TensorFlow 1.15 CI error

[albertz]

...
Collecting protobuf>=3.6.1
  Downloading protobuf-4.21.9-cp37-abi3-manylinux2014_x86_64.whl (408 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 408.4/408.4 kB 87.0 MB/s eta 0:00:00
Requirement already satisfied: h5py in /home/runner/.local/lib/python3.7/site-packages (from keras-applications>=1.0.8->tensorflow==1.15.3) (3.7.0)
Collecting markdown>=2.6.8
  Downloading Markdown-3.4.1-py3-none-any.whl (93 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 93.3/93.3 kB 34.9 MB/s eta 0:00:00
Collecting werkzeug>=0.11.15
  Downloading Werkzeug-2.2.2-py3-none-any.whl (232 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 232.7/232.7 kB 63.7 MB/s eta 0:00:00
Requirement already satisfied: setuptools>=41.0.0 in /home/runner/.local/lib/python3.7/site-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow==1.15.3) (65.5.1)
Requirement already satisfied: importlib-metadata>=4.4 in /home/runner/.local/lib/python3.7/site-packages (from markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15.3) (5.0.0)
Collecting MarkupSafe>=2.1.1
  Downloading MarkupSafe-2.1.1-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (25 kB)
Requirement already satisfied: typing-extensions>=3.6.4 in /home/runner/.local/lib/python3.7/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15.3) (4.4.0)
Requirement already satisfied: zipp>=0.5 in /home/runner/.local/lib/python3.7/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15.3) (3.10.0)
Building wheels for collected packages: gast
  Building wheel for gast (setup.py): started
  Building wheel for gast (setup.py): finished with status 'done'
  Created wheel for gast: filename=gast-0.2.2-py3-none-any.whl size=7537 sha256=2572272d2b77ff5dba44a996df76f95664b74ff5cea0a5abcb3c833a37fa60aa
  Stored in directory: /home/runner/.cache/pip/wheels/cf/87/6f/3f34218ef184368cec9ee65bdfd65baf117811f0a0ce1263ff
Successfully built gast
Installing collected packages: tensorflow-estimator, wrapt, termcolor, six, protobuf, opt-einsum, MarkupSafe, gast, astor, absl-py, werkzeug, markdown, keras-preprocessing, keras-applications, grpcio, google-pasta, tensorboard, tensorflow
Successfully installed MarkupSafe-2.1.1 absl-py-1.3.0 astor-0.8.1 gast-0.2.2 google-pasta-0.2.0 grpcio-1.50.0 keras-applications-1.0.8 keras-preprocessing-1.1.2 markdown-3.4.1 opt-einsum-3.3.0 protobuf-4.21.9 six-1.16.0 tensorboard-1.15.0 tensorflow-1.15.3 tensorflow-estimator-1.15.1 termcolor-2.1.0 werkzeug-2.2.2 wrapt-1.14.1

Then:
While importing TensorFlow:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow/__init__.py", line 102, in <module>
    from tensorflow_core import *
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/__init__.py", line 28, in <module>
    from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import
  File "<frozen importlib._bootstrap>", line 1019, in _handle_fromlist
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow/__init__.py", line 50, in __getattr__
    module = self._load()
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow/__init__.py", line 44, in _load
    module = _importlib.import_module(self.__name__)
  File "/opt/hostedtoolcache/Python/3.7.15/x64/lib/python3.7/importlib/__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/python/__init__.py", line 52, in <module>
    from tensorflow.core.framework.graph_pb2 import *
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/graph_pb2.py", line 16, in <module>
    from tensorflow.core.framework import node_def_pb2 as tensorflow_dot_core_dot_framework_dot_node__def__pb2
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/node_def_pb2.py", line 16, in <module>
    from tensorflow.core.framework import attr_value_pb2 as tensorflow_dot_core_dot_framework_dot_attr__value__pb2
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/attr_value_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__pb2
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/tensor_pb2.py", line 16, in <module>
    from tensorflow.core.framework import resource_handle_pb2 as tensorflow_dot_core_dot_framework_dot_resource__handle__pb2
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/resource_handle_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_shape_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2
  File "/home/runner/.local/lib/python3.7/site-packages/tensorflow_core/core/framework/tensor_shape_pb2.py", line 42, in <module>
    serialized_options=None, file=DESCRIPTOR),
  File "/home/runner/.local/lib/python3.7/site-packages/google/protobuf/descriptor.py", line 560, in __new__
    _message.Message._CheckCalledFromGeneratedFile()
TypeError: Descriptors cannot not be created directly.
If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
If you cannot immediately regenerate your protos, some other possible workarounds are:
 1. Downgrade the protobuf package to 3.20.x or lower.
 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).

More information: https://developers.google.com/protocol-buffers/docs/news/20[22](https://github.com/rwth-i6/returnn/actions/runs/3440254625/jobs/5738513821#step:7:23)-05-06#python-updates

Via

[albertz]

Related:
https://stackoverflow.com/questions/72441758/typeerror-descriptors-cannot-not-be-created-directly
protocolbuffers/protobuf#10051

Actually seems to be exactly the same:
protocolbuffers/protobuf#10051 (comment)

Solution seems to be to just use protobuf==3.20.1 in this case.

[albertz]

Ok, fixed now.

pip install --user "protobuf<=3.20.1"

For all older TF versions.