下面是我正在使用的代码。请让我知道为什么我的验证和培训准确性这么低？验证精度仅为0.0000e + 00，并且培训精度约为37％。可能出了什么问题？我的训练集有10500行和172列我的测试集有3150行和172列我的第一列是响应（类），因此我仅将其用作Y，其余列用作X。我的响应为3类：默认，低频和射频
from __future__ import print_function
import numpy as np
import pandas
from keras.models import Sequential
from keras.layers.core import Dense, Activation
from keras.optimizers import SGD
from keras.utils import np_utils
from sklearn.preprocessing import LabelEncoder
np.random.seed(1671)
NB_EPOCH = 5
BATCH_SIZE = 128
VERBOSE = 1
NB_CLASSES = 3
OPTIMIZER = SGD()
N_HIDDEN = 128
VALIDATION_SPLIT=0.1
RESHAPED = 171
dataframe_train = pandas.read_csv("TrainingEdgesToAction.csv", header=None)
dataset_train = dataframe_train.values
X_train = dataset_train[1:,1:172].astype(float)
#X_train = dataset_train[1:,0:172]
Y_train = dataset_train[1:,0]
dataframe_test = pandas.read_csv("TestingEdgesToAction.csv", header=None)
dataset_test = dataframe_test.values …