meaning of val_log result

[kinfeparty]

Hello I got some result on the dataset.

topn: {'top1_10NN': 0.080424786} I know it's not the accuracy. I read your paper,but don't know the meaning of Nearest neighbor validation performances.

What does it mean?

[chengxuz]

This nearest neighbor validation is done by getting 5 clips from one validation video, finding 10 nearest neighbor for each clip separately in the memory bank of the training videos, combining the lables of these 10 neighbors in a weighted manner (see Instance Discrimination paper for details) to get class probabilities for one clip, and finally averaging the class prababilities across all 5 clips chosen.

[chengxuz]

Just added this explanation into the document.

[kinfeparty]

And may I know the result of transfer learning result? such as top1_5,top1_7. wha's the meaning of the 5 or 7?

[chengxuz]

ResNet-18 has 9 "layers": the first convolution-pooling layer, and the remaining eight residual blocks. top1_5 means reading out from 5th layer (CONV3 in the paper), top1_7 means reading out from 7th layer (CONV4 in the paper). You should also see a top1_9, that's reading out from the top layer (CONV5 in the paper). I will also add this to the document, thanks for asking.

[kinfeparty]

Thanks for the answering.