ML and DL specialisation by Andrew Ng in Coursera is very good course for beginners. https://www.coursera.org/learn/python-machine-learning?specialization=data-science-python#syllabus Another ueful course: https://machinelearning-ai.netlify.app/

https://www.kaggle.com/c/digit-recognizer/overview

1. I downloaded the csv files from the website and was trying to access them, then I faced the error :

'function' object is not subscriptable

Then I got to know that my file should be in the same directory as my jupyter file iam working on or i would have to specify the path to the file

Then I faced an error :

(unicode error) 'unicodeescape' codec can't decode bytes in position 2-3: truncated \UXXXXXXXX escape

Then I found solution in this website : https://stackoverflow.com/questions/37400974/unicode-error-unicodeescape-codec-cant-decode-bytes-in-position-2-3-trunca

2. other problems i faced are when creating the csv file to be submitted in kaggle i referred to these websites : https://www.pythontutorial.net/python-basics/python-write-csv-file/ https://www.geeksforgeeks.org/how-to-append-pandas-dataframe-to-existing-csv-file/

1. I split the training data into inputs and labels
2. I used KNN classifier with 3 neighbours
3. I predicted the outputs for the test data using the model
4. appended this data into the csv file
5. submitted it in kaggle
6. I got the accuracy of 0.96803 and leaderboard position of 835