Merge branch 'main' into PHS-993-os-error-source-code-not-available-w…

…hen-trying-to-push-benchmark-from-jupyter-notebook

CHANGELOG.md

@@ -10,10 +10,13 @@
 - Add `create_project` bool to `StudioClient.__init__()` to enable users to automatically create their Studio projects
 - Add progressbar to the `Runner` to be able to track the `Run`
 - Add `StudioClient.submit_benchmark_lineages` function and include it in `StudioClient.submit_benchmark_execution`
+
+#### DocumentIndexClient 
 - Add method `DocumentIndexClient.chunks()` for retrieving all text chunks of a document.
+- Add metadata filter `FilterOps.IS_NULL`, that allows to filter fields based on whether their value is null.
 
 ### Fixes
-...
+- The Document Index `SearchQuery` now correctly allows searches with a negative `min_score`.
 
 ### Deprecations
 ...

pyproject.toml

@@ -61,7 +61,7 @@ pandas-stubs = "^2.2.3.241126"
 # docs
 sphinx-rtd-theme = "^3.0.2"
 sphinx = "^8.1.3"
-matplotlib = "^3.9.2"
+matplotlib = "^3.10.0"
 
 [tool.poetry.group.docs.dependencies]
 fastapi = "^0.115.5"

src/documentation/elo_qa_eval.ipynb

@@ -448,7 +448,7 @@
    "outputs": [],
    "source": [
     "newly_added_models = [\n",
-    "    Llama3InstructModel(name=\"llama-3.1-70b-instruct\", client=aa_client),\n",
+    "    Llama3InstructModel(name=\"llama-3.3-70b-instruct\", client=aa_client),\n",
     "]\n",
     "\n",
     "for model in newly_added_models:\n",

src/intelligence_layer/connectors/document_index/document_index.py

@@ -227,6 +227,7 @@ class FilterOps(Enum):
     BEFORE = "before"
     AT_OR_BEFORE = "at_or_before"
     EQUAL_TO = "equal_to"
+    IS_NULL = "is_null"
 
 
 class FilterField(BaseModel):
@@ -293,16 +294,15 @@ class SearchQuery(BaseModel):
         query: Actual text to be searched with.
         max_results: Max number of search results to be retrieved by the query.
             Must be larger than 0.
-        min_score: Filter out results with a similarity score below this value.
-            Must be between 0 and 1.
-            For searches on hybrid indexes, the Document Index applies the min_score
-            to the semantic results before fusion of result sets. As fusion re-scores results,
+        min_score: Filter out results with a similarity score below this value. Must be between
+            -1 and 1. For searches on hybrid indexes, the Document Index applies the min_score to
+            the semantic results before fusion of result sets. As fusion re-scores results,
             returned scores may exceed this value.
     """
 
     query: str
     max_results: int = Field(ge=0, default=1)
-    min_score: float = Field(ge=0.0, le=1.0, default=0.0)
+    min_score: float = Field(ge=-1.0, le=1.0, default=0.0)
     filters: Optional[list[Filters]] = None
 
 

src/intelligence_layer/core/model.py

@@ -261,7 +261,7 @@ def __init__(
         )
         if name not in [model["name"] for model in self._client.models()]:
             warnings.warn(
-                "The provided model is not a recommended model for this model class."
+                "The provided model is not a recommended model for this model class. "
                 "Make sure that the model you have selected is suited to be use for the prompt template used in this model class."
             )
         self._complete: Task[CompleteInput, CompleteOutput] = _Complete(
@@ -414,7 +414,7 @@ def __init__(
     ) -> None:
         if name not in self.RECOMMENDED_MODELS or name == "":
             warnings.warn(
-                "The provided model is not a recommended model for this model class."
+                "The provided model is not a recommended model for this model class. "
                 "Make sure that the model you have selected is suited to be use for the prompt template used in this model class."
             )
         super().__init__(name, client)

tests/conftest_document_index.py

@@ -158,6 +158,7 @@ def document_contents_with_metadata() -> list[DocumentContents]:
 
     metadata_1: JsonSerializable = {
         "string-field": "example_string_1",
+        "option-field": None,
         "integer-field": 123,
         "float-field": 123.45,
         "boolean-field": True,
@@ -168,6 +169,7 @@ def document_contents_with_metadata() -> list[DocumentContents]:
 
     metadata_2: JsonSerializable = {
         "string-field": "example_string_2",
+        "option-field": "example_string_2",
         "integer-field": 456,
         "float-field": 678.90,
         "boolean-field": False,
@@ -178,6 +180,7 @@ def document_contents_with_metadata() -> list[DocumentContents]:
 
     metadata_3: JsonSerializable = {
         "string-field": "example_string_3",
+        "option-field": "example_string_3",
         "integer-field": 789,
         "float-field": 101112.13,
         "boolean-field": True,
@@ -237,6 +240,10 @@ def filter_index_configs(
             "field-name": "string-field",
             "field-type": "string",
         },
+        random_identifier(): {
+            "field-name": "option-field",
+            "field-type": "string",
+        },
         random_identifier(): {
             "field-name": "integer-field",
             "field-type": "integer",