fix: Update device deserialization for components that use local models (#7686)

* fix: Update device deserializtion for SentenceTransformersTextEmbedder

* Add unit test

* Fix unit test

* Make same change to doc embedder

* Add release notes

* Add same change to Diversity Ranker and Named Entity Extractor

* Add unit test

* Add the same for whisper local

* Update release notes

haystack/components/audio/whisper_local.py

@@ -90,9 +90,9 @@ class LocalWhisperTranscriber:
         :returns:
             The deserialized component.
         """
-        serialized_device = data["init_parameters"]["device"]
-        data["init_parameters"]["device"] = ComponentDevice.from_dict(serialized_device)
-
+        init_params = data["init_parameters"]
+        if init_params["device"] is not None:
+            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
         return default_from_dict(cls, data)
 
     @component.output_types(documents=List[Document])

haystack/components/embedders/sentence_transformers_document_embedder.py

@@ -125,9 +125,9 @@ class SentenceTransformersDocumentEmbedder:
         :returns:
             Deserialized component.
         """
-        serialized_device = data["init_parameters"]["device"]
-        data["init_parameters"]["device"] = ComponentDevice.from_dict(serialized_device)
-
+        init_params = data["init_parameters"]
+        if init_params["device"] is not None:
+            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
         deserialize_secrets_inplace(data["init_parameters"], keys=["token"])
         return default_from_dict(cls, data)
 

haystack/components/embedders/sentence_transformers_text_embedder.py

@@ -115,9 +115,9 @@ class SentenceTransformersTextEmbedder:
         :returns:
             Deserialized component.
         """
-        serialized_device = data["init_parameters"]["device"]
-        data["init_parameters"]["device"] = ComponentDevice.from_dict(serialized_device)
-
+        init_params = data["init_parameters"]
+        if init_params["device"] is not None:
+            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
         deserialize_secrets_inplace(data["init_parameters"], keys=["token"])
         return default_from_dict(cls, data)
 

haystack/components/extractors/named_entity_extractor.py

@@ -215,7 +215,8 @@ class NamedEntityExtractor:
         """
         try:
             init_params = data["init_parameters"]
-            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
+            if init_params["device"] is not None:
+                init_params["device"] = ComponentDevice.from_dict(init_params["device"])
             return default_from_dict(cls, data)
         except Exception as e:
             raise DeserializationError(f"Couldn't deserialize {cls.__name__} instance") from e

haystack/components/rankers/sentence_transformers_diversity.py

@@ -141,9 +141,9 @@ class SentenceTransformersDiversityRanker:
         :returns:
             The deserialized component.
         """
-        serialized_device = data["init_parameters"]["device"]
-        data["init_parameters"]["device"] = ComponentDevice.from_dict(serialized_device)
-
+        init_params = data["init_parameters"]
+        if init_params["device"] is not None:
+            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
         deserialize_secrets_inplace(data["init_parameters"], keys=["token"])
         return default_from_dict(cls, data)
 

releasenotes/notes/fix-device-deserialization-st-embedder-c4efad96dd3869d5.yaml

@@ -0,0 +1,5 @@
+---
+fixes:
+  - |
+    Updates the from_dict method of SentenceTransformersTextEmbedder, SentenceTransformersDocumentEmbedder, NamedEntityExtractor, SentenceTransformersDiversityRanker and LocalWhisperTranscriber to allow None as a valid value for device when deserializing from a YAML file.
+    This allows a deserialized pipeline to auto-determine what device to use using the ComponentDevice.resolve_device logic.

test/components/audio/test_whisper_local.py

@@ -72,6 +72,17 @@ class TestLocalWhisperTranscriber:
         assert transcriber.whisper_params == {}
         assert transcriber._model is None
 
+    def test_from_dict_none_device(self):
+        data = {
+            "type": "haystack.components.audio.whisper_local.LocalWhisperTranscriber",
+            "init_parameters": {"model": "tiny", "device": None, "whisper_params": {}},
+        }
+        transcriber = LocalWhisperTranscriber.from_dict(data)
+        assert transcriber.model == "tiny"
+        assert transcriber.device == ComponentDevice.resolve_device(None)
+        assert transcriber.whisper_params == {}
+        assert transcriber._model is None
+
     def test_warmup(self):
         with patch("haystack.components.audio.whisper_local.whisper") as mocked_whisper:
             transcriber = LocalWhisperTranscriber(model="large-v2", device=ComponentDevice.from_str("cpu"))

test/components/embedders/test_sentence_transformers_document_embedder.py

@@ -137,6 +137,38 @@ class TestSentenceTransformersDocumentEmbedder:
         assert component.trust_remote_code
         assert component.meta_fields_to_embed == ["meta_field"]
 
