speedy-utils 1.1.15__py3-none-any.whl → 1.1.17__py3-none-any.whl

llm_utils/lm/async_lm/async_llm_task.py

@@ -11,8 +11,6 @@ from venv import logger
 
 from openai.types.chat import ChatCompletionMessageParam
 from pydantic import BaseModel
-from pytest import Cache
-from speedy_utils import jdumps
 from speedy_utils.all import dump_json_or_pickle, identify
 
 from llm_utils.chat_format.display import get_conversation_one_turn

llm_utils/vector_cache/core.py

@@ -167,12 +167,11 @@ class VectorCache:
         return "vllm"
     def _try_infer_model_name(self, model_name: Optional[str]) -> Optional[str]:
         """Infer model name for OpenAI backend if not explicitly provided."""
-        # if self.backend != "openai":
-            # return model_name
         if model_name:
             return model_name
         if 'https://' in self.url_or_model:
             model_name =  "text-embedding-3-small"
+        
         if 'http://localhost' in self.url_or_model:
             from openai import OpenAI
             client = OpenAI(base_url=self.url_or_model, api_key='abc')
@@ -277,6 +276,8 @@ class VectorCache:
 
     def _get_embeddings(self, texts: list[str]) -> list[list[float]]:
         """Get embeddings using the configured backend."""
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         if self.backend == "openai":
             return self._get_openai_embeddings(texts)
         elif self.backend == "vllm":
@@ -288,6 +289,8 @@ class VectorCache:
 
     def _get_openai_embeddings(self, texts: list[str]) -> list[list[float]]:
         """Get embeddings using OpenAI API."""
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         # Assert valid model_name for OpenAI backend
         model_name = self.config["model_name"]
         assert model_name is not None and model_name.strip(), f"Invalid model_name for OpenAI backend: {model_name}. Model name must be provided and non-empty."
@@ -304,6 +307,8 @@ class VectorCache:
 
     def _get_vllm_embeddings(self, texts: list[str]) -> list[list[float]]:
         """Get embeddings using vLLM."""
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         if self._model is None:
             self._load_model()
         
@@ -313,6 +318,8 @@ class VectorCache:
 
     def _get_transformers_embeddings(self, texts: list[str]) -> list[list[float]]:
         """Get embeddings using transformers directly."""
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         if self._model is None:
             self._load_model()
         
@@ -380,6 +387,8 @@ class VectorCache:
         handle very large input lists. A tqdm progress bar is shown while
         computing missing embeddings.
         """
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         if not texts:
             return np.empty((0, 0), dtype=np.float32)
         t = time()
@@ -436,6 +445,8 @@ class VectorCache:
         return np.vstack([hit_map[h] for h in hashes])
 
     def __call__(self, texts: list[str], cache: bool = True) -> np.ndarray:
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         return self.embeds(texts, cache)
 
     def _bulk_insert(self, data: list[tuple[str, str, bytes]]) -> None:
@@ -454,6 +465,8 @@ class VectorCache:
         Precompute embeddings for a large list of texts efficiently.
         This is optimized for bulk operations when you know all texts upfront.
         """
+        assert isinstance(texts, list), "texts must be a list"
+        assert all(isinstance(t, str) for t in texts), "all elements in texts must be strings"
         if not texts:
             return
 

{speedy_utils-1.1.15.dist-info → speedy_utils-1.1.17.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: speedy-utils
-Version: 1.1.15
+Version: 1.1.17
 Summary: Fast and easy-to-use package for data science
 Author: AnhVTH
 Author-email: anhvth.226@gmail.com

