tensorflow Pipeline 之TextLineReader 和decode_csv多分割替代方案

1、TextLineReader

reader = tf.TextLineReader()    
key, value = reader.read(filename_queue)    
record_defaults = [1,1,1,1,1]    
a,b,c,d,e =  tf.decode_csv(records=value,record_defaults=record_defaults, field_delim=" ")

2、decode_csv用法

1>、decode_csv的 field_delim只能1个
2>、可以多个decode_csv套用

    reader = tf.TextLineReader()
    key, value = reader.read(filename_queue)
    record_defaults = [[''], ['']]
    cols = tf.decode_csv(value, record_defaults, field_delim='\t')

    default = [[1.0]] * (lenth) 
    img = tf.decode_csv(cols[0], default, ',')
    label = tf.decode_csv(cols[1], [[0]] * 10, ',')

 

参考自 http://cn.voidcc.com/question/p-mdfqowwd-ts.html