Machine Learning

Mean Aveage Precision (mAP) is actually not an easy-understanding metric. Confusing Matrix T\P A B C A 11 2 3 B 3 12 1 C 0 7 9 Normally people draw confusing matrix like this above. Left column is ground Truth (T), first row is the Prediction (P). In this confusing matrix, we have 3 classes, A, B, and C. Each class has 16 samples (sum of each row) which means data is balanced. If the model does a good job, its confusing matrix should have a very clear diagonal. In this example, the model is very confused about class C. Given class C, nearly half of the predictions are B (7/16). ...

While training a model, a few important mathematical tricks have to be employed. Normalization and regularization techiniques belong to them. They are very important to successfully train models, and they are often confusing and sometimes hard to understand the details as well. Additionally, learning rate scheduling is also very important while training models, especially large ones. Normalization vs Regularization To put it simple, normalization is applied on data. Before training a model, data should be normalized. There are a few irresistible benefits to do it. It stablizes the training process, speeds up model convergence significantly, and improve model’s generalization capability. ...