graph_components.validators.finetuning_validator.FinetuningValidator'>, constructor_name='create', fn='validate', config={'validate_core': True, 'validate_nlu': True}, eager=False, is_target=False, is_input=True, resource=None),
        'nlu_training_data_provider': SchemaNode(needs={'importer': 'finetuning_validator'}, uses=<class 'rasa.graph_components.providers.nlu_training_data_provider.NLUTrainingDataProvider'>, constructor_name='create', fn='provide', config={'language': 'en', 'persist': False}, eager=False, is_target=False, is_input=True, resource=None),
        'run_WhitespaceTokenizer0': SchemaNode(needs={'training_data': 'nlu_training_data_provider'}, uses=<class 'rasa.nlu.tokenizers.whitespace_tokenizer.WhitespaceTokenizer'>, constructor_name='load', fn='process_training_data', config={}, eager=False, is_target=False, is_input=False, resource=None),
        'train_RegexFeaturizer1': SchemaNode(needs={'training_data': 'run_WhitespaceTokenizer0'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.regex_featurizer.RegexFeaturizer'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'run_RegexFeaturizer1': SchemaNode(needs={'training_data': 'run_WhitespaceTokenizer0', 'resource': 'train_RegexFeaturizer1'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.regex_featurizer.RegexFeaturizer'>, constructor_name='load', fn='process_training_data', config={}, eager=False, is_target=False, is_input=False, resource=None),
        'train_LexicalSyntacticFeaturizer2': SchemaNode(needs={'training_data': 'run_RegexFeaturizer1'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.lexical_syntactic_featurizer.LexicalSyntacticFeaturizer'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'run_LexicalSyntacticFeaturizer2': SchemaNode(needs={'training_data': 'run_RegexFeaturizer1', 'resource': 'train_LexicalSyntacticFeaturizer2'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.lexical_syntactic_featurizer.LexicalSyntacticFeaturizer'>, constructor_name='load', fn='process_training_data', config={}, eager=False, is_target=False, is_input=False, resource=None),
        'train_CountVectorsFeaturizer3': SchemaNode(needs={'training_data': 'run_LexicalSyntacticFeaturizer2'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'run_CountVectorsFeaturizer3': SchemaNode(needs={'training_data': 'run_LexicalSyntacticFeaturizer2', 'resource': 'train_CountVectorsFeaturizer3'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer'>, constructor_name='load', fn='process_training_data', config={}, eager=False, is_target=False, is_input=False, resource=None),
        'train_CountVectorsFeaturizer4': SchemaNode(needs={'training_data': 'run_CountVectorsFeaturizer3'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer'>, constructor_name='create', fn='train', config={'analyzer': 'char_wb', 'min_ngram': 1, 'max_ngram': 4}, eager=False, is_target=True, is_input=False, resource=None),
        'run_CountVectorsFeaturizer4': SchemaNode(needs={'training_data': 'run_CountVectorsFeaturizer3', 'resource': 'train_CountVectorsFeaturizer4'}, uses=<class 'rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer.CountVectorsFeaturizer'>, constructor_name='load', fn='process_training_data', config={'analyzer': 'char_wb', 'min_ngram': 1, 'max_ngram': 4}, eager=False, is_target=False, is_input=False, resource=None),
        'train_DIETClassifier5': SchemaNode(needs={'training_data': 'run_CountVectorsFeaturizer4'}, uses=<class 'rasa.nlu.classifiers.diet_classifier.DIETClassifier'>, constructor_name='create', fn='train', config={'epochs': 100, 'constrain_similarities': True}, eager=False, is_target=True, is_input=False, resource=None),
        'train_EntitySynonymMapper6': SchemaNode(needs={'training_data': 'run_CountVectorsFeaturizer4'}, uses=<class 'rasa.nlu.extractors.entity_synonyms.EntitySynonymMapper'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'train_ResponseSelector7': SchemaNode(needs={'training_data': 'run_CountVectorsFeaturizer4'}, uses=<class 'rasa.nlu.selectors.response_selector.ResponseSelector'>, constructor_name='create', fn='train', config={'epochs': 100, 'constrain_similarities': True}, eager=False, is_target=True, is_input=False, resource=None),
        'domain_provider': SchemaNode(needs={'importer': 'finetuning_validator'}, uses=<class 'rasa.graph_components.providers.domain_provider.DomainProvider'>, constructor_name='create', fn='provide_train', config={}, eager=False, is_target=True, is_input=True, resource=None),
        'domain_for_core_training_provider': SchemaNode(needs={'domain': 'domain_provider'}, uses=<class 'rasa.graph_components.providers.domain_for_core_training_provider.DomainForCoreTrainingProvider'>, constructor_name='create', fn='provide', config={}, eager=False, is_target=False, is_input=True, resource=None),
        'story_graph_provider': SchemaNode(needs={'importer': 'finetuning_validator'}, uses=<class 'rasa.graph_components.providers.story_graph_provider.StoryGraphProvider'>, constructor_name='create', fn='provide', config={'exclusion_percentage': None}, eager=False, is_target=False, is_input=True, resource=None),
        'training_tracker_provider': SchemaNode(needs={'story_graph': 'story_graph_provider', 'domain': 'domain_for_core_training_provider'}, uses=<class 'rasa.graph_components.providers.training_tracker_provider.TrainingTrackerProvider'>, constructor_name='create', fn='provide', config={'debug_plots': False, 'augmentation_factor': 50}, eager=False, is_target=False, is_input=False, resource=None),
        'train_MemoizationPolicy0': SchemaNode(needs={'training_trackers': 'training_tracker_provider', 'domain': 'domain_for_core_training_provider'}, uses=<class 'rasa.core.policies.memoization.MemoizationPolicy'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'train_RulePolicy1': SchemaNode(needs={'training_trackers': 'training_tracker_provider', 'domain': 'domain_for_core_training_provider'}, uses=<class 'rasa.core.policies.rule_policy.RulePolicy'>, constructor_name='create', fn='train', config={}, eager=False, is_target=True, is_input=False, resource=None),
        'train_UnexpecTEDIntentPolicy2': SchemaNode(needs={'training_trackers': 'training_tracker_provider', 'domain': 'domain_for_core_training_provider'}, uses=<class 'rasa.core.policies.unexpected_intent_policy.UnexpecTEDIntentPolicy'>, constructor_name='create', fn='train', config={'max_history': 5, 'epochs': 100}, eager=False, is_target=True, is_input=False, resource=None),
        'train_TEDPolicy3': SchemaNode(needs={'training_trackers': 'training_tracker_provider', 'domain': 'domain_for_core_training_provider'}, uses=<class 'rasa.core.policies.ted_policy.TEDPolicy'>, constructor_name='create', fn='train', config={'max_history': 5, 'epochs': 100, 'constrain_similarities': True}, eager=False, is_target=True, is_input=False, resource=None)
        }
    )