inhomogeneous shape error when running the immitation learning example

[Fisher-Wang]

My environment

• OS: Ubuntu 23.04
• GPU: RTX 3090
• CoppeliaSim: CoppeliaSim_Edu_V4_1_0_Ubuntu20_04
• Python: 3.11.4

Package         Version
--------------- -------
cffi            1.14.2
html-testRunner 1.2.1
Jinja2          3.1.3
MarkupSafe      2.1.5
natsort         8.4.0
numpy           1.26.4
pillow          10.2.0
pip             23.0.1
pycparser       2.21
pyquaternion    0.9.9
PyRep           4.1.0.3
rlbench         1.2.0
setuptools      66.1.1
wheel           0.38.4

How to reproduce

run

python examples/imitation_learning.py

i got this error

Traceback (most recent call last):
  File "/home/fs/cod/RLBench/examples/imitation_learning.py", line 39, in <module>
    demos = np.array(demos).flatten()
            ^^^^^^^^^^^^^^^
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (2,) + inhomogeneous part.
QObject::~QObject: Timers cannot be stopped from another thread
QMutex: destroying locked mutex

[hahans]

hello I meet the same error, I solved this problem by modifying this code: "demos = np.array(demos,dtype=object).flatten()" and then downloading the related linux dependency package:"sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev "