search

tyagi-iiitv / PointPillars

GNU General Public License v3.0
105 stars 47 forks source link

Failed to find data adapter that can handle input: <class 'processors.SimpleDataGenerator'>, <class 'NoneType'> #45

Open Gurk97 opened 2 years ago

Gurk97 commented 2 years ago

Hi!

I have been trying to launch the "point_pillars_training_run.py" file using python 3.8 on cuda 11.6. I have followed the instructions strictly (I belive). Now i am using a conda enviroment but i recieve the same error while on an virtual enviroment. I download the KITTY files and unpack them in the given directory structure. And run the "python setup.py install" without any issues or warnings.

i recieve the following problem when running "python point_pillars_training_run.py":

super(Adam, self).init(name, **kwargs) Traceback (most recent call last): File "point_pillars_training_run.py", line 58, in pillar_net.fit(training_gen, File "/home/gustavni/anaconda3/envs/py38/lib/python3.8/site-packages/keras/utils/traceback_utils.py", line 67, in error_handler raise e.with_traceback(filtered_tb) from None File "/home/gustavni/anaconda3/envs/py38/lib/python3.8/site-packages/keras/engine/data_adapter.py", line 984, in select_data_adapter raise ValueError( ValueError: Failed to find data adapter that can handle input: <class 'processors.SimpleDataGenerator'>, <class 'NoneType'>

Any tips on how to get the code running?

Kind regards

EricCurl commented 2 years ago

I overcame this issue, and I believe it was a simple fix. In the processors.py script, just change:

from tensorflow.python.keras.utils.data_utils import Sequence to from tensorflow.keras.utils import Sequence