[Chainer]indexError: index 207 is out of bounds for axis 0 with size 5の解決法がわかりません

googlecolab上でChainerを用いて224枚のシカの画像を学習させ、その学習データを生成するプログラムを作成しているのですが、おそらく、L.Convolution2DとL.linearの使い方がいまいちわからず、不適切な値を入れているせいで以下のエラーが出てしまいます。どうしたらよろしいでしょうか。

エラー文
Exception in main training loop: index 218 is out of bounds for axis 0 with size 3
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/chainer/training/trainer.py", line 343, in run
update()
File "/usr/local/lib/python3.6/dist-packages/chainer/training/updaters/standard_updater.py", line 240, in update
self.update_core()
File "/usr/local/lib/python3.6/dist-packages/chainer/training/updaters/standard_updater.py", line 253, in update_core
optimizer.update(loss_func, *in_arrays)
File "/usr/local/lib/python3.6/dist-packages/chainer/optimizer.py", line 874, in update
loss = lossfun(*args, kwds) File "/usr/local/lib/python3.6/dist-packages/chainer/link.py", line 287, in call out = forward(*args, kwargs)
File "/usr/local/lib/python3.6/dist-packages/chainer/links/model/classifier.py", line 144, in forward
self.loss = self.lossfun(self.y, t)
File "/usr/local/lib/python3.6/dist-packages/chainer/functions/loss/softmax_cross_entropy.py", line 556, in softmax_cross_entropy
loss, = func.apply((x, t))
File "/usr/local/lib/python3.6/dist-packages/chainer/function_node.py", line 334, in apply
outputs = self.forward(in_data)
File "/usr/local/lib/python3.6/dist-packages/chainer/function_node.py", line 592, in forward
return self.forward_cpu(inputs)
File "/usr/local/lib/python3.6/dist-packages/chainer/functions/loss/softmax_cross_entropy.py", line 162, in forward_cpu
log_p = log_yd[t.ravel(), numpy.arange(t.size)]
Will finalize trainer extensions and updater before reraising the exception.

IndexError                                Traceback (most recent call last)
<ipython-input-1-3bf3422f7400> in <module>()
81 trainer.extend(extensions.PlotReport(['main/loss'],'epoch',file_name='loss.png'))
82 trainer.extend(extensions.PlotReport(['main/accuracy'],'epoch',file_name='accuracy.png'))
---> 83 trainer.run()
84 
85 serializer.save_npz("mymodel.npz",model)

/usr/local/lib/python3.6/dist-packages/chainer/functions/loss/softmax_cross_entropy.py in forward_cpu(self, inputs)
160         t_valid = t != self.ignore_label
161         t = t * t_valid
--> 162         log_p = log_yd[t.ravel(), numpy.arange(t.size)]
163 
164         log_p *= t_valid.ravel()

IndexError: index 218 is out of bounds for axis 0 with size 3

コード

import chainer
import glob
from itertools import chain
from chainer.datasets import LabeledImageDataset
from chainer import iterators,training,optimizers,datasets,serializers
from chainer.training import extensions,triggers
from chainer.dataset import concat_examples
import chainer.functions as F
import chainer.links as L

#-----MyChain-----

class MyChain(chainer.Chain):

    def __init__(self):
        super(MyChain,self).__init__()
        with self.init_scope():
            self.conv1 = L.Convolution2D(None,16,3,pad=2)
            self.conv2 = L.Convolution2D(None,32,3,pad=2)
            self.l3 = L.Linear(None,256)
            self.l4 = L.Linear(None,3)

    def __call__(self,x):
        h = F.max_pooling_2d(F.relu(self.conv1(x)),ksize=3,stride=2,pad=2)
        h = F.max_pooling_2d(F.relu(self.conv2(x)),ksize=3,stride=2,pad=2)
        h = F.dropout(F.relu(self.l3(h)))
        y = self.l4(h)
        return y


#-----img-----

train_path = glob.glob('img/Deers/Deers_train/*')

tr = 0
label = []
for index,item in enumerate(train_path):
    label.append(tr)
    tr = tr+1

pa_tuple = tuple(train_path)
la_tuple = tuple(label)

train_data = [(0,0)] * len(label)

tr = 0
for index,item in enumerate(train_path):
    train_data[tr] = [train_path[tr],label[tr]]
    tr = tr + 1

dataset_train = chainer.datasets.LabeledImageDataset(train_data)

from chainercv.transforms import resize
from chainer.datasets import TransformDataset

def transform(data):
    img,label = data
    img = resize(img,(500,500))
    return img,label

deerset_train = chainer.datasets.TransformDataset(dataset_train,transform)

#-----train-----

epoch = 10
batch = 5

model = L.Classifier(MyChain())
optimizer = optimizers.Adam()
optimizer.setup(model)

train_iter = iterators.SerialIterator(deerset_train,batch)
updater = training.StandardUpdater(train_iter,optimizer)
trainer = training.Trainer(updater,(epoch,'epoch'),out='result')

trainer.extend(extensions.dump_graph('main/loss'))
trainer.extend(extensions.snapshot(),trigger=(epoch,'epoch'))
trainer.extend(extensions.LogReport())
trainer.extend(extensions.PrintReport(['epoch','main/loss','main/accuracy']))
trainer.extend(extensions.ProgressBar())
trainer.extend(extensions.PlotReport(['main/loss'],'epoch',file_name='loss.png'))
trainer.extend(extensions.PlotReport(['main/accuracy'],'epoch',file_name='accuracy.png'))
trainer.run()

serializer.save_npz("mymodel.npz",model)