ShaoqingRen / SPP_net

SPP_net : Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
364 stars 237 forks source link

how to continue finetune the full-conection layers? #52

Closed wuqiangch closed 9 years ago

wuqiangch commented 9 years ago

@ShaoqingRen when i finetune the full-conection layers, it crashed . I have saved some finetune models. How can I continue to finetune it with the saved models?