mindformers.models

Base Classes

`mindformers.models.PreTrainedModel`	Base class for all models.
`mindformers.models.PretrainedConfig`	Base class for all configuration classes.
`mindformers.models.PreTrainedTokenizer`	Base class for all slow tokenizers.
`mindformers.models.PreTrainedTokenizerFast`	Base class for all fast tokenizers (wrapping HuggingFace tokenizers library).
`mindformers.models.multi_modal.ModalContentTransformTemplate`	Base class of modal content transform template.

`mindformers.models.LlamaForCausalLM`	Provide llama training loss or logits through network.
`mindformers.models.LlamaConfig`	Llama config class which defines the model size.
`mindformers.models.LlamaTokenizer`	Construct a Llama tokenizer based on byte-level Byte-Pair-Encoding.
`mindformers.models.LlamaTokenizerFast`	Construct a Llama tokenizer.

`mindformers.models.glm2.ChatGLM2ForConditionalGeneration`	Provide ChatGLM2 training loss or logits through network.
`mindformers.models.glm2.ChatGLM2Config`	ChatGLM2 model config class which defines the model size.
`mindformers.models.glm2.ChatGLM3Tokenizer`	Construct a ChatGLM3 tokenizer.
`mindformers.models.glm2.ChatGLM4Tokenizer`	Construct a ChatGLM4 tokenizer.