[BUG] Segmentation fault while importing TextPredictor

• I have checked that this bug exists on the latest stable version of AutoGluon
• and/or I have checked that this bug exists on the latest mainline of AutoGluon via source installation

Describe the bug

>>> from autogluon.text import TextPredictor
/home/cjj/anaconda3/envs/pytorch/lib/python3.9/site-packages/torchvision/io/image.py:11: UserWarning: Failed to load image Python extension: libc10_cuda.so: cannot open shared object file: No such file or directory
  warn(f"Failed to load image Python extension: {e}")
Segmentation fault (core dumped)

Expected behavior

import TextPredictor

To Reproduce

from autogluon.text import TextPredictor

Screenshots

Installed Versions

/home/cjj/anaconda3/envs/pytorch/lib/python3.9/site-packages/torchvision/io/image.py:11: UserWarning: Failed to load image Python extension: libc10_cuda.so: cannot open shared object file: No such file or directory
  warn(f"Failed to load image Python extension: {e}")
/home/cjj/anaconda3/envs/pytorch/lib/python3.9/site-packages/gluoncv/__init__.py:40: UserWarning: Both `mxnet==1.9.1` and `torch==1.10.2+cpu` are installed. You might encounter increased GPU memory footprint if both framework are used at the same time.
  warnings.warn(f'Both `mxnet=={mx.__version__}` and `torch=={torch.__version__}` are installed. '

INSTALLED VERSIONS
------------------
date                 : 2022-05-28
time                 : 08:05:29.672541
python               : 3.9.12.final.0
OS                   : Linux
OS-release           : 4.15.0-76-generic
Version              : #86-Ubuntu SMP Fri Jan 17 17:24:28 UTC 2020
machine              : x86_64
processor            : x86_64
num_cores            : 40
cpu_ram_mb           : 257841
cuda version         : 11.515.43.04
num_gpus             : 10
gpu_ram_mb           : [2886, 12195, 12195, 6636, 3766, 3766, 3766, 1598, 6930, 4332]
avail_disk_size_mb   : 1487138

autogluon.common     : 0.4.2b20220528
autogluon.core       : 0.4.2b20220528
autogluon.features   : 0.4.2b20220528
autogluon.tabular    : 0.4.1
autogluon.text       : 0.4.2b20220528
autogluon.vision     : 0.4.1
autogluon_contrib_nlp: 0.0.1
boto3                : 1.23.8
catboost             : 1.0.6
dask                 : 2021.11.2
distributed          : 2021.11.2
fairscale            : 0.4.6
fastai               : 2.5.6
gluoncv              : 0.11.0
lightgbm             : 3.3.2
matplotlib           : 3.5.2
networkx             : 2.8.2
nptyping             : 1.4.4
numpy                : 1.22.4
omegaconf            : 2.1.2
pandas               : 1.3.5
PIL                  : 9.0.1
psutil               : 5.8.0
pytorch_lightning    : 1.6.3
ray                  : 1.10.0
requests             : 2.27.1
scipy                : 1.7.3
sentencepiece        : None
setuptools           : 59.5.0
skimage              : 0.19.2
sklearn              : 1.0.2
smart_open           : 5.2.1
timm                 : 0.5.4
torch                : 1.10.2+cpu
torchmetrics         : 0.7.3
tqdm                 : 4.64.0
transformers         : 4.16.2
xgboost              : 1.4.2

[damirpolat]

Only pip3 install -U protobuf==3.18.1 worked for me.

[sxjscience]

@damirpolat Is it possible for you to help test if the following versions of protobuf will work? This can help us nail down the version cap.

pip3 install -U protobuf==3.20.1
pip3 install -U protobuf==3.20.0
pip3 install -U protobuf==3.19.4
pip3 install -U protobuf==3.18.1