declare-lab/MELD

Issues

• Label indices do not match class weights or paper notation

  #51 opened 5 months ago by florianblume
  3

• Video file corrupted: train_splits/dia125_utt3.mp4

  #39 opened 3 years ago by tnlin
  3

• Train_set video dia125_utt3 can't be opened correctly

  #52 opened 4 months ago by ShockCloud10
  14

• Baseline results=0

  #36 opened 3 years ago by Aidenfaustine
  1

• class weight {0: 4.0, 1: 15.0, 2: 15.0, 3: 3.0, 4: 1.0, 5: 6.0, 6: 3.0}, it still is zero

  #37 opened 3 years ago by Aidenfaustine
  1

• Wrong video files

  #20 opened 4 years ago by ATriantafyllopoulos
  1

• BC-LSTM text unimodal checkpoint is broken

  #38 opened 3 years ago by SeanXiaoyuSun
  1

• How do I convert a video to the data format required for this model?

  #42 opened 2 years ago by She-yh
  1

• Is there any face region information in videos?

  #48 opened a year ago by ByeongjunCho
  1

• How can I know which face on the video sample is speaking to extract my own visual features ??

  #50 opened a year ago by david-gimeno
  1

• How to get the audio

  #53 opened 4 months ago by jasonppy
  0

• Absence of Audio in the Test File

  #49 opened a year ago by ClarenceKiva
  1

• No audio modal in the test set.

  #47 opened 2 years ago by Jinjikiko
  0

• Really poor results using the given features

  #46 opened 2 years ago by rodaslemos
  0

• How do you divide the dataset?

  #45 opened 2 years ago by mrFocusXin
  0

• requirements.txt missing

  #34 opened 4 years ago by laxmena
  2

• audio features

  #11 opened 5 years ago by Mazagov
  1

• Using pre-trained models with my own audio/video files.

  #16 opened 5 years ago by monilshah98
  2

• data preprocessing code of MELD

  #18 opened 4 years ago by Yuri-Kim
  1

• ValueError: `class_weight` not supported for 3+ dimensional targets. with class_weight

  #43 opened 2 years ago by s1guma
  0

• About the audio data

  #41 opened 2 years ago by 1511827481zzh
  0

• missing dialog 110 utterance 7

  #27 opened 4 years ago by jgabriel98
  1

• Cannot download the raw data.

  #40 opened 2 years ago by YuichiroKasuga
  1

• Baseline results

  #10 opened 5 years ago by stdoo
  9

• ValueError: "input_length" is 33, but received input has shape (None, 50)

  #15 opened 5 years ago by ajinkyabedekar
  1

• Data download link

  #33 opened 4 years ago by eonglints
  1

• Unable to download the Raw Data from the website.

  #17 opened 4 years ago by manish-agni
  1

• error: bad character range \|-t at position 12

  #35 opened 4 years ago by Aidenfaustine
  0

• Mismatch in data_emotion.p

  #32 opened 4 years ago by yujianll
  0

• Videos are not well-aligned with the texts.

  #30 opened 4 years ago by tae898
  0

• About visual_features.tar.gz

  #29 opened 4 years ago by ffcarina
  0

• Absense of No.60 dialogue in data/MELD/train.csv

  #25 opened 4 years ago by mianzhang
  0

• Error on the function method test_model

  #23 opened 4 years ago by tae898
  1

• Missing video in dev set and utility of additional videos

  #22 opened 4 years ago by youcaiSUN
  2

• FileNotFoundError: [Errno 2] No such file or directory: './data/pickles/data_emotion.p'

  #21 opened 4 years ago by elyorman
  1

• Use of the dataset for training models in relation to the license

  #19 opened 5 years ago by AngledLuffa
  4

• The Dataloader class is not getting loaded

  #14 opened 5 years ago by saanilandhare
  0

• Using pre-trained models with our own audio files.

  #12 opened 5 years ago by vsakkas96
  2

• baseline.py not working

  #13 opened 5 years ago by tamrakars
  1

• guidance for developing a LSTM model

  #8 opened 6 years ago by Thani011
  1

• Multiple issues in the dataset.

  #9 opened 6 years ago by saxenarohit
  1

• Failed to load pretrained models

  #7 opened 6 years ago by KuzmaNg
  4

• About the sequence length and sentence length.

  #6 opened 6 years ago by KuzmaNg
  2

• missing training data

  #5 opened 6 years ago by ChenJiaDong9219
  6

• The meanings about the features.

  #4 opened 6 years ago by XinyuLyu
  1

• Error in train.tar.gz

  #3 opened 6 years ago by saxenarohit
  2

• Hello

  #1 opened 6 years ago by spkll
  3