sirish-gambhira
commented
2 years ago Hello @ssbilakeri.
I am also trying to run the DyGLIP pipeline in an end-to-end manner on MCT Dataset. I generated the bounding boxes, re-id features, maskrcnn features using the Step 1. However, when trying to run Step 2 (test_emb_to_mot.py) to generate merged bounding boxes across multiple cameras, I am facing few issues. I also noticed that trained checkpoint files are missing in the repo. I wanted to know if you were able to run the evaluation pipeline. If so, it would be great if you could help me in this regard.
Tagging @truongthanhdat @qkgia @pha-nguyen for better reach.
Thank you for your time.
during evaluation im facing key error ad trace back is here, i ran the code without any changes
Traceback (most recent call last): File "graph_baseline.py", line 210, in
main(args)
File "graph_baseline.py", line 183, in main
summary = compare_dataframes(gt, ts)
File "/home/mca/DyGLIP-master/Step3_Matching/motmetrics/apps/eval_motchallenge.py", line 73, in compare_dataframes
accs.append(mm.utils.compare_to_groundtruth(gts[k], tsacc, 'iou', distth=0.5))
File "/home/mca/DyGLIP-master/Step3_Matching/motmetrics/utils.py", line 185, in compare_to_groundtruth
fid_to_fgt = dict(iter(gt.groupby('FrameId')))
File "/home/mca/anaconda3/lib/python3.8/site-packages/pandas/core/frame.py", line 7712, in groupby
return DataFrameGroupBy(
File "/home/mca/anaconda3/lib/python3.8/site-packages/pandas/core/groupby/groupby.py", line 882, in init
grouper, exclusions, obj = get_grouper(
File "/home/mca/anaconda3/lib/python3.8/site-packages/pandas/core/groupby/grouper.py", line 882, in get_grouper
raise KeyError(gpr)
KeyError: 'FrameId'