A$.$ Model Evaluation:

$-$ Accuracy

$-$ Precision

$-$ Recall

$-$ F$1$ Score

$-$ ROC$-$AUC Score

$-$ Use cross$-$validation to ensure the reliability of your results.

B$.$ Comparison and Analysis:

$-$ Compare the models based on the performance metrics.

$-$ Discuss the strengths and weaknesses of each model in the context of digit

recognition.

$-$ Provide insights into the challenges of using machine learning for digit recognition, if any.

There are $7291$ training observations and $2007$ test observations, distributed as follows:

$0123456789$ Total

Train $119410057316586525566646455426447291$

Test $3592641981662001601701471661772007$

or as proportions:

$0123456789$

Train $0\mathrm{.}160\mathrm{.}140\mathrm{.}10\mathrm{.}090\mathrm{.}090\mathrm{.}080\mathrm{.}090\mathrm{.}090\mathrm{.}070\mathrm{.}09$

Test $0\mathrm{.}180\mathrm{.}130\mathrm{.}10\mathrm{.}080\mathrm{.}100\mathrm{.}080\mathrm{.}080\mathrm{.}070\mathrm{.}080\mathrm{.}09$

Using KNN $5-$fold cross validation, ANN, OR SVM

Create Models showing before and after cross validation and compare training and test data set.