[ad_1]

In the video lectures, Professor Kim showed you how to use TensorFlow to train neural networks to predict outcomes. Now it is your turn to do it.

  1. Go to: http://playground.tensorflow.org/
  2. Start with the data set shown in the attachment.
  3. Play with the various options you have available on the user interface to train the model.
  4. Your goal is to predict this outcome with a test loss of less than 0.03 in as few epochs as possible.
    Note: There is a trade-off here. The more hidden layers you add, the more features you add. With more features, it might take longer to train a model. On the other hand, without enough features (hidden layers and neurons) you may never reach the complexity needed to get a test loss below 0.03

Sample Solution

The post Technology;Tensor Flow appeared first on acestar tutors.

[ad_2]

Source link