add-azureembedder-cookbook-example-phi-1918

cookbook/embedders/azure_embedder.py

@@ -0,0 +1,19 @@
+from phi.agent import AgentKnowledge
+from phi.vectordb.pgvector import PgVector
+from phi.embedder.azure_openai import AzureOpenAIEmbedder
+
+embeddings = AzureOpenAIEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
+knowledge_base = AgentKnowledge(
+    vector_db=PgVector(
+        db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
+        table_name="azure_openai_embeddings",
+        embedder=AzureOpenAIEmbedder(),
+    ),
+    num_documents=2,
+)

cookbook/embedders/fireworks_embedder.py

@@ -1,27 +1,19 @@
-import os
-
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.fireworks import FireworksEmbedder
 
-# Create knowledge base
+embeddings = FireworksEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="fireworks_embeddings",
-        embedder=FireworksEmbedder(
-            api_key=os.getenv("FIREWORKS_API_KEY"),
-            dimensions=768,
-        ),
+        embedder=FireworksEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/gemini_embedder.py

@@ -1,27 +1,19 @@
-import os
-
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.google import GeminiEmbedder
 
-# Create knowledge base
+embeddings = GeminiEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="gemini_embeddings",
-        embedder=GeminiEmbedder(
-            api_key=os.getenv("GOOGLE_API_KEY"),
-            dimensions=768,
-        ),
+        embedder=GeminiEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/huggingface_embedder.py

@@ -0,0 +1,19 @@
+from phi.agent import AgentKnowledge
+from phi.vectordb.pgvector import PgVector
+from phi.embedder.huggingface import HuggingfaceCustomEmbedder
+
+embeddings = HuggingfaceCustomEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
+knowledge_base = AgentKnowledge(
+    vector_db=PgVector(
+        db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
+        table_name="huggingface_embeddings",
+        embedder=HuggingfaceCustomEmbedder(),
+    ),
+    num_documents=2,
+)

cookbook/embedders/mistral_embedder.py

@@ -1,26 +1,19 @@
-import os
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.mistral import MistralEmbedder
 
-# Create knowledge base
+embeddings = MistralEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="mistral_embeddings",
-        embedder=MistralEmbedder(
-            api_key=os.getenv("MISTRAL_API_KEY"),
-            dimensions=1024,
-        ),
+        embedder=MistralEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/ollama_embedder.py

@@ -1,25 +1,19 @@
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.ollama import OllamaEmbedder
 
-# Create knowledge base
+embeddings = OllamaEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="ollama_embeddings",
-        embedder=OllamaEmbedder(
-            model="openhermes",
-            dimensions=4096,
-        ),
+        embedder=OllamaEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/openai_embedder.py

@@ -1,27 +1,19 @@
-import os
-
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.openai import OpenAIEmbedder
 
-# Create knowledge base
+embeddings = OpenAIEmbedder().get_embedding("Embed me")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="openai_embeddings",
-        embedder=OpenAIEmbedder(
-            api_key=os.getenv("OPENAI_API_KEY"),
-            dimensions=1536,
-        ),
+        embedder=OpenAIEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/qdrant_fastembed.py

@@ -1,20 +1,19 @@
+from phi.agent import AgentKnowledge
+from phi.vectordb.pgvector import PgVector
 from phi.embedder.fastembed import FastEmbedEmbedder
-from typing import List
 
-documents: List[str] = [
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta",
-    "with crispy pancetta, which adds a savory crunch and depth of flavor,",
-    "creamy eggs that create a luscious sauce when mixed with the hot pasta,",
-    "and a generous sprinkle of freshly grated Parmesan cheese",
-    "for a comforting, flavorful dish that's sure to impress any pasta lover.",
-    "Finish with a dash of black pepper and a garnish of parsley for a touch of freshness.",
-]
+embeddings = FastEmbedEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
 
-"""FastEmbed supported models can be found here: https://qdrant.github.io/fastembed/examples/Supported_Models/"""
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
 
-embedder = FastEmbedEmbedder(
-    # model="BAAI/bge-small-en-v1.5"
+# Example usage:
+knowledge_base = AgentKnowledge(
+    vector_db=PgVector(
+        db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
+        table_name="qdrant_embeddings",
+        embedder=FastEmbedEmbedder(),
+    ),
+    num_documents=2,
 )
-
-embeddings = embedder.get_embedding("\n".join(documents))
-print(embeddings[0])

cookbook/embedders/sentence_transformer_embedder.py

@@ -0,0 +1,19 @@
+from phi.agent import AgentKnowledge
+from phi.vectordb.pgvector import PgVector
+from phi.embedder.sentence_transformer import SentenceTransformerEmbedder
+
+embeddings = SentenceTransformerEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
+knowledge_base = AgentKnowledge(
+    vector_db=PgVector(
+        db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
+        table_name="sentence_transformer_embeddings",
+        embedder=SentenceTransformerEmbedder(),
+    ),
+    num_documents=2,
+)

cookbook/embedders/together_embedder.py

@@ -1,27 +1,19 @@
-import os
-
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.together import TogetherEmbedder
 
-# Create knowledge base
+embeddings = TogetherEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="together_embeddings",
-        embedder=TogetherEmbedder(
-            api_key=os.getenv("TOGETHER_API_KEY"),
-            dimensions=768,
-        ),
+        embedder=TogetherEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

cookbook/embedders/voyageai_embedder.py

@@ -1,27 +1,19 @@
-import os
-
-from phi.agent import Agent, AgentKnowledge
+from phi.agent import AgentKnowledge
 from phi.vectordb.pgvector import PgVector
 from phi.embedder.voyageai import VoyageAIEmbedder
 
-# Create knowledge base
+embeddings = VