pre-trained

  Segmentation models with pretrained backbones. Keras and TensorFlow Keras.

翻译 - 具有预训练骨干的细分模型。 Keras和TensorFlow Keras。

  A LITE BERT FOR SELF-SUPERVISED LEARNING OF LANGUAGE REPRESENTATIONS, 海量中文预训练ALBERT模型

  RoBERTa中文预训练模型: RoBERTa for Chinese

  Pre-trained Neural Network models in Axon (+ 🤗 Models integration)

  MEAL V2: Boosting Vanilla ResNet-50 to 80%+ Top-1 Accuracy on ImageNet without Tricks. In NeurIPS 2020 workshop.

翻译 - 餐点V2：在没有技巧的情况下将ImageNet上的Vanilla ResNet-50提高到80％+ Top-1准确性