[CHORE]: Change store_tokens to get_tokens

Author: tanujnay112

chromadb/utils/embedding_functions/chroma_bm25_embedding_function.py

@@ -22,6 +22,7 @@
 
 DEFAULT_CHROMA_BM25_STOPWORDS: List[str] = list(_DEFAULT_STOPWORDS)
 
+
 class _HashedToken:
     __slots__ = ("hash", "label")
 
@@ -47,7 +48,7 @@ class ChromaBm25Config(TypedDict, total=False):
     avg_doc_length: float
     token_max_length: int
     stopwords: List[str]
-    store_tokens: bool
+    get_tokens: bool
 
 
 class ChromaBm25EmbeddingFunction(SparseEmbeddingFunction[Documents]):
@@ -58,15 +59,15 @@ def __init__(
         avg_doc_length: float = DEFAULT_AVG_DOC_LENGTH,
         token_max_length: int = DEFAULT_TOKEN_MAX_LENGTH,
         stopwords: Optional[Iterable[str]] = None,
-        store_tokens: bool = False,
+        get_tokens: bool = False,
     ) -> None:
         """Initialize the BM25 sparse embedding function."""
 
         self.k = float(k)
         self.b = float(b)
         self.avg_doc_length = float(avg_doc_length)
         self.token_max_length = int(token_max_length)
-        self.store_tokens = bool(store_tokens)
+        self.get_tokens = bool(get_tokens)
 
         if stopwords is not None:
             self.stopwords: Optional[List[str]] = [str(word) for word in stopwords]
@@ -87,28 +88,28 @@ def _encode(self, text: str) -> SparseVector:
 
         doc_len = float(len(tokens))
         counts = Counter(
-            _HashedToken(self._hasher.hash(token), token if self.store_tokens else None)
+            _HashedToken(self._hasher.hash(token), token if self.get_tokens else None)
             for token in tokens
         )
 
         sorted_keys = sorted(counts.keys())
         indices: List[int] = []
         values: List[float] = []
-        tokens: Optional[List[str]] = [] if self.store_tokens else None
+        labels: Optional[List[str]] = [] if self.get_tokens else None
 
         for key in sorted_keys:
             tf = float(counts[key])
             denominator = tf + self.k * (
                 1 - self.b + (self.b * doc_len) / self.avg_doc_length
             )
             score = tf * (self.k + 1) / denominator
-            
+
             indices.append(key.hash)
             values.append(score)
-            if tokens is not None:
-                tokens.append(key.label)
+            if labels is not None and key.label is not None:
+                labels.append(key.label)
 
-        return SparseVector(indices=indices, values=values, labels=tokens)
+        return SparseVector(indices=indices, values=values, labels=labels)
 
     def __call__(self, input: Documents) -> SparseVectors:
         sparse_vectors: SparseVectors = []
@@ -138,7 +139,7 @@ def build_from_config(
             avg_doc_length=config.get("avg_doc_length", DEFAULT_AVG_DOC_LENGTH),
             token_max_length=config.get("token_max_length", DEFAULT_TOKEN_MAX_LENGTH),
             stopwords=config.get("stopwords"),
-            store_tokens=config.get("store_tokens", False),
+            get_tokens=config.get("get_tokens", False),
         )
 
     def get_config(self) -> Dict[str, Any]:
@@ -147,7 +148,7 @@ def get_config(self) -> Dict[str, Any]:
             "b": self.b,
             "avg_doc_length": self.avg_doc_length,
             "token_max_length": self.token_max_length,
-            "store_tokens": self.store_tokens,
+            "get_tokens": self.get_tokens,
         }
 
         if self.stopwords is not None:
@@ -158,11 +159,18 @@ def get_config(self) -> Dict[str, Any]:
     def validate_config_update(
         self, old_config: Dict[str, Any], new_config: Dict[str, Any]
     ) -> None:
-        mutable_keys = {"k", "b", "avg_doc_length", "token_max_length", "stopwords", "store_tokens"}
+        mutable_keys = {
+            "k",
+            "b",
+            "avg_doc_length",
+            "token_max_length",
+            "stopwords",
+            "get_tokens",
+        }
         for key in new_config:
             if key not in mutable_keys:
                 raise ValueError(f"Updating '{key}' is not supported for {NAME}")
 
