DeepLearningについて勉強しています。
NVIDIAのDIGITSを用いて学習、テストをしようと試みているのですが、
ネットワーク構築の時点でエラーが出て学習に進むことができません。
- 環境
OS:Ubuntu 16.04.2
CPU:Intel(R)Core(TM)i7-6850@3.60GHz
Memory:128GB
GPU:NVIDIA Quadro P6000(24GB)
DIGITSのバージョンは5.0.0
caffeは0.15.14です。
input: "data" input_shape{dim: 2 dim: 1 dim: 144 dim: 144 dim: 144} input: "label" input_shape{dim: 2 dim: 1 dim: 144 dim: 144 dim: 144} layer { name: "conv_in128_chan16" type: "Convolution" bottom: "data" top: "conv_in128_chan16" param { lr_mult: 1.0 decay_mult: 1.0 } param { lr_mult: 2.0 decay_mult: 0.0 } convolution_param { num_output: 16 kernel_size: 5 pad: 2 stride: 1 weight_filler { type: "msra" variance_norm: 2 } bias_filler { type: "constant" value: 0.0 } } } layer { name: "split1_1" type: "Split" bottom: "data" top: "t1_1" top: "t1_2" top: "t1_3" top: "t1_4" top: "t1_5" top: "t1_6" top: "t1_7" top: "t1_8" top: "t1_9" top: "t1_10" top: "t1_11" top: "t1_12" top: "t1_13" top: "t1_14" top: "t1_15" top: "t1_16" } layer { name: "concat1_1" bottom: "t1_1" bottom: "t1_2" bottom: "t1_3" bottom: "t1_4" bottom: "t1_5" bottom: "t1_6" bottom: "t1_7" bottom: "t1_8" bottom: "t1_9" bottom: "t1_10" bottom: "t1_11" bottom: "t1_12" bottom: "t1_13" bottom: "t1_14" bottom: "t1_15" bottom: "t1_16" top: "tiled1_1" type: "Concat" concat_param { axis: 1 } } layer { name: "addLayer1_1" type: "Eltwise" bottom: "conv_in128_chan16" bottom: "tiled1_1" top: "outBlock1_1" eltwise_param { operation: SUM } } layer { name: "outBlock1_1_RELU" type: "PReLU" bottom: "outBlock1_1" top: "outBlock1_1" } ##途中の畳み込み層などは割愛させていただきます。 layer { name: "reshapelab" type: "Reshape" bottom: "label" top: "label_flat" reshape_param { shape { dim: 0 # copy the dimension from below dim: 1 dim: 2985984 } } } layer { name: "reshaperes" type: "Reshape" bottom: "conv_in128_chan2_2_right" top: "conv_in128_chan2_right_flat" reshape_param { shape { dim: 0 # copy the dimension from below dim: 2 dim: 2985984 } } } layer { name: "softmax" type: "Softmax" bottom: "conv_in128_chan2_right_flat" top: "softmax_out" } layer { type: 'Python' name: 'loss' top: 'loss' bottom: 'softmax_out' bottom: "label_flat" python_param { # the module name -- usually the filename -- that needs to be in $PYTHONPATH module: 'pyLayer' # the layer name -- the class name in the module layer: 'DiceLoss' } # set loss weight so Caffe knows this is a loss layer. # since PythonLayer inherits directly from Layer, this isn't automatically # known to Caffe loss_weight: 1 }
エラー文は
Traceback (most recent call last):
File "digits/scheduler.py", line 508, in run_task
task.run(resources)
File "digits/task.py", line 188, in run
self.before_run()
File "digits/model/tasks/caffe_train.py", line 217, in before_run
self.save_files_generic()
File "digits/model/tasks/caffe_train.py", line 724, in save_files_generic
CaffeTrainTask.net_sanity_check(deploy_network, caffe_pb2.TEST)
File "digits/model/tasks/caffe_train.py", line 1635, in net_sanity_check
layer.name, bottom, "TRAIN" if phase == caffe_pb2.TRAIN else "TEST"))
CaffeTrainSanityCheckError: Layer 'reshapelab' references bottom 'label' at the TEST stage however this blob is not included at that stage. Please consider using an include directive to limit the scope of this layer.
です。
よろしくお願いいたします。
回答1件
あなたの回答
tips
プレビュー
バッドをするには、ログインかつ
こちらの条件を満たす必要があります。
2017/09/20 06:48