Day 6 of #31DaysofML
πΉ Training & Loss πΉ
In supervised learning training a model means learning good values for weights & bias from labeled examples. In doing so it attempts to find a model that minimizes loss. Process is called empirical risk minimization
A π§΅
πΉ Training & Loss πΉ
In supervised learning training a model means learning good values for weights & bias from labeled examples. In doing so it attempts to find a model that minimizes loss. Process is called empirical risk minimization
A π§΅
2/4
π Loss indicates how bad model's prediction was on an example
π Loss = 0 if the model's prediction is perfect otherwise it's greater.
π Goal of training is to find weights & biases that have low loss, on average, across dataset
#31DaysofML
π Loss indicates how bad model's prediction was on an example
π Loss = 0 if the model's prediction is perfect otherwise it's greater.
π Goal of training is to find weights & biases that have low loss, on average, across dataset
#31DaysofML
3/4
Squared loss or L2 loss is a popular Loss function that aggregates the individual losses
L2 loss for a single example = (observation - prediction)^2
#31DaysofML
Squared loss or L2 loss is a popular Loss function that aggregates the individual losses
L2 loss for a single example = (observation - prediction)^2
#31DaysofML
4/4
Mean square error (MSE) is the average squared loss per example over the whole dataset. To calculate MSE, sum up all the squared losses for individual examples and then divide by the number of examples. It looks like this π
#31DaysofML
Mean square error (MSE) is the average squared loss per example over the whole dataset. To calculate MSE, sum up all the squared losses for individual examples and then divide by the number of examples. It looks like this π
#31DaysofML
β’ β’ β’
Missing some Tweet in this thread? You can try to
force a refresh
γ
