Hello everyone. I am looking for some guidance. Help!
I completed the first course of the specialization and I’m not sure how to proceed. I know the obvious thing to do would be to move on to the next course, but i would like to practice some more what I learned on the first course.
Are there some projects available on the internet, or are we just supposed to find a dataset from somewhere and do whatever we want with it ?
I tried looking at the projects tab here as well as on Kaggle (supposedly one of the best platforms to learn and practice) but besides the courses I can’t find much. Some of the projects i could find are too advanced for what i currently know.
Maybe i don’t know where to look, or i don’t know what to look for. Please help me in clarifying the situation. Your input is much appreciated.
Tl;dr : I want to practice what i learned, but I don’t know how. Help!
Yes, just completing the first course in the MLS specialization isn’t really enough to get you very far into more complicated data sets.
I recommend you continue in the course, and also look on some of the internet dataset repositories. You can get lots of data for your own experiments, for free.
Try the “UCI Machine Learning Respository”, for example.
Obviously completing the first course alone is not going to get me very far, but i still wanted to have more practice on the concepts learned aka regression and classification. What i’m looking for here is to increase my confidence in implementing those models.
I obviously plan on completing the rest of the courses. However, they tackle different concepts and are not there to help with regression and classification proficiency.
Thank you for the repository, i will check it out and experiment with it.
Hey Coricara - I am in the same boat as you are. I have been trying a few things to just get my head around some of the computation and the process follwed in the 1st course. This is what I have been doing
Install PyCharm and run the course labs on my own. In some cases copied the code, compiled and played around with it.
To understand the process/steps follwed in the course . I tried the following approach. As an example - error calculations, loss function etc, I have been using some mock data and working them out in excel etc. I do this just to get a feel for data and the numbers that I will be dealing with.
Just like you I looked at Kaggle, it has some very good examples and data sets but I could not get much far. beyond admiring what people had done.
Please be very careful running the labs in your own environment.
It’s fine if it’s just for your own experiments.
But if you work on a notebook file locally and then upload it back to Coursera for grading, it’s very likely that the notebook metadata will be mangled in ways that make the grader very unhappy.