前提・実現したいこと
Chainerを使って画像を二種類に分類したいのですが、エラーが出てしましい分類できません。
作業は以下のように行いました。
まず、自分でイメージスキャナから得た19872枚の画像にデータセット制作のコードを実行することでラベルを振りました。
次に、以下のChainerのコードを実行した時に、trainer.run()でエラーが発生しました。
エラー文で示されているin_types[0]はデータの値なのか、ネットワークの変数なのかどちらでしょうか。
修正する箇所を教えていただけるとありがたいです。
発生している問題・エラーメッセージ
Exception in main training loop: Invalid operation is performed in: Convolution2DFunction (Forward) Expect: in_types[0].dtype.kind == f Actual: O != f Traceback (most recent call last): File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 306, in run update() File "/usr/local/lib/python2.7/dist-packages/chainer/training/updaters/standard_updater.py", line 149, in update self.update_core() File "/usr/local/lib/python2.7/dist-packages/chainer/training/updaters/standard_updater.py", line 160, in update_core optimizer.update(loss_func, *in_arrays) File "/usr/local/lib/python2.7/dist-packages/chainer/optimizer.py", line 640, in update loss = lossfun(*args, **kwds) File "chainclass.py", line 25, in __call__ return F.softmax_cross_entropy(self.fwd(x),t) File "chainclass.py", line 28, in fwd h1 = F.max_pooling_2d(F.relu(self.cn1(x)),2) File "/usr/local/lib/python2.7/dist-packages/chainer/links/connection/convolution_2d.py", line 175, in __call__ groups=self.groups) File "/usr/local/lib/python2.7/dist-packages/chainer/functions/connection/convolution_2d.py", line 582, in convolution_2d y, = fnode.apply(args) File "/usr/local/lib/python2.7/dist-packages/chainer/function_node.py", line 243, in apply self._check_data_type_forward(in_data) File "/usr/local/lib/python2.7/dist-packages/chainer/function_node.py", line 328, in _check_data_type_forward self.check_type_forward(in_type) File "/usr/local/lib/python2.7/dist-packages/chainer/functions/connection/convolution_2d.py", line 58, in check_type_forward w_type.ndim == 4, File "/usr/local/lib/python2.7/dist-packages/chainer/utils/type_check.py", line 524, in expect expr.expect() File "/usr/local/lib/python2.7/dist-packages/chainer/utils/type_check.py", line 482, in expect '{0} {1} {2}'.format(left, self.inv, right)) Will finalize trainer extensions and updater before reraising the exception. Traceback (most recent call last): File "chainclass.py", line 50, in <module> trainer.run() File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 320, in run six.reraise(*sys.exc_info()) File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 306, in run update() File "/usr/local/lib/python2.7/dist-packages/chainer/training/updaters/standard_updater.py", line 149, in update self.update_core() File "/usr/local/lib/python2.7/dist-packages/chainer/training/updaters/standard_updater.py", line 160, in update_core optimizer.update(loss_func, *in_arrays) File "/usr/local/lib/python2.7/dist-packages/chainer/optimizer.py", line 640, in update loss = lossfun(*args, **kwds) File "chainclass.py", line 25, in __call__ return F.softmax_cross_entropy(self.fwd(x),t) File "chainclass.py", line 28, in fwd h1 = F.max_pooling_2d(F.relu(self.cn1(x)),2) File "/usr/local/lib/python2.7/dist-packages/chainer/links/connection/convolution_2d.py", line 175, in __call__ groups=self.groups) File "/usr/local/lib/python2.7/dist-packages/chainer/functions/connection/convolution_2d.py", line 582, in convolution_2d y, = fnode.apply(args) File "/usr/local/lib/python2.7/dist-packages/chainer/function_node.py", line 243, in apply self._check_data_type_forward(in_data) File "/usr/local/lib/python2.7/dist-packages/chainer/function_node.py", line 328, in _check_data_type_forward self.check_type_forward(in_type) File "/usr/local/lib/python2.7/dist-packages/chainer/functions/connection/convolution_2d.py", line 58, in check_type_forward w_type.ndim == 4, File "/usr/local/lib/python2.7/dist-packages/chainer/utils/type_check.py", line 524, in expect expr.expect() File "/usr/local/lib/python2.7/dist-packages/chainer/utils/type_check.py", line 482, in expect '{0} {1} {2}'.format(left, self.inv, right)) chainer.utils.type_check.InvalidType: Invalid operation is performed in: Convolution2DFunction (Forward) Expect: in_types[0].dtype.kind == f Actual: O != f
該当のソースコード
Python
1 2import numpy as np 3import math 4import random 5import chainer 6from chainer import cuda,Function,report,training,utils,Variable 7from chainer import datasets,iterators,optimizers,serializers 8from chainer import Link,Chain,ChainList 9import chainer.functions as F 10import chainer.links as L 11from chainer.training import extensions 12from chainer.datasets import tuple_dataset 13 14class MyChain(Chain): 15 def __init__(self): 16 super(MyChain,self).__init__( 17 cn1 = L.Convolution2D(3,30,9,stride=1,pad=0), 18 cn2 = L.Convolution2D(30,48,9,stride=1,pad=1), 19 cn3 = L.Convolution2D(48,60,9,stride=1,pad=0), 20 l1 = L.Linear(2160,100), 21 l2 = L.Linear(100,2), 22 ) 23 24 def __call__(self,x,t): 25 return F.softmax_cross_entropy(self.fwd(x),t) 26 27 def fwd(self,x): 28 h1 = F.max_pooling_2d(F.relu(self.cn1(x)),2) 29 h2 = F.max_pooling_2d(F.relu(self.cn2(h1)),2) 30 h3 = F.max_pooling_2d(F.relu(self.cn3(h2)),2) 31 h4 = F.relu(self.l1(h3)) 32 return self.l2(h4) 33 34model = MyChain() 35optimizer = optimizers.Adam() 36optimizer.setup(model) 37 38validation_data = np.load('validation.npz') 39 40data = Variable(np.array(validation_data['x'],dtype=np.float32)) 41label = Variable(np.array(validation_data['t'],dtype=np.float32)) 42 43tuple_data = tuple_dataset.TupleDataset(data,label) 44 45iterator = iterators.SerialIterator(tuple_data,132) 46updater = training.StandardUpdater(iterator,optimizer) 47trainer = training.Trainer(updater,(10,'epoch')) 48trainer.extend(extensions.ProgressBar()) 49 50trainer.run() 51
試したこと
ここに問題に対して試したことを記載してください。
補足情報(FW/ツールのバージョンなど)
ここにより詳細な情報を記載してください。
あなたの回答
tips
プレビュー