Greate short course but terrible colab code, too much bugs and so many details (class, api etc) non-transparent. Can Sharon Zhou check the notebook code again?
1 Like
Hi @SBFALLOUT,
Thank you for your feedback.
Firstly, by colab, do you mean Jupyter notebook ? I’m asking this, because the platform doesn’t use any colabs.
Secondly, the API key is hidden from the learners for obvious reasons, as API keys are not for free and what you use in the courses is DeepLearning.AI’s own personal key. Now imagine, if everyone could use that key and the billing it would cost us. Especially when all these courses are for free. So instead, we have provided you with the key in a hidden manner so that people can fully experience the courses, without exploiting the key.
I don’t think apart from the API key, anything else is hidden from the learners. When you open the notebook, click File --> Open..., this will take you to your workspace. In there, you can see and the relevant files being used by the notebook.
As for the code and bugs, could you share the screenshot of what you are referring to ? Please mention the lesson name as well.
Thanks,
Mubsi
1 Like
Thank you. I can comfirm that everything works well in the official jupyter page. Just a suggestion, if people can try in their own environment (such as colab or locally), they can benefit more.
- I know maybe they are just far beyond the scope of this short course
something I think is not that transparent: the hyperparameter, the API (in fact you can run free with lamini example notebook, but not mentioned here) and the wrapper Huggingface Trainer class etc.
1 Like
Hi @SBFALLOUT,
In order to avoid inconveniences in running things in a colab or locally, please make sure you are using exact package versions as they are being used in the platform.
Best,
Mubsi
1 Like