Skip to content
GitLab
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • D DeepPavlov
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 18
    • Issues 18
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 22
    • Merge requests 22
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Packages and registries
    • Packages and registries
    • Package Registry
    • Infrastructure Registry
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • DeepPavlov
  • DeepPavlov
  • Issues
  • #978
Closed
Open
Issue created Aug 29, 2019 by Andrei Glinskii@glinskii.avDeveloper

Hard-coded context length in DAM ranking model?

Created by: kauttoj

I'm training a DAM neural ranking model following the example for Ubuntu V2 dataset. I want to repeat the same analysis with my own data with varying context length size. However, I think there is a problem in line 99 of "deep_attention_matching_network.py". The call

super(DAMNetwork, self).__init__(*args, **kwargs)

does not allow custom context lengths, but always reverts to the default 10 (in "tf_base_matching_model.py" line 46). The correct behavior (at least no errors) is achieved by replacing above call with:

super(DAMNetwork, self).__init__(num_context_turns=num_context_turns,*args, **kwargs)

Could you check if this is a bug in the code. Or maybe there is something missing in "ranking_ubuntu_v2_mt_word2vec_dam.json" config file.

Assignee
Assign to
Time tracking