     @staticmethod
     def validate_config(config: Dict[str, Any]) -> None:
-        validate_config_schema(config, NAME)
+        validate_config_schema(config, NAME)

chromadb/utils/embedding_functions/chroma_cloud_splade_embedding_function.py

@@ -1,5 +1,6 @@
 from chromadb.api.types import (
     SparseEmbeddingFunction,
+    SparseVector,
     SparseVectors,
     Documents,
 )
@@ -21,7 +22,7 @@ def __init__(
         self,
         api_key_env_var: str = "CHROMA_API_KEY",
         model: ChromaCloudSpladeEmbeddingModel = ChromaCloudSpladeEmbeddingModel.SPLADE_PP_EN_V1,
-        store_tokens: bool = False,
+        get_tokens: bool = False,
     ):
         """
         Initialize the ChromaCloudSpladeEmbeddingFunction.
@@ -50,7 +51,7 @@ def __init__(
                 f"or in any existing client instances"
             )
         self.model = model
-        self.store_tokens = bool(store_tokens)
+        self.get_tokens = bool(get_tokens)
         self._api_url = "https://embed.trychroma.com/embed_sparse"
         self._session = httpx.Client()
         self._session.headers.update(
@@ -89,7 +90,7 @@ def __call__(self, input: Documents) -> SparseVectors:
             "texts": list(input),
             "task": "",
             "target": "",
-            "fetch_tokens": "true" if self.store_tokens is True else "false",
+            "fetch_tokens": "true" if self.get_tokens is True else "false",
         }
 
         try:
@@ -123,14 +124,14 @@ def _parse_response(self, response: Any) -> SparseVectors:
             if isinstance(emb, dict):
                 indices = emb.get("indices", [])
                 values = emb.get("values", [])
-                raw_labels = emb.get("labels") if self.store_tokens else None
+                raw_labels = emb.get("labels") if self.get_tokens else None
                 labels: Optional[List[str]] = raw_labels if raw_labels else None
             else:
                 # Already a SparseVector, extract its data
-                assert(isinstance(emb, SparseVector))
+                assert isinstance(emb, SparseVector)
                 indices = emb.indices
                 values = emb.values
-                labels = emb.labels if self.store_tokens else None
+                labels = emb.labels if self.get_tokens else None
 
             normalized_vectors.append(
                 normalize_sparse_vector(indices=indices, values=values, labels=labels)
@@ -155,23 +156,25 @@ def build_from_config(
         return ChromaCloudSpladeEmbeddingFunction(
             api_key_env_var=api_key_env_var,
             model=ChromaCloudSpladeEmbeddingModel(model),
-            store_tokens=config.get("store_tokens", False),
+            get_tokens=config.get("get_tokens", False),
         )
 
     def get_config(self) -> Dict[str, Any]:
         return {
             "api_key_env_var": self.api_key_env_var,
             "model": self.model.value,
-            "store_tokens": self.store_tokens,
+            "get_tokens": self.get_tokens,
         }
 
     def validate_config_update(
         self, old_config: Dict[str, Any], new_config: Dict[str, Any]
     ) -> None:
-        immutable_keys = {"store_tokens", "model"}
+        immutable_keys = {"get_tokens", "model"}
         for key in immutable_keys:
             if key in new_config and new_config[key] != old_config.get(key):
-                raise ValueError(f"Updating '{key}' is not supported for chroma-cloud-splade")
+                raise ValueError(
+                    f"Updating '{key}' is not supported for chroma-cloud-splade"
+                )
 
     @staticmethod
     def validate_config(config: Dict[str, Any]) -> None:

rust/chroma/src/embed/bm25.rs

@@ -31,7 +31,7 @@ where
     H: TokenHasher,
 {
     /// Whether to store tokens in the created sparse vectors.
-    pub store_tokens: bool,
+    pub get_tokens: bool,
     /// Tokenizer for converting text into tokens.
     pub tokenizer: T,
     /// Hasher for converting tokens into u32 identifiers.
@@ -57,7 +57,7 @@ impl BM25SparseEmbeddingFunction<Bm25Tokenizer, Murmur3AbsHasher> {
     /// - hasher: Murmur3 with seed 0, abs() behavior
     pub fn default_murmur3_abs() -> Self {
         Self {
-            store_tokens: true,
+            get_tokens: true,
             tokenizer: Bm25Tokenizer::default(),
             hasher: Murmur3AbsHasher::default(),
             k: 1.2,
@@ -78,7 +78,7 @@ where
 
         let doc_len = tokens.len() as f32;
 
-        if self.store_tokens {
+        if self.get_tokens {
             let mut token_ids = Vec::with_capacity(tokens.len());
             for token in tokens {
                 let id = self.hasher.hash(&token);

schemas/embedding_functions/chroma-cloud-splade.json

@@ -17,7 +17,7 @@
             "description": "Environment variable name that contains your API key for the Chroma Embedding API",
             "default": "CHROMA_API_KEY"
         },
-        "store_tokens": {
+        "get_tokens": {
             "type": "boolean",
             "description": "Whether to store token labels in the sparse vector output",
             "default": false
@@ -28,4 +28,4 @@
         "model"
     ],
     "additionalProperties": false
-}
+}

schemas/embedding_functions/chroma_bm25.json

@@ -28,10 +28,10 @@
                 "type": "string"
             }
         },
-        "store_tokens": {
+        "get_tokens": {
             "type": "boolean",
             "description": "Whether to store token strings in the sparse vectors (default: true)"
         }
     },
     "additionalProperties": false
-}
+}