Cannot reproduce the results for 1-shot 5-way Mini-ImageNet about supervised-reptile HOT 10 OPEN

from supervised-reptile.

It seems the dataset is indeed incomplete (see the output below). However, it does not lose too many images (38392/38400, 9593/9600, 11996/12000). Thus, I wonder if there are some other reasons behind the performance.

/m/c/w/c/F/R/supervised-reptile/data/miniimagenet
find ./train -name '*.JPEG' | wc -l
38392
/m/c/w/c/F/R/supervised-reptile/data/miniimagenet
find ./val -name '*.JPEG' | wc -l
9593
/m/c/w/c/F/R/supervised-reptile/data/miniimagenet
find ./test -name '*.JPEG' | wc -l
11996

By the way, could you send me a copy of the data? Thank you very much!

from supervised-reptile.

See if 7f815bc fixes your problem. The command in the README didn't quite match the hyperparameters in the paper. The correct command is:

python -u run_miniimagenet.py --shots 1 --inner-batch 10 --inner-iters 8 --meta-step 1 --meta-batch 5 --meta-iters 100000 --eval-batch 5 --eval-iters 50 --learning-rate 0.001 --meta-step-final 0 --train-shots 15 --checkpoint ckpt_m15t

from supervised-reptile.

Thank you very much! Will try the new command with complete data.

from supervised-reptile.

@unixpickle By the way, 7f815bc changes --eval-batch 15 --train-shots 5 to --eval-batch 5 --train-shots 15. I could understand the change of eval-batch, which behaves the same as reducing the learning rate (to 1/3 of original one) in the inner loop during the evaluation stage. However, I wonder why train-shots should be increased. In fact, it seems train-shots=1 matches the best when considering the similarity between the training and the evaluation stage.

from supervised-reptile.

@xwjabc for whatever reason, we found that training on more "shots" helped Reptile's performance, probably because it allows you to take more diverse gradient steps during each inner-loop. Table 4 of Appendix A in the paper specifies the hyper-parameters, this included.

from supervised-reptile.

@unixpickle
Got it. Thank you for the clarification!

from supervised-reptile.

@unixpickle @xwjabc I had the same problem with the incomplete miniimagenet data downloaded from fetch_data.sh,most of folder is empty images. Could you send me a complete data？My email address is [email protected] ,thanks.

from supervised-reptile.

I used to have the dataset on my OpenAI machine and in the cloud. Unfortunately, I no longer have access to either copy. I'll see if I can find it sitting anywhere else, but I doubt I can.

from supervised-reptile.

What a pity! Thanks for your reply

from supervised-reptile.