Title and Abstract(0.75/0.75) - Good - add which machine learning model(s) you plan to use for this problem.
Background (1/1) -
Please fix intext citation formatting so they link to appropriate citations in the Footnotes.
Otherwise - Good!
Problem Statement (1/1) -
Good - but add which other algorithms you plan to explore in this section as well. We expect to see a compare/contrast of at least 3 models.
Data (1.25/1.25) -
Clear and detailed - good work!
You can include the data cleaning steps as code cells here. Add markdown cells in between explaining each step.
Proposed Solution (1.25/1.25)-
Clearly state which algorithms you plan to use first - Gradient Boosting, DT, Logreg then mention train/test split, cross-validation techniques, hyperparameters etc.
Otherwise OK.
Metrics (1.25/1.25) -
Fix formatting on the mathematical formulae.
Preliminary Results (1.5/1.5) -
First - conduct some preliminary EDA on your data - perhaps show some scatterplots/bar plots for numeric features, and correlation heatmaps to show the correlation between critical features and the target variable.
Regarding Model Selection - Good! I would add markdown cells between each step to make it clear what you are accomplishing in the notebook - which variables you are dropping/not dropping etc. Analyze the confusion matrices, classification reports, etc. It's hard for me as a grader to follow comments in the code to see what exactly you are doing.
You can also add a comprehensive table comparing model performance for different evaluation metrics (with variables/without variables) at the end. You could also do this per model as a summary.
I see you've implemented Gradient Boosting and SVM - but in your proposed solution you mention DT and Logreg - fix this.
Otherwise - great work!
Ethics and Privacy (0.5/0.5) - Good, keep adding to this section as you continue the analysis of your data.
Team Expectations (0.25/0.25) -
Timeline (0.25/0.25) -
Other Comments -
Good work so far - you guys are on the right track :)
You can reply to this feedback below. Contact me anytime if you want help improving your project or have any questions at all!
Project Checkpoint Grade - 9/9
Title and Abstract(0.75/0.75) - Good - add which machine learning model(s) you plan to use for this problem.
Background (1/1) - Please fix intext citation formatting so they link to appropriate citations in the Footnotes. Otherwise - Good!
Problem Statement (1/1) - Good - but add which other algorithms you plan to explore in this section as well. We expect to see a compare/contrast of at least 3 models.
Data (1.25/1.25) - Clear and detailed - good work! You can include the data cleaning steps as code cells here. Add markdown cells in between explaining each step.
Proposed Solution (1.25/1.25)- Clearly state which algorithms you plan to use first - Gradient Boosting, DT, Logreg then mention train/test split, cross-validation techniques, hyperparameters etc. Otherwise OK.
Metrics (1.25/1.25) - Fix formatting on the mathematical formulae.
Preliminary Results (1.5/1.5) - First - conduct some preliminary EDA on your data - perhaps show some scatterplots/bar plots for numeric features, and correlation heatmaps to show the correlation between critical features and the target variable. Regarding Model Selection - Good! I would add markdown cells between each step to make it clear what you are accomplishing in the notebook - which variables you are dropping/not dropping etc. Analyze the confusion matrices, classification reports, etc. It's hard for me as a grader to follow comments in the code to see what exactly you are doing. You can also add a comprehensive table comparing model performance for different evaluation metrics (with variables/without variables) at the end. You could also do this per model as a summary. I see you've implemented Gradient Boosting and SVM - but in your proposed solution you mention DT and Logreg - fix this. Otherwise - great work!
Ethics and Privacy (0.5/0.5) - Good, keep adding to this section as you continue the analysis of your data.
Team Expectations (0.25/0.25) -
Timeline (0.25/0.25) -
Other Comments - Good work so far - you guys are on the right track :)
You can reply to this feedback below. Contact me anytime if you want help improving your project or have any questions at all!