search

microsoft / SwinBERT

Research code for CVPR 2022 paper "SwinBERT: End-to-End Transformers with Sparse Attention for Video Captioning"
https://arxiv.org/abs/2111.13196
MIT License
238 stars 34 forks source link

evalcap preparation #37

Closed franciszchen closed 2 years ago

franciszchen commented 2 years ago

Thanks for sharing the amazing work. Here I encounter a problem that I have to run the script in conda envs, instead of dockers. From the setup.sh, maybe /evalcap/coco_caption and /evalcap/cider are required in the project, as the error information:

Traceback (most recent call last):
  File "/home/SwinBERT/src/tasks/run_caption_VidSwinBert.py", line 22, in <module>
    from src.evalcap.utils_caption_evaluate import evaluate_on_coco_caption
  File "/home/SwinBERT/src/evalcap/utils_caption_evaluate.py", line 13, in <module>
    from .coco_caption.pycocotools.coco import COCO
ModuleNotFoundError: No module named 'src.evalcap.coco_caption'

Is there any instruction to prepare coco_caption and cider? Thanks for your attention!

Space-Xun commented 1 year ago

hello, I want to run the project in conda envs too, but it seems that the eval source codes are not provided. Have you solved the problem? Thanks!

Markkk111 commented 1 year ago

Hi,Have you solved the problem? Thanks!