Edward_wang
Edward_wang
hi , when using the pretrained weights file (ssdmobilenetv2) shared in google drive , the train-loss value has been reduced while the val-loss value is almost constant . after testing...
https://drive.google.com/file/d/1zts95H6A4-lrRD0HAvLH2oNHA8CgU6m1/view?usp=sharing
https://github.com/Adamdad/Object-Detection-Metrics
i would be glad to receive your answer , thx!
the code is almost the same as your repo , maybe there is something diffierent in SSD_training.ipnb . Perhaps you can find something wrong by running the training code.
https://drive.google.com/drive/folders/1c4XUoHl1q9PPxCCG94m4A7sjI2H8XnJo?usp=sharing
@tanakataiki i am sorry , the link above should be useful .
hi,i just train my model SSD300shufflnetV2 by following your strategy in SSD_train.ipynb ,but it seems hard to converge to a low loss (loss=4) . is there something wrong? thanks for...
could you tell me mAP on your dataset?