Reduce Console Logging (#45)

- feat: reduced default noise of the model (cf456d3eb4b965685af2a435dc88510ed005d762)
- feat: default verbosity 1 (191d83caa0a8f06c2c1c8ca849df21bdf8857c4c)

configuration_jina_embeddings_v4.py

@@ -2,6 +2,7 @@ from transformers.models.qwen2_5_vl import Qwen2_5_VLConfig
 
 from typing import Optional
 
+
 class JinaEmbeddingsV4Config(Qwen2_5_VLConfig):
     """
     Configuration for the JinaEmbeddingsV4 model.
@@ -12,10 +13,11 @@ class JinaEmbeddingsV4Config(Qwen2_5_VLConfig):
         single_vector_pool_strategy: str = "mean",
         multi_vector_projector_dim: int = 128,
         pretrained_peft_model_name_or_path: Optional[str] = None,
+        verbosity: int = 1,
         **kwargs,
     ):
         super().__init__(**kwargs)
         self.single_vector_pool_strategy = single_vector_pool_strategy
         self.multi_vector_projector_dim = multi_vector_projector_dim
         self.pretrained_peft_model_name_or_path = pretrained_peft_model_name_or_path
-
+        self.verbosity = verbosity

modeling_jina_embeddings_v4.py

@@ -146,6 +146,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             self.name_or_path, trust_remote_code=True, use_fast=True
         )
         self.multi_vector_projector_dim = config.multi_vector_projector_dim
+        self.verbosity = config.verbosity
         self._task = None
 
     @property
@@ -336,7 +337,7 @@ class JinaEmbeddingsV4Model(Qwen2_5_VLForConditionalGeneration):
             ), "`return_numpy` is not supported when `return_multivector=True` and more than one data is encoded"
         results = []
         self.eval()
-        for batch in tqdm(dataloader, desc=desc):
+        for batch in tqdm(dataloader, desc=desc, disable=self.verbosity == 0):
             with torch.no_grad():
                 batch = {k: v.to(self.device) for k, v in batch.items()}
                 with torch.autocast(