jzxwang
commented
3 years ago hi, there, I would like to reproduce this task and face the error above. I google that protocol, google said protocol 5 only python 3.8 above is supported, but if I use python3.8, the TensorFlow version can not be supported.
Traceback (most recent call last): File "task1b.py", line 692, in <module> sys.exit(main(sys.argv)) File "task1b.py", line 233, in main overwrite=overwrite File "/mnt/batch/tasks/shared/LS_root/mounts/clusters/at007484/code/Users/dcase2020_task1_baseline-master/task1a.py", line 637, in do_learning init_parameters=init_parameters, File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/processors/processing_chain.py", line 359, in push_processor 'output_type': output_type, File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/processors/processing_chain.py", line 27, in __init__ self.init_processor_class() File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/processors/processing_chain.py", line 54, in init_processor_class self.processor_class = eval('processor_module.' + processor_name.split('.')[-1])(**processor_init_parameters) File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/processors/data.py", line 448, in __init__ normalizer = Normalizer().load(filename=filename) File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/containers/containers.py", line 64, in load self.__dict__.update(Serializer.load_cpickle(filename=self.filename)) File "/anaconda/envs/tf-dcase-old/lib/python3.7/site-packages/dcase_util/files/serialization.py", line 122, in load_cpickle return pickle.load(open(filename, "rb")) ValueError: unsupported pickle protocol: 5