A class that keeps all conversation history.

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

Collate examples for supervised fine-tuning.

Multi-headed attention. See "Attention Is All You Need" for more details.

Base class for Qwen2VL causal language model (or autoregressive) outputs. Args: loss (`torch.FloatTensor` of shape `(1,)`, *optional

Base class for Qwen2_5_VL causal language model (or autoregressive) outputs. Args: loss (`torch.FloatTensor` of shape `(1,)`, *optio

r""" Sampler that samples indices in a way that groups together features of the dataset of roughly the same length while keeping a bit of rand

Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequ

r""" This is the configuration class to store the configuration of a [`Qwen2VLModel`]. It is used to instantiate a Qwen2-VL model according to

Qwen2VL flash attention module, following Qwen2VL attention module. This module inherits from `Qwen2VLAttention` as the weights of the module

r""" Constructs a Qwen2-VL image processor that dynamically resizes images based on the original images. Args: do_resize (`bool`, *op

r""" Constructs a Qwen2-VL processor which wraps a Qwen2-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2VLProcessor`

Qwen2 attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from `Qwen2Attention` as the weights of t

Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequ

r""" This is the configuration class to store the configuration of a [`Qwen2_5_VLModel`]. It is used to instantiate a Qwen2-VL model according

Qwen2_5_VL flash attention module, following Qwen2_5_VL attention module. This module inherits from `Qwen2_5_VLAttention` as the weights of t

r""" Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLPro

r""" Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLPro

Qwen2 attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from `Qwen2Attention` as the weights of t