HOW’S MY TRAINING
When you finish the text training, there will be 3 outcome reference on the bottom line to help you asess your training performance.Perplexity
The perplexity measures the amount of “randomness” in our model. If the perplexity is 3 (per word) then that means the model had a 1-in-3 chance of guessing (on average) the next word in the text.
Loss
Loss function is a method that evaluates how well the algorithm learns the data and produces correct outputs.
Score
Overall efficiency and performance of this training, generated by the playform.
Some tips will also be generated to guide users to improve text training. As for the left side, there are always insructions to guide user to start with text training first. If they want to chat with model first, which is ok, they will talk to a plain untrained model.