+    def test_from_dict_none_device(self):
+        init_parameters = {
+            "model": "model",
+            "device": None,
+            "token": {"env_vars": ["ENV_VAR"], "strict": False, "type": "env_var"},
+            "prefix": "prefix",
+            "suffix": "suffix",
+            "batch_size": 64,
+            "progress_bar": False,
+            "normalize_embeddings": True,
+            "embedding_separator": " - ",
+            "meta_fields_to_embed": ["meta_field"],
+            "trust_remote_code": True,
+        }
+        component = SentenceTransformersDocumentEmbedder.from_dict(
+            {
+                "type": "haystack.components.embedders.sentence_transformers_document_embedder.SentenceTransformersDocumentEmbedder",
+                "init_parameters": init_parameters,
+            }
+        )
+        assert component.model == "model"
+        assert component.device == ComponentDevice.resolve_device(None)
+        assert component.token == Secret.from_env_var("ENV_VAR", strict=False)
+        assert component.prefix == "prefix"
+        assert component.suffix == "suffix"
+        assert component.batch_size == 64
+        assert component.progress_bar is False
+        assert component.normalize_embeddings is True
+        assert component.embedding_separator == " - "
+        assert component.trust_remote_code
+        assert component.meta_fields_to_embed == ["meta_field"]
+
     @patch(
         "haystack.components.embedders.sentence_transformers_document_embedder._SentenceTransformersEmbeddingBackendFactory"
     )

test/components/embedders/test_sentence_transformers_text_embedder.py

@@ -122,6 +122,32 @@ class TestSentenceTransformersTextEmbedder:
         assert component.normalize_embeddings is False
         assert component.trust_remote_code is False
 
+    def test_from_dict_none_device(self):
+        data = {
+            "type": "haystack.components.embedders.sentence_transformers_text_embedder.SentenceTransformersTextEmbedder",
+            "init_parameters": {
+                "token": {"env_vars": ["HF_API_TOKEN"], "strict": False, "type": "env_var"},
+                "model": "model",
+                "device": None,
+                "prefix": "",
+                "suffix": "",
+                "batch_size": 32,
+                "progress_bar": True,
+                "normalize_embeddings": False,
+                "trust_remote_code": False,
+            },
+        }
+        component = SentenceTransformersTextEmbedder.from_dict(data)
+        assert component.model == "model"
+        assert component.device == ComponentDevice.resolve_device(None)
+        assert component.token == Secret.from_env_var("HF_API_TOKEN", strict=False)
+        assert component.prefix == ""
+        assert component.suffix == ""
+        assert component.batch_size == 32
+        assert component.progress_bar is True
+        assert component.normalize_embeddings is False
+        assert component.trust_remote_code is False
+
     @patch(
         "haystack.components.embedders.sentence_transformers_text_embedder._SentenceTransformersEmbeddingBackendFactory"
     )

test/components/extractors/test_named_entity_extractor.py

@@ -40,3 +40,17 @@ def test_named_entity_extractor_serde():
     with pytest.raises(DeserializationError, match=r"Couldn't deserialize"):
         serde_data["init_parameters"].pop("backend")
         _ = NamedEntityExtractor.from_dict(serde_data)
+
+
+@pytest.mark.unit
+def test_named_entity_extractor_serde_none_device():
+    extractor = NamedEntityExtractor(
+        backend=NamedEntityExtractorBackend.HUGGING_FACE, model="dslim/bert-base-NER", device=None
+    )
+
+    serde_data = extractor.to_dict()
+    new_extractor = NamedEntityExtractor.from_dict(serde_data)
+
+    assert type(new_extractor._backend) == type(extractor._backend)
+    assert new_extractor._backend.model_name == extractor._backend.model_name
+    assert new_extractor._backend.device == extractor._backend.device

test/components/rankers/test_sentence_transformers_diversity.py

@@ -113,6 +113,37 @@ class TestSentenceTransformersDiversityRanker:
         assert ranker.meta_fields_to_embed == []
         assert ranker.embedding_separator == "\n"
 
+    def test_from_dict_none_device(self):
+        data = {
+            "type": "haystack.components.rankers.sentence_transformers_diversity.SentenceTransformersDiversityRanker",
+            "init_parameters": {
+                "model": "sentence-transformers/all-MiniLM-L6-v2",
+                "top_k": 10,
+                "device": None,
+                "similarity": "cosine",
+                "token": {"env_vars": ["HF_API_TOKEN"], "strict": False, "type": "env_var"},
+                "query_prefix": "",
+                "document_prefix": "",
+                "query_suffix": "",
+                "document_suffix": "",
+                "meta_fields_to_embed": [],
+                "embedding_separator": "\n",
+            },
+        }
+        ranker = SentenceTransformersDiversityRanker.from_dict(data)
+
+        assert ranker.model_name_or_path == "sentence-transformers/all-MiniLM-L6-v2"
+        assert ranker.top_k == 10
+        assert ranker.device == ComponentDevice.resolve_device(None)
+        assert ranker.similarity == "cosine"
+        assert ranker.token == Secret.from_env_var("HF_API_TOKEN", strict=False)
+        assert ranker.query_prefix == ""
+        assert ranker.document_prefix == ""
+        assert ranker.query_suffix == ""
+        assert ranker.document_suffix == ""
+        assert ranker.meta_fields_to_embed == []
+        assert ranker.embedding_separator == "\n"
+
     def test_to_dict_with_custom_init_parameters(self):
         component = SentenceTransformersDiversityRanker(
             model="sentence-transformers/msmarco-distilbert-base-v4",