`bridge.recipes.qwen_vl.qwen25_vl`#

Module Contents#

Classes#

Qwen25VLCommonKwargs

Typed options accepted by Qwen2.5-VL recipe helper functions.

Functions#

`qwen25_vl_3b_finetune_config`	Return a fine-tuning config for Qwen2.5-VL 3B Instruct.
`qwen25_vl_7b_finetune_config`	Return a fine-tuning config for Qwen2.5-VL 7B Instruct.
`qwen25_vl_32b_finetune_config`	Return a fine-tuning config for Qwen2.5-VL 32B Instruct.
`qwen25_vl_72b_finetune_config`	Return a fine-tuning config for Qwen2.5-VL 72B Instruct.
`_qwen25_vl_common`	Create a fine-tuning configuration for Qwen2.5-VL models using a given HuggingFace path.

API#

class bridge.recipes.qwen_vl.qwen25_vl.Qwen25VLCommonKwargs#

Bases: typing_extensions.TypedDict

Typed options accepted by Qwen2.5-VL recipe helper functions.

Initialization

Initialize self. See help(type(self)) for accurate signature.

hf_path: str#: None

dir: Optional[str]#: None

name: str#: None

train_data_path: Optional[List[str]]#: None

valid_data_path: Optional[List[str]]#: None

test_data_path: Optional[List[str]]#: None

dataset_type: Optional[str]#: None

image_folder: Optional[str]#: None

tokenizer_model: Optional[str]#: None

tensor_parallelism: int#: None

pipeline_parallelism: int#: None

pipeline_parallelism_dtype: Optional[torch.dtype]#: None

virtual_pipeline_parallelism: Optional[int]#: None

context_parallelism: int#: None

sequence_parallelism: bool#: None

use_megatron_fsdp: bool#: None

train_iters: int#: None

global_batch_size: int#: None

micro_batch_size: int#: None

seq_length: int#: None

lr: float#: None

min_lr: float#: None

lr_warmup_iters: int#: None

lr_decay_iters: Optional[int]#: None

eval_interval: int#: None

save_interval: int#: None

precision_config: Optional[Union[megatron.bridge.training.mixed_precision.MixedPrecisionConfig, str]]#: None

comm_overlap_config: Optional[megatron.bridge.training.comm_overlap.CommOverlapConfig]#: None

freeze_language_model: bool#: None

freeze_vision_model: bool#: None

freeze_vision_projection: bool#: None

pretrained_checkpoint: Optional[str]#: None

bridge.recipes.qwen_vl.qwen25_vl.qwen25_vl_3b_finetune_config(

**user_kwargs: typing_extensions.Unpack[bridge.recipes.qwen_vl.qwen25_vl.Qwen25VLCommonKwargs],

) → megatron.bridge.training.config.ConfigContainer#

Return a fine-tuning config for Qwen2.5-VL 3B Instruct.

See _qwen25_vl_common for the full list of parameters.

bridge.recipes.qwen_vl.qwen25_vl.qwen25_vl_7b_finetune_config(

**user_kwargs: typing_extensions.Unpack[bridge.recipes.qwen_vl.qwen25_vl.Qwen25VLCommonKwargs],

) → megatron.bridge.training.config.ConfigContainer#

Return a fine-tuning config for Qwen2.5-VL 7B Instruct.

See _qwen25_vl_common for the full list of parameters.

bridge.recipes.qwen_vl.qwen25_vl.qwen25_vl_32b_finetune_config(

**user_kwargs: typing_extensions.Unpack[bridge.recipes.qwen_vl.qwen25_vl.Qwen25VLCommonKwargs],

) → megatron.bridge.training.config.ConfigContainer#

Return a fine-tuning config for Qwen2.5-VL 32B Instruct.

See _qwen25_vl_common for the full list of parameters.

bridge.recipes.qwen_vl.qwen25_vl.qwen25_vl_72b_finetune_config(

**user_kwargs: typing_extensions.Unpack[bridge.recipes.qwen_vl.qwen25_vl.Qwen25VLCommonKwargs],

) → megatron.bridge.training.config.ConfigContainer#

Return a fine-tuning config for Qwen2.5-VL 72B Instruct.

See _qwen25_vl_common for the full list of parameters.

bridge.recipes.qwen_vl.qwen25_vl._qwen25_vl_common( hf_path: str, dir: Optional[str] = None, name: str = 'qwen25_vl_finetune', pretrained_checkpoint: Optional[str] = None, train_data_path: Optional[List[str]] = None, valid_data_path: Optional[List[str]] = None, test_data_path: Optional[List[str]] = None, dataset_type: Optional[str] = None, image_folder: Optional[str] = None, tokenizer_model: Optional[str] = None, tensor_parallelism: int = 2, pipeline_parallelism: int = 1, pipeline_parallelism_dtype: Optional[torch.dtype] = None, virtual_pipeline_parallelism: Optional[int] = None, context_parallelism: int = 1, sequence_parallelism: bool = False, use_megatron_fsdp: bool = False, train_iters: int = 300000, global_batch_size: int = 32, micro_batch_size: int = 2, seq_length: int = 4096, lr: float = 0.0003, min_lr: float = 3e-05, lr_warmup_iters: int = 500, lr_decay_iters: Optional[int] = None, eval_interval: int = 500, save_interval: int = 500, precision_config: Optional[Union[megatron.bridge.training.mixed_precision.MixedPrecisionConfig, str]] = 'bf16_mixed', comm_overlap_config: Optional[megatron.bridge.training.comm_overlap.CommOverlapConfig] = None, freeze_language_model: bool = False, freeze_vision_model: bool = False, freeze_vision_projection: bool = False, ) → megatron.bridge.training.config.ConfigContainer#

Create a fine-tuning configuration for Qwen2.5-VL models using a given HuggingFace path.

The dataset pipeline is conversation-based. To train multimodal tokens, ensure your preprocessed data includes placeholders (e.g., ) as needed.

bridge.recipes.qwen_vl.qwen25_vl#

Module Contents#

Classes#

Functions#

API#

`bridge.recipes.qwen_vl.qwen25_vl`#