training Squeezenet won't end!!

[Farima Fatahi]

Hi everybody,

I'm having trouble training the squeezenet. I used tiny-imagenet as my dataset. it consists of 100,000 training photos and 10,000 validation photos. I chose --iterations 1000 in 00_quantize_squeeznet.sh (although 2000 would'nt make any difference) and I'm working in CPU mode. each batch's accuracy is often 0 and it can't make quantized.prototxt and just starts testing and training repeatedly from the beginning. attached pictures are screenshots of my command line during the run of that bash file. can you please tell me what can be wrong in the whole process?

thanks in advance.

[tanvide...@gmail.com]

Hi Farima,

Can you please provide information on how you created the lmdb for training and test dataset from tiny-imagenet?

Thanks,
Tanvi