wptoux/albert-chinese-large-qa

We combined the two datasets together and created a new dataset in squad format . The new dataset includes 705139 samples for training and 69638 samples for validation . We finetune the model based on the albert chinese large model . The model was pretrained from the baidu webqa and baidU dureader datasets . Please visit https://://://github.com/wptoux/albert-chinese-large-qa for details of the training and validation of the new dataset . The final model is based on a model that was trained on the model of a large QA model . It is now being used to test the accuracy of the data .,