Remove unnecessary null pointer checks

[elfring]

An extra null pointer check is not needed in functions like the following.

• SentiClsDnn::global_destroy
• WordSegment::thread_destroy

[csaimd]

eval的时候有问题，
for data in test_reader():
# infer a batch
pred = exe.run(inference_program,
feed=utils.data2tensor(data, place),
fetch_list=fetch_targets,
return_numpy=True)
for i, val in enumerate(data):
class3_label, class2_label = utils.get_predict_label(pred[0][i, 1])
true_label = val[1]
if class2_label == true_label:
class2_acc += 1
if class3_label == true_label:
class3_acc += 1
if true_label == 1.0:
neu_count += 1

        total_count += len(data)

    class2_acc = class2_acc / (total_count - neu_count)
    class3_acc = class3_acc / total_count

为什么for循环进不去，直接跑到class2_acc = class2_acc / (total_count - neu_count)
class3_acc = class3_acc / total_count，然后除数为零，就出问题了

[ChinaLiuHao]

test_reader()里是否塞数据了呢？

[csaimd]

解决了，batchsize在测试的时候设置为1，然后就好了

[csaimd]

默认batchsize=256，但是测试的例子中只有200条数据

[elfring]

Will any redundant null pointer checks be deleted?