各位大佬，我最近在訓練mask rcnn的程序中，一直卡在
Starting at epoch 0. LR=0.001
Checkpoint Path: D:\daima\Mask_RCNN-master\logs\shapes20190515T2105\mask_rcnn_shapes_{epoch:04d}.h5
Selecting layers to train
fpn_c5p5 (Conv2D)
fpn_c4p4 (Conv2D)
fpn_c3p3 (Conv2D)
fpn_c2p2 (Conv2D)
fpn_p5 (Conv2D)
fpn_p2 (Conv2D)
fpn_p3 (Conv2D)
fpn_p4 (Conv2D)
In model: rpn_model
rpn_conv_shared (Conv2D)
rpn_class_raw (Conv2D)
rpn_bbox_pred (Conv2D)
mrcnn_mask_conv1 (TimeDistributed)
mrcnn_mask_bn1 (TimeDistributed)
mrcnn_mask_conv2 (TimeDistributed)
mrcnn_mask_bn2 (TimeDistributed)
mrcnn_class_conv1 (TimeDistributed)
mrcnn_class_bn1 (TimeDistributed)
mrcnn_mask_conv3 (TimeDistributed)
mrcnn_mask_bn3 (TimeDistributed)
mrcnn_class_conv2 (TimeDistributed)
mrcnn_class_bn2 (TimeDistributed)
mrcnn_mask_conv4 (TimeDistributed)
mrcnn_mask_bn4 (TimeDistributed)
mrcnn_bbox_fc (TimeDistributed)
mrcnn_mask_deconv (TimeDistributed)
mrcnn_class_logits (TimeDistributed)
mrcnn_mask (TimeDistributed)
C:\Users\86182\Anaconda3\envs\tenrorflow1\lib\site-packages\tensorflow\python\ops\gradients_impl.py:112: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.
"Converting sparse IndexedSlices to a dense Tensor of unknown shape. "
Epoch 1/10
下面無法顯示loss值的變化，我看了GPU，發現GPU已經滿了，應該是正在訓練的。
我把多執行緒改成了單執行緒，發現問題還在。
而且訓練一直停不下來，各位大佬能幫看一下，這是什么原因嗎
怎樣可以列印出loss值。
uj5u.com熱心網友回復：
是不是你的batch_size定的太大了，mrcnn/config.py里的原話：# Number of images to train with on each GPU. A 12GB GPU can typically
# handle 2 images of 1024x1024px.
# Adjust based on your GPU memory and image sizes. Use the highest
# number that your GPU can handle for best performance.
uj5u.com熱心網友回復：
我也遇到這個問題了，都除錯兩天了還沒弄好，不知道你解決這個問題了嗎。。。
uj5u.com熱心網友回復：
除錯的demo這個檔案嗎uj5u.com熱心網友回復：
我解決了，卸載keras2.0.8 換上2.1.6就可以跑了轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/119258.html
標籤：機器視覺
上一篇：大佬們，請問阿里云 oss 用多少 G 之內是免費的？
下一篇：求推薦一款離線md檔案編輯器