df_btest[df_btest[labels] == 1])
print(t, v)
# #划分训练测试集
X_train_tra, X_test_tra, df_btest = data_seperate(df_train,
df_btest,
size=0.3,
label=labels,
cri=None,
undeal_column=None)
# # 划分label
print(df_train.columns)
# 划分label
x_train, y_train = seperate_label(X_train_tra, label=labels)
x_test, y_test = seperate_label(X_test_tra, label=labels)
print("x_train", x_train.shape)
#SBS
# sbs = SBS(rf, k_features=9,test_size=0.3, random_state=0,scoring=f1_score)
# sbs.fit(x.drop(["studentNo", "teacherId"], axis=1),y)
# k_feat = [len(k) for k in sbs.subsets_]
# import matplotlib.pyplot as plt
# plt.plot(k_feat, sbs.scores_, marker='o')
# plt.ylim([0.7, 1.1])
# plt.ylabel('Accuracy')
# plt.xlabel('Number of features')
# plt.grid()
# plt.tight_layout()
# # plt.savefig('./sbs.png', dpi=300)
df_train = df_voice[(df_voice["order_apply_time"] >= "2018-09-01") & (
df_voice["order_apply_time"] < "2019-01-10")].drop(["order_apply_time","content"], axis=1)
df_btest = df_voice[(df_voice["order_apply_time"] >= "2019-01-10") & (
df_voice["order_apply_time"] < "2019-01-15")].drop(["order_apply_time","content"], axis=1)
# # #划分训练测试集
X_train_tra, X_test_tra, df_btest = data_seperate(df_train, df_btest,
size=0.3,
label=labels,
cri=None,
undeal_column=None
)
# 划分label
x_trains, y_train = seperate_label(X_train_tra, label=labels)
x_tests, y_test = seperate_label(X_test_tra, label=labels)
x_btests, y_btest = seperate_label(df_btest, label=labels)
x_train = np.array(list(x_trains['sent']))
x_test = np.array(list(x_tests['sent']))
x_btest = np.array(list(x_btests['sent']))
print("x_train", x_train.shape)
model = Sequential()
model.add(Embedding(input_dim=len(dict), output_dim=32, input_length=maxlen))
# model.add(
# Embedding(input_dim=128, output_dim=64, input_length=maxlen))/