iarfmoose/t5-base-question-generator

The question generator is based on a pretrained t5-base model . It takes an answer and context as an input, and generates a question as an output . The model performs best with full sentence answers, but can also be used with single word or short phrase answers . It is limited to generating questions in the same style as those found in SQuAD, CoQA, and MSMARCO. The generated questions can potentially be leading or reflect biases that are present in the context. The max sequence length is 512 tokens. The model was trained for 20 epochs over the training set with a learning rate of 1e-3. The full training set was roughly 200,000 examples,