search

mit-han-lab / temporal-shift-module

[ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding
https://arxiv.org/abs/1811.08383
MIT License
2.07k stars 417 forks source link

As for how to train my dataset #191

Open dy-gdut opened 3 years ago

dy-gdut commented 3 years ago

As for how to train my data set, I have the folder of video frame and the label of the corresponding folder. How can I make train_videofolder.txt and val_videofolder.txt.

dy-gdut commented 3 years ago

Train_videofolder.txt and val_videofolder.txt I used the format 'image frame path, frame number, label', is this format correct?

dy-gdut commented 3 years ago

NaN or INF found in input tensor at runtime

melon-coder commented 3 years ago

Did you solve this problem? I have the same question

melon-coder commented 3 years ago

Train_videofolder.txt and val_videofolder.txt I used the format 'image frame path, frame number, label', is this format correct?

I think it should be like this "image frame path, frame number, label (for numbers, the category corresponds to the first few lines of the label file kinetics_label_map.txt)"

liumf001 commented 3 years ago

Did you two solve the problem?

sajjad-z commented 2 years ago

Did you three solve this problem?

can you share this txt files?