{speedy_utils-1.1.15.dist-info → speedy_utils-1.1.17.dist-info}/RECORD

@@ -7,7 +7,7 @@ llm_utils/group_messages.py,sha256=Oe2tlhg-zRodG1-hodYebddrR77j9UdE05LzJw0EvYI,3
 llm_utils/lm/__init__.py,sha256=rX36_MsnekM5GHwWS56XELbm4W5x2TDwnPERDTfo0eU,194
 llm_utils/lm/async_lm/__init__.py,sha256=PUBbCuf5u6-0GBUu-2PI6YAguzsyXj-LPkU6vccqT6E,121
 llm_utils/lm/async_lm/_utils.py,sha256=P1-pUDf_0pDmo8WTIi43t5ARlyGA1RIJfpAhz-gfA5g,6105
-llm_utils/lm/async_lm/async_llm_task.py,sha256=iXSTbf-KekXncVVnic-v4dTq5HBDjbyLwhgo0Y-wp7Q,19167
+llm_utils/lm/async_lm/async_llm_task.py,sha256=IT63_qtKWs1Svk6tic12ALxHA9du2WnAjX5xXuYH9jA,19110
 llm_utils/lm/async_lm/async_lm.py,sha256=slGOUXFeWAy3Ak7Xj7Z4JNqCgAUBo21Hjg3RD75Ul2Q,13396
 llm_utils/lm/async_lm/async_lm_base.py,sha256=Qh9Bx_S-FboO7huUi6TqP3KiTVHDH0C-Tfbd_UJC7Cc,8122
 llm_utils/lm/async_lm/lm_specific.py,sha256=KmqdCm3SJ5MqN-dRJd6S5tq5-ve1X2eNWf2CMFtc_3s,3926
@@ -18,7 +18,7 @@ llm_utils/scripts/vllm_load_balancer.py,sha256=TT5Ypq7gUcl52gRFp--ORFFjzhfGlcaX2
 llm_utils/scripts/vllm_serve.py,sha256=gJ0-y4kybMfSt8qzye1pJqGMY3x9JLRi6Tu7RjJMnss,14771
 llm_utils/vector_cache/__init__.py,sha256=i1KQuC4OhPewYpFl9X6HlWFBuASCTx2qgGizhpZhmn0,862
 llm_utils/vector_cache/cli.py,sha256=DMXTj8nZ2_LRjprbYPb4uzq04qZtOfBbmblmaqDcCuM,6251
-llm_utils/vector_cache/core.py,sha256=-g_y3U4hdJltHMVi8oYJTsW8p7lR973LlDqcLZ9kECk,22725
+llm_utils/vector_cache/core.py,sha256=O1C3azjPT7wyp8NOGPSSd3BPacM0fo-oSc3BdaISr6I,23769
 llm_utils/vector_cache/types.py,sha256=ru8qmUZ8_lNd3_oYpjCMtpXTsqmwsSBe56Z4hTWm3xI,435
 llm_utils/vector_cache/utils.py,sha256=dwbbXlRrARrpmS4YqSlYQqrTURg0UWe8XvaAWcX05MM,1458
 speedy_utils/__init__.py,sha256=nJpUb5Oa3STDbqPSiWXoI-IvKntyRYzYxkYW4GM2i_Q,5740
@@ -39,7 +39,7 @@ speedy_utils/multi_worker/thread.py,sha256=f02VjJV8nudg0eA_AcfPEX7tHY4-czesuzthK
 speedy_utils/scripts/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 speedy_utils/scripts/mpython.py,sha256=IvywP7Y0_V6tWfMP-4MjPvN5_KfxWF21xaLJsCIayCk,3821
 speedy_utils/scripts/openapi_client_codegen.py,sha256=f2125S_q0PILgH5dyzoKRz7pIvNEjCkzpi4Q4pPFRZE,9683
-speedy_utils-1.1.15.dist-info/METADATA,sha256=GOftJglyMxP320bZFW8XX-UurB8vr_01jDNGV7ecKh8,7483
-speedy_utils-1.1.15.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-speedy_utils-1.1.15.dist-info/entry_points.txt,sha256=T1t85jwx8fK6m5msdkBGIXH5R5Kd0zSL0S6erXERPzg,237
-speedy_utils-1.1.15.dist-info/RECORD,,
+speedy_utils-1.1.17.dist-info/METADATA,sha256=3o7AJX8v5MS-d9j-FgQWZHKikyVPmlzkm4p1-gj2ISM,7483
+speedy_utils-1.1.17.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+speedy_utils-1.1.17.dist-info/entry_points.txt,sha256=T1t85jwx8fK6m5msdkBGIXH5R5Kd0zSL0S6erXERPzg,237
+speedy_utils-1.1.17.dist-info/RECORD,,