francescotonini/human-gaze-target-detection-transformer

Issues

• The issue of model accuracy result

  #5 opened 6 months ago by Y445n9
  1

• Missing key(s) in state_dict: "gaze_watch_outside_embed.layers.0.weight", "gaze_watch_outside_embed.layers.0.bias". size mismatch for backbone.query_embed.weight: copying a param with shape torch.Size([20, 256]) from checkpoint, the shape in current model is torch.Size([64, 256]).

  #6 opened 6 months ago by weiyingxue
  1

• Parameters in weights file do not match

  #3 opened a year ago by chenxi-Guo
  4

• Regarding the issue of model accuracy

  #4 opened 10 months ago by chenwenhe11
  3

• model in paper

  #2 opened a year ago by chenxi-Guo
  1