1 个仓库
Processes for training encoder-decoder architectures to map input sequences to target sequences.
Distinct from Sequence-to-Sequence Models: Distinct from Sequence-to-Sequence Models: focuses on the training process and execution logic rather than the neural network architecture.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Training Workflows. Refine with filters or upvote what's useful.
SimpleTransformers 是一个用于训练和微调 Transformer 模型以执行各种自然语言处理任务的高级框架。它作为一个开发文本分类、命名实体识别和问答模型的工具包,同时也可用作序列到序列工具和文本嵌入生成器。 该库通过提供能够处理和分类结合文本和图像输入数据的多模态模型训练器而脱颖而出。它进一步支持用于对话式 AI 训练、语言模型生成和用于信息检索系统的密集文档检索的专用工作流。 该框架涵盖了广泛的功能,包括具有提前停止功能的训练工作流管理、通过量化进行的模型性能优化,以及创建自定义领域特定分词器。它还通过外部遥测仪表板集成了实验跟踪和预测可视化。
Trains encoder-decoder architectures for text transformation tasks such as language translation and summarization.