large amount of memory.

Rasa Open Source
1

UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(“gradients/cond_grad/Identity_1:0”, shape=(None,), dtype=int64), values=Tensor(“gradients/cond_grad/Identity:0”, shape=(None,), dtype=float32), dense_shape=Tensor(“gradients/cond_grad/Identity_2:0”, shape=(1,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.

2

Please read “How to ask a great question”:

  • Please explain:
    • How you encountered this problem
    • What you’ve already tried to solve it; any obstacles that prevented you from solving it yourself
    • Minimal steps to reproduce the issue or other context (e.g. deployment method)

Can you share your config file?

Also, this is a warning, not an error. You can ignore it if everything works fine later.

3

I’ve also been met with this warning, it appears 4 times in training the core model (not nlu).

I’m using a docker image based on rasa/rasa:2.8.12-full.

My config file looks like this:

# Configuration for Rasa NLU.
# https://rasa.com/docs/rasa/nlu/components/
language: en

pipeline:
# # No configuration for the NLU pipeline was provided. The following default pipeline was used to train your model.
# # If you'd like to customize it, uncomment and adjust the pipeline.
# # See https://rasa.com/docs/rasa/tuning-your-model for more information.
  - name: WhitespaceTokenizer
  - name: custom_components.SimpleNameExtractor
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: char_wb
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 100
    constrain_similarities: true
  - name: EntitySynonymMapper
  - name: ResponseSelector
    epochs: 100
    constrain_similarities: true
  - name: FallbackClassifier
    threshold: 0.3
    ambiguity_threshold: 0.1

# Configuration for Rasa Core.
# https://rasa.com/docs/rasa/core/policies/
policies:
# # No configuration for policies was provided. The following default policies were used to train your model.
# # If you'd like to customize them, uncomment and adjust the policies.
# # See https://rasa.com/docs/rasa/policies for more information.
  - name: MemoizationPolicy
  - name: RulePolicy
  - name: UnexpecTEDIntentPolicy
    max_history: 5
    epochs: 100
  - name: TEDPolicy
    max_history: 5
    epochs: 100
    constrain_similarities: true

This is pretty much the default for this version of rasa apart from the addition of the SimpleNameExtractor component.

4

This warning has appeared with the new tensorflow 2.6 version, not sure why and if it can cause any further issue, but you can train the same way.

2 Likes
5

I have the same problem. Does the problem affect the model inference speed? My inference speed is slower than other tensorflow version.