问题
I have cloned a github repository into google-colab [https://github.com/limbo018/DREAMPlace] and installed the necessary requirements.
When I run the main script (!python dreamplace/placer.py) from either the base repo directory or the cmake install directory in jupyter I receive a modulenotfound error. I am able to import the module from a jupyter cell.
The placer.py script imports it's parent module (dreamplace), which I have added to the path. I have a feeling colab/jupyter spawns a separate shell when I run the program through colab/jupyter. How can I ensure that the dreamplace module is added to the global path such that placer.py and other child-scripts can find it? I would prefer not having to edit the program, but if this is the best solution, it's no problem.
Here is a link to my colaboratory which includes my steps to set up the environment [it takes a couple minutes to complete]: https://colab.research.google.com/drive/1hS42ZegFZhjO28On1Hk_1muC1opasBxA
I tried to add the dreamplace directory to the path, but it's no use.
I would appreciate any advice or help on this issue. Thanks.
Note this issue has been mentioned in the repository: https://github.com/limbo018/DREAMPlace/issues/2
来源：https://stackoverflow.com/questions/56939487/google-colaboratory-local-module-modulenotfounderror