trustgraph-embeddings-hf 0.22.5__tar.gz → 0.22.7__tar.gz

{trustgraph-embeddings-hf-0.22.5 → trustgraph-embeddings-hf-0.22.7}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.1
 Name: trustgraph-embeddings-hf
-Version: 0.22.5
+Version: 0.22.7
 Summary: HuggingFace embeddings support for TrustGraph.
 Home-page: https://github.com/trustgraph-ai/trustgraph
-Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.22.5.tar.gz
+Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.22.7.tar.gz
 Author: trustgraph.ai
 Author-email: security@trustgraph.ai
 Classifier: Programming Language :: Python :: 3

trustgraph-embeddings-hf-0.22.7/trustgraph/embeddings/hf/hf.py

@@ -0,0 +1,48 @@
+
+"""
+Embeddings service, applies an embeddings model selected from HuggingFace.
+Input is text, output is embeddings vector.
+"""
+
+from ... base import EmbeddingsService
+
+from langchain_huggingface import HuggingFaceEmbeddings
+
+default_ident = "embeddings"
+
+default_model="all-MiniLM-L6-v2"
+
+class Processor(EmbeddingsService):
+
+    def __init__(self, **params):
+
+        model = params.get("model", default_model)
+
+        super(Processor, self).__init__(
+            **params | { "model": model }
+        )
+
+        print("Get model...", flush=True)
+        self.embeddings = HuggingFaceEmbeddings(model_name=model)
+
+    async def on_embeddings(self, text):
+
+        embeds = self.embeddings.embed_documents([text])
+        print("Done.", flush=True)
+        return embeds
+
+    @staticmethod
+    def add_args(parser):
+
+        EmbeddingsService.add_args(parser)
+
+        parser.add_argument(
+            '-m', '--model',
+            default="all-MiniLM-L6-v2",
+            help=f'LLM model (default: all-MiniLM-L6-v2)'
+        )
+
+def run():
+
+    Processor.launch(default_ident, __doc__)
+

trustgraph-embeddings-hf-0.22.7/trustgraph/embeddings_hf_version.py

@@ -0,0 +1 @@
+__version__ = "0.22.7"

{trustgraph-embeddings-hf-0.22.5 → trustgraph-embeddings-hf-0.22.7}/trustgraph_embeddings_hf.egg-info/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.1
 Name: trustgraph-embeddings-hf
-Version: 0.22.5
+Version: 0.22.7
 Summary: HuggingFace embeddings support for TrustGraph.
 Home-page: https://github.com/trustgraph-ai/trustgraph
-Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.22.5.tar.gz
+Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.22.7.tar.gz
 Author: trustgraph.ai
 Author-email: security@trustgraph.ai
 Classifier: Programming Language :: Python :: 3

trustgraph-embeddings-hf-0.22.5/trustgraph/embeddings/hf/hf.py

@@ -1,100 +0,0 @@
-
-"""
-Embeddings service, applies an embeddings model selected from HuggingFace.
-Input is text, output is embeddings vector.
-"""
-
-from langchain_huggingface import HuggingFaceEmbeddings
-
-from trustgraph.schema import EmbeddingsRequest, EmbeddingsResponse, Error
-from trustgraph.schema import embeddings_request_queue
-from trustgraph.schema import embeddings_response_queue
-from trustgraph.log_level import LogLevel
-from trustgraph.base import ConsumerProducer
-
-module = ".".join(__name__.split(".")[1:-1])
-
-default_input_queue = embeddings_request_queue
-default_output_queue = embeddings_response_queue
-default_subscriber = module
-default_model="all-MiniLM-L6-v2"
-
-class Processor(ConsumerProducer):
-
-    def __init__(self, **params):
-
-        input_queue = params.get("input_queue", default_input_queue)
-        output_queue = params.get("output_queue", default_output_queue)
-        subscriber = params.get("subscriber", default_subscriber)
-        model = params.get("model", default_model)
-
-        super(Processor, self).__init__(
-            **params | {
-                "input_queue": input_queue,
-                "output_queue": output_queue,
-                "subscriber": subscriber,
-                "input_schema": EmbeddingsRequest,
-                "output_schema": EmbeddingsResponse,
-            }
-        )
-
-        self.embeddings = HuggingFaceEmbeddings(model_name=model)
-
-    async def handle(self, msg):
-
-        v = msg.value()
-
-        # Sender-produced ID
-        id = msg.properties()["id"]
-
-        print(f"Handling input {id}...", flush=True)
-
-        try:
-
-            text = v.text
-            embeds = self.embeddings.embed_documents([text])
-
-            print("Send response...", flush=True)
-            r = EmbeddingsResponse(vectors=embeds, error=None)
-            await self.send(r, properties={"id": id})
-
-            print("Done.", flush=True)
-
-
-        except Exception as e:
-
-            print(f"Exception: {e}")
-
-            print("Send error response...", flush=True)
-
-            r = EmbeddingsResponse(
-                error=Error(
-                    type = "llm-error",
-                    message = str(e),
-                ),
-                response=None,
-            )
-
-            await self.send(r, properties={"id": id})
-
-            self.consumer.acknowledge(msg)
-            
-
-    @staticmethod
-    def add_args(parser):
-
-        ConsumerProducer.add_args(
-            parser, default_input_queue, default_subscriber,
-            default_output_queue,
-        )
-
-        parser.add_argument(
-            '-m', '--model',
-            default="all-MiniLM-L6-v2",
-            help=f'LLM model (default: all-MiniLM-L6-v2)'
-        )
-
-def run():
-
-    Processor.launch(module, __doc__)
-

trustgraph-embeddings-hf-0.22.5/trustgraph/embeddings_hf_version.py

@@ -1 +0,0 @@
-__version__ = "0.22.5"