A class that keeps all conversation history.

:class:`transformers.LogitsProcessor` that enforces that when specified sequences appear, stop geration. Args: stop_words_ids (:obj:

Fake file-like stream object that redirects writes to a logger instance.

Multi-head Attention block with relative position embeddings.

Apply a bitmap-style transformation to an image. This transformation replaces all pixel values below a certain threshold with a specified va

Transformer blocks with support of window attention and residual propagation blocks

Apply dilation operation to an image. Dilation is a morphological operation that expands the white regions in a binary image. Args:

Apply erosion operation to an image. Erosion is a morphological operation that shrinks the white regions in a binary image. Args:

Multi-headed attention from 'Attention Is All You Need' paper

Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`OPTDecoderLayer`] Args: config: OPTConfig

This module learns positional embeddings up to a fixed maximum size.

Image to Patch Embedding.

Conversation format dataset stage2 fine-tuning.

Conversation format dataset stage2 fine-tuning.

QWen tokenizer.

Different separator style.