facebookresearch/dynalab

Issues

• submitting a different kind of model for sentiment analysis

  #131 opened 2 years ago by seinecle
  0

• `init.py` fails on Windows if `checkpoint.pt` or `handler.py` is in root path

  #114 opened 2 years ago by kokrui
  0

• tqdm is not installed with the pip install of dynalab

  #125 opened 3 years ago by TristanThrush
  0

• Add BART style NLI example handler

  #119 opened 3 years ago by Ishita-0112
  0

• Add BERT style NLI example handler

  #123 opened 3 years ago by fzchriha
  0

• Link the task IO documentation from the main README if not done already

  #120 opened 3 years ago by TristanThrush
  0

• `KeyError` when calling `get_task_submission_limit()`

  #116 opened 3 years ago by ktirumalafb
  0

• Could an example task.json be provided

  #113 opened 3 years ago by walterwang
  1

• The same model works successfully for small tracks but takendown for the full track

  #109 opened 3 years ago by BaohaoLiao
  1

• getting no model_files path specified issue

  #110 opened 3 years ago by nishantgupta08
  2

• Test failing after step 25/25

  #111 opened 3 years ago by bpopeters
  0

• Large model not evaluating on the devtest set

  #103 opened 3 years ago by jcblaisecruz02
  17

• Questions about the second and final submission period

  #108 opened 3 years ago by BaohaoLiao
  0

• Taken down for full track issue[hurry]

  #106 opened 3 years ago by haoyanghua
  1

• Successfully evaluated on dev set but failed on devtest set

  #107 opened 3 years ago by BaohaoLiao
  2

• Taken Down in the Dynabench for Large Track

  #90 opened 3 years ago by CSJianYang
  13

• Submit models with own extensions to fairseq

  #105 opened 3 years ago by dannigt
  1

• Showing "takendown" for my submission of flores task

  #98 opened 3 years ago by BaohaoLiao
  4

• Translate with ensemble models instead of one model

  #97 opened 3 years ago by BaohaoLiao
  3

• Different model sizes between pretrained model and finetuned model

  #99 opened 3 years ago by BaohaoLiao
  4

• The shown average score is different with the manual calculation of mean of individual scores

  #100 opened 3 years ago by BaohaoLiao
  5

• Failed to submit large model

  #102 opened 3 years ago by yingwaner
  2

• The status of the model is "takendown"

  #87 opened 3 years ago by yingwaner
  8

• Large Size Model > 2GB

  #91 opened 3 years ago by CSJianYang
  9

• Submission shows "Takendown" status

  #96 opened 3 years ago by jcblaisecruz02
  2

• Help with submission

  #93 opened 3 years ago by gwenzek
  5

• 504 Server Error: Gateway Time-out

  #94 opened 3 years ago by mazhiyi
  3

• The same model failed today, but worked before

  #101 opened 3 years ago by BaohaoLiao
  5

• Installing on py37 gives error AttributeError: type object 'Callable' has no attribute '_abc_registry'

  #73 opened 3 years ago by chauhang
  2

• What is in .git that prevents tarball from being parsed?

  #52 opened 3 years ago by mazhiyi
  1

• Submission Limit Exceeded Warning: Provide Additional Details

  #83 opened 3 years ago by maxbartolo
  5

• Better management of submitted files

  #89 opened 3 years ago by mazhiyi
  0

• Improved Reporting of Model Upload Status

  #85 opened 3 years ago by maxbartolo
  1

• Non-Leaderboard Datasets Leaderboard

  #86 opened 3 years ago by maxbartolo
  0

• Add a warning explaining that '_' is not allowed in filename (or general model naming convention guidelines)

  #81 opened 3 years ago by maxbartolo
  3

• Question about the dockerfile

  #77 opened 3 years ago by CSJianYang
  8

• [open source] pull dockerfiles from github file and remove install by --editable requirement

  #67 opened 3 years ago by mazhiyi
  0

• Double check to make sure that labels are unified with dynabench and the template handlers are ok

  #60 opened 3 years ago by TristanThrush
  4

• In verify_response, prevent extra keys too

  #53 opened 3 years ago by mazhiyi
  0

• Add prob key to response

  #54 opened 3 years ago by mazhiyi
  0

• Add some tips in how to read integrated test and torchserve log

  #59 opened 3 years ago by mazhiyi
  0

• Integrated test taking too much memory

  #58 opened 3 years ago by mazhiyi
  0

• dynalab init --rename

  #42 opened 3 years ago by mazhiyi
  0

• [be] use schema validation for verify_response

  #56 opened 3 years ago by mazhiyi
  0

• [BE][dynalab + dynabench]Use list in setup_config files

  #40 opened 3 years ago by mazhiyi
  0

• [optimization] [dynalab + dynabench] Remove checkpoint from the docker file

  #39 opened 3 years ago by mazhiyi
  0

• docker image name cannot contain upper case letters

  #43 opened 3 years ago by mazhiyi
  0

• [dynalab + dynabench] API to generate model response signature

  #35 opened 3 years ago by mazhiyi
  1

• docker build timeout when libraries in requirements.txt or setup.py are too big

  #32 opened 3 years ago by mazhiyi
  1

• Add a check before dynalab-cli upload really uploads to prevent accidental submission

  #34 opened 3 years ago by mazhiyi
  0