Activitynet results: report results on the val_2 split

minghangz / cpl

CPL: Weakly Supervised Temporal Sentence Grounding with Gaussian-based Contrastive Proposal Learning

56 stars 5 forks source link

Activitynet results: report results on the val_2 split #7

Open Xiyu-AI opened 9 months ago

Xiyu-AI commented 9 months ago

Hello, in experiments section of your paper: "we report our results on the val_2 split". But, in your code, I find that you use the test_set to evaluate the model. Should I revise the test_set to val_set in main_runner.py???