accuracy is 100% for 1st epoch 10itr/10batches. further it decreases to 0 or very less value about moco HOT 3 CLOSED

Accuracy of 100% at first few batched is also true for ImageNet and other datasets. This is because you are comparing the positive embedding with negative embedding and the remaining randomly initialized keys in the queue. So, your negative embedding easily gets classified.

Regarding your problem, seems like you have a very small dataset.

from moco.

The code release in this repo is meant for reproducing the results of research papers.

We help users who have trouble using the code to reproduce results in the paper, but in general we do not provide suggestions on other issues such as how to apply the method to a new task or dataset.

from moco.

I am trying to run code on my own data (apart from datasets mentioned in paper). During training, it is seen that accuracy is 100% for 1st epoch 1st 10 batches, however it is decreasing to 0 or comparatively very small value throughout further training. Also loss is increasing all time. Snapshot for reference.

During the first 10 iterations, the queue is nearly empty (queue size is 65535), so the training accuracy can be very large (easy classification task).

from moco.