Depthwise卷积与Pointwise卷积残差网络CNN的平移不变性机器学习相关交叉熵/相对熵A tutorial of transformersFAQbatch normalization标准化、正则化、归一化one-hot encoding和label encoder编码对深度学习的启发性理解Vision Transformer预训练模型PTMsword2vec和word embeddingRepresentation Learning