From the course: Full-Stack Deep Learning with Python

Unlock this course with a free trial

Join today to access over 24,400 courses taught by industry experts.

Visualizing charts, metrics, and parameters on MLflow

Visualizing charts, metrics, and parameters on MLflow - Python Tutorial

From the course: Full-Stack Deep Learning with Python

Start my 1-month free trial Buy for my team

Visualizing charts, metrics, and parameters on MLflow

“

- [Instructor] Now that we've trained our convolutional neural network and tracked its metrics and parameters using MLflow runs, let's head over to the MLflow UI and here you can see the experiment that we've just created and used the emnist_letters_prediction_using_cnn. You can see this exactly one run here within this experiment, mysterious-fly-874. That is our current training run. Let's take a look at the chart first so that we can see how training accuracy and validation accuracy changed over time. The very first chart that we get, let me expand the view first is of the test accuracy. This is a single value and you can see that it's rather high, 0.91. If you scroll down, you'll see how the training loss changed over time. You can see the loss was high and it gradually fell over the 10 epochs of training. Maybe we could have trained further to get a better model. You can see the validation accuracy here. It arose up…

Contents