← Back to Teacher Trial

PHL Classroom Engine Science Educator Evaluation Programme

Give your learners time to investigate how a training dataset shapes a model’s predictions. This PHL-led evaluation suits educators planning a sequence of practical science or data lessons in the Virtual Lab.

Can a model distinguish virtual fungi?

  1. Collect repeated examples of each labelled virtual fungus. Reserve some examples for testing.
  2. Train a classification model. In its confusion matrix, compare the expected and predicted labels. Which fungi does the model most often confuse?
  3. Predict the labels of unknown specimens. Discuss whether collecting more examples of a particular class could improve the result.

Virtual specimens generate synthetic educational sensor responses. Measuring the chemistry of real specimens requires physical equipment.

We ask educators to try their proposed investigation and share feedback on what worked in class and where learners needed help. If you want to repeat the work with measured data, tell us about your interest in a sensor demonstration, pilot or purchase.

Access for your classroom

The standard evaluation runs for 30 days from activation. Your classroom shares 120 saved captures and 30 flushes across all participating users and devices.

Student participation is off by default and requires explicit inclusion. Programme approval covers the Virtual Lab. Physical sensors require the appropriate separate licence. Your existing AI, classroom, task, calendar and physical-device allowances continue to apply.

PHL reviews each application individually and contacts you by email. If approved, you can onboard through Classroom Engine Settings, including after an ordinary Teacher Trial has expired.

Apply for an evaluation

Tell us about your teaching plans. PHL uses these details to assess your application and contact you about it. Submitting an application does not activate access.