Python3环境导入tensorflowjs时崩溃并抛出冗长错误

Gentoo Linux系统下，Python3环境执行import tensorflowjs时程序直接崩溃，该问题可稳定复现，输出的错误日志篇幅较长可读性差。日志开头的CUDA动态库加载警告属于无GPU场景下的正常提示，真正触发崩溃的是末尾的重复类型注册断言错误。
完整错误日志如下：

2022-06-30 07:54:28.109638: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-06-30 07:54:28.109660: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflowjs/__init__.py", line 21, in <module>
    from tensorflowjs import converters
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflowjs/converters/__init__.py", line 21, in <module>
    from tensorflowjs.converters.converter import convert
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflowjs/converters/converter.py", line 37, in <module>
    from tensorflowjs.converters import tf_saved_model_conversion_v2
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflowjs/converters/tf_saved_model_conversion_v2.py", line 42, in <module>
    import tensorflow_hub as hub
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_hub/__init__.py", line 88, in <module>
    from tensorflow_hub.estimator import LatestModuleExporter
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_hub/estimator.py", line 62, in <module>
    class LatestModuleExporter(tf.compat.v1.estimator.Exporter):
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/util/lazy_loader.py", line 58, in __getattr__
    module = self._load()
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/util/lazy_loader.py", line 41, in _load
    module = importlib.import_module(self.__name__)
  File "/usr/lib/python3.9/importlib/__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_estimator/__init__.py", line 8, in <module>
    from tensorflow_estimator._api.v1 import estimator
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_estimator/_api/v1/estimator/__init__.py", line 8, in <module>
    from tensorflow_estimator._api.v1.estimator import experimental
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_estimator/_api/v1/estimator/experimental/__init__.py", line 8, in <module>
    from tensorflow_estimator.python.estimator.canned.dnn import dnn_logit_fn_builder
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/canned/dnn.py", line 27, in <module>
    from tensorflow_estimator.python.estimator import estimator
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 31, in <module>
    from tensorflow.python.checkpoint import checkpoint as trackable_util
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/checkpoint/checkpoint.py", line 32, in <module>
    from tensorflow.python.checkpoint import graph_view as graph_view_lib
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/checkpoint/graph_view.py", line 19, in <module>
    from tensorflow.python.checkpoint import trackable_view
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/checkpoint/trackable_view.py", line 20, in <module>
    from tensorflow.python.trackable import converter
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/trackable/converter.py", line 23, in <module>
    from tensorflow.python.trackable import data_structures
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/trackable/data_structures.py", line 1103, in <module>
    revived_types.register_revived_type(
  File "/home/moth/.local/lib/python3.9/site-packages/tensorflow/python/saved_model/revived_types.py", line 133, in register_revived_type
    raise AssertionError(f"Duplicate registrations for type '{identifier}'")
AssertionError: Duplicate registrations for type 'trackable_dict_wrapper'

从安装路径能看出，问题根源是混用了Portage系统包管理器安装的Python包、和pip带--user参数安装到用户目录的tensorflow相关组件。tensorflow、tensorflow-estimator、tensorflow-hub、tensorflowjs四个包有严格的版本对应要求，跨安装源混装会导致版本错位，同一个可追踪对象类型被重复注册，最终触发断言崩溃。

• 先彻底清理所有残留的tensorflow相关包，避免版本冲突：
  清理pip用户安装的相关包：pip uninstall -y tensorflow tensorflow-cpu tensorflow-estimator tensorflow-hub tensorflowjs
  清理Portage安装的相关包：sudo emerge -C dev-python/tensorflow dev-python/tensorflow-estimator dev-python/tensorflow-hub dev-python/tensorflowjs
• 二选一选择统一的安装渠道，禁止混装系统包和pip用户包：
  • 方案1：全用Portage安装，由包管理器自动匹配依赖版本：sudo emerge dev-python/tensorflowjs
  • 方案2：用Python虚拟环境隔离安装，不修改系统全局包：

    python -m venv ~/tfjs-venv
    source ~/tfjs-venv/bin/activate
    pip install --upgrade pip
    pip install tensorflowjs
    

• 日志开头的libcudart.so.11.0加载警告无需处理，没有配置Nvidia GPU的机器出现该提示属于正常情况，不影响CPU模式下的功能使用。

内容的提问来源于stack exchange，提问作者moth