I would like to learn from CSV using a Chainer and determine whether CSV for input is valid or invalid by 0 or 1.

Gender, "Vocational (1=Student, 2=Working, 3=Housewife), Age, Annual Income, Available Money, Date of Purchase, Time of Purchase, Item id previously viewed, "Season (0=Spring, 1=Summer, 2=Autumn, 3=Winter), Region (JIS), Points of Purchase, Outdoor Temperature, Climate (0=Cloud, 2=Rain), Dollar (Foreign Exchange, Yen), Yen, =Foreign Exchange, ( ( " ) 方法 ) , 送料, ネット, 口コミ (, 円 検索 , 1 ( 知った 外為 ) 2 " ,
1,3,23,400,3000000,1456758000,79200,2,0,1,0.03,16,1,103,103,0,1,1,0
1,2,34,1000,5000000,1468508400,68400,3,1,13,0.01,16,0,112,114,100,1,2,0
1,2,60,1000,5000000,1470150000,61200,2,1,13,0.05,18,0,103,112,100,2,1,1
2,1,22,400,300000,1470236400,61200,3,1,13,0.05,24,0,100,109,100,2,2,2
2,2,22,400,300000,1481641200,57600,2,3,13,0.05,3,2,101,100,0,1,1,1

1,3,23,400,3000000,1456758000,79200,2,0,1,0.03,16,1,103,103,0,1,1,0

#-*-coding:utf-8-*-
import numpy as np
import chain
from chain import cuda, function, gradient_check, report, training, utils, variable
from chain import data sets, iterators, optimizers, serializers
from chain import Link, Chain, ChainList
import chain.functions as F
import chain.links as L
from chain.training import extensions
import pandas aspd

mnist=pd.read_csv('./data/bough_history_utf8.csv')
mnist_data, mnist_label = np.split(mnist, [1], Axis=1)
x_train, x_test=np.split(mnist_data, [50000])
y_train, y_test=np.split(mnist_label, [50000])

x_train=np.array(x_train,dtype=np.float32)
y_train=np.array(y_train,dtype=np.int32)
x_test=np.array(x_test,dtype=np.float32)
y_test=np.array(y_test,dtype=np.int32)

print('x_train:'+str(x_train.shape))
print('y_train:'+str(y_train.shape))
print('x_test:'+str(x_test.shape))
print('y_test:'+str(y_test.shape))

# train someone a hundred times
train_iter=iterators.SerialIterator(x_train,batch_size=100)
test_iter=iterators.SerialIterator(x_test, batch_size=100, repeat=False, shuffle=False)

class MLP (chain.Chain):

    def_init__(self, n_units, n_out):
        super(MLP,self).__init__(
            # The size of the inputs to each layer will be inferred
            l1 = L. Linear (None, n_units), #n_in->n_units
            l2 = L. Linear (None, n_units), #n_units ->n_units
            l3 = L. Linear (None, n_out), #n_units ->n_out
        )

    def__call__(self, x):
        h1 = F.relu(self.l1(x))
        h2 = F.relu(self.l2(h1))
        return self.l3(h2)

model = L. Classifier (MLP(1000,1))

# Setup an optimizer
optimizer=chainer.optimizers.Adam()
optimizer.setup(model)

updater=training.StandardUpdater(train_iter, optimizer)
trainer=training.Trainer(updater, (20, 'epoch'), out='result')
trainer.extend(extensions.Evaluator(test_iter,model))
trainer.run()

x_train:(5,1)
y_train: (5,18)
x_test —(0,1)
y_test —(0,18)
Traceback (most recent call last):
  File "DynamicPricing.py", line 55, in <module>
    trainer.run()
  File"/usr/lib64/python 2.7/site-packages/chainer/training/trainer.py", line 266, in run
    update()
  File"/usr/lib64/python 2.7/site-packages/chainer/training/updater.py", line 170, in update
    self.update_core()
  File"/usr/lib64/python 2.7/site-packages/chainer/training/updater.py", line 189, in update_core
    optimizer.update(loss_func, in_var)
  File"/usr/lib64/python 2.7/site-packages/chainer/optimizer.py", line 392, in update
    loss=lossfun(*args,**kwds)
  File"/usr/lib64/python 2.7/site-packages/chainer/links/model/classifier.py", line61, in__call_
    assertlen(args)>=2
AssertionError