Running visualize_predictions_sequences.ipynb

[berkiu]

Hello, i've got a problem while running sequence mode predictions

Traceback (most recent call last) ────────────────────────────────╮
│ in :21 │
│ │
│ 18 │ del pred │
│ 19 │
│ 20 uvt_p = torch.stack([p["uvr_max"] for p in preds]) │
│ ❱ 21 uvt_seq = torch.stack([p["uvr_max_seq"] for p in preds]) │
│ 22 xy_p = torch.stack([c.to_xy(uv) for c, uv in zip(canvas, uvt_p[:,:2])]) │
│ 23 xy_seq = torch.stack([c.to_xy(uv) for c, uv in zip(canvas, uvt_seq[:,:2])]) │
│ 24 logprobs = torch.stack([p["log_probs"] for p in preds]) │
│ │
│ in :21 │
│ │
│ 18 │ del pred │
│ 19 │
│ 20 uvt_p = torch.stack([p["uvr_max"] for p in preds]) │
│ ❱ 21 uvt_seq = torch.stack([p["uvr_max_seq"] for p in preds]) │
│ 22 xy_p = torch.stack([c.to_xy(uv) for c, uv in zip(canvas, uvt_p[:,:2])]) │
│ 23 xy_seq = torch.stack([c.to_xy(uv) for c, uv in zip(canvas, uvt_seq[:,:2])]) │
│ 24 logprobs = torch.stack([p["log_probs"] for p in preds]) │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
KeyError: 'uvr_max_seq'

It seems like model is not providing a key "uvr_max_seq" in predictions
If i run preds[0].keys() i got dict_keys(['map', 'pixel_scales', 'bev', 'scores', 'log_probs', 'uvr_max', 'uv_max', 'yaw_max', 'uvr_expectation', 'uv_expectation', 'yaw_expectation', 'features_image', 'features_bev', 'valid_bev', 'log_probs_seq', 'xyr_max_seq'])

[xiaofeifei-1]

Me,too. Can you solve it?

[martin-liao]

change the xyr_max_seq to uvr_max_seq in L16.