Inconsistent model even after using 'random_seed' in config

Rasa Open Source
1

Hi, I am training a model with the below config:

language: en
pipeline:
  - name: WhitespaceTokenizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: char_wb
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 100
    constrain_similarities: true
    random_seed: 42
  - name: EntitySynonymMapper
  - name: ResponseSelector
    epochs: 100
    constrain_similarities: true
    random_seed: 42
  - name: FallbackClassifier
    threshold: 0.1
    ambiguity_threshold: 0.05

policies:
  - name: MemoizationPolicy
    random_seed: 42
  - name: TEDPolicy
    max_history: 10
    epochs: 100
    constrain_similarities: true
    random_seed: 42
  - name: RulePolicy
    random_seed: 42

I get different results(trained models) on i7 and i9 machines. Any idea why?

Thanks!

2

Could you elaborate on how the results are different?