From my personal experience, Auto-encoders are amazing for dense input (images, audio etc), more specifically, when the input feature space is not large. However, in many real-world problems such as recommendation, ranking etc. the feature space is generally very sparse for eg clicks, purchase of items (say 100M items). In such cases, scaling can be challenging with neural models esp Autoencoder.
You do not have to do anything. You can simply stop working on the course at any time you choose to do so.
What happens if I drop the course?
For the prototype course, learners achieving grades of "A," "B," or "C" will receive an electronic Certificate of completion with the learner's name and grade on it. If you receive a grade below a "C" or do not complete the course, you will not receive a Certificate and no grade record attaching your name to your participation in the class will be disclosed outside of MITx. You can also choose to opt for a no record at any time. However, the posts you make while enrolled in the class will remain visible."
Well i think you need to adjust your routine. I think both startup and relationship are equally important and you have to manage accordingly.
Relationship stands on top of understanding and i think you both should understand each other. If your girl understands you and you do the same then i dont think it wont be any problem.
From my personal experience, Auto-encoders are amazing for dense input (images, audio etc), more specifically, when the input feature space is not large. However, in many real-world problems such as recommendation, ranking etc. the feature space is generally very sparse for eg clicks, purchase of items (say 100M items). In such cases, scaling can be challenging with neural models esp Autoencoder.