请问下loadData()里面 Time = df.index报错是为什么啊？

[estimate123]

我使用的DCRNN下载下来的METR.h5文件，使用pandas对其进行读取，生成Time Embedding时，代码中TIME = df.index报错，如下： ssh://root@192.168.1.104:22/home/tank/anaconda3/envs/lpb/bin/python3.6 -u /home/tank/lxl/GMAN/GMAN-master/METR/analyzeData.py Traceback (most recent call last): File "/home/tank/lxl/GMAN/GMAN-master/METR/analyzeData.py", line 37, in print(df.index) File "/home/tank/anaconda3/envs/lpb/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 852, in repr attrs = self._format_attrs() File "/home/tank/anaconda3/envs/lpb/lib/python3.6/site-packages/pandas/core/indexes/datetimelike.py", line 381, in _format_attrs freq = self.freqstr File "/home/tank/anaconda3/envs/lpb/lib/python3.6/site-packages/pandas/core/indexes/extension.py", line 54, in fget result = getattr(self.data, name) File "/home/tank/anaconda3/envs/lpb/lib/python3.6/site-packages/pandas/core/arrays/datetimelike.py", line 1104, in freqstr return self.freq.freqstr AttributeError: 'numpy.bytes' object has no attribute 'freqstr'

Process finished with exit code 1

请问是我的数据集不对吗？还是我的Pandas版本(1.1.4)不对啊,为什么无法获取到这个index呢？万分感谢

[Coolgiserz]

我也遇到了类似的错误: AttributeError: 'numpy.bytes' object has no attribute 'delta'

[Hdjdnsj]

解决了吗？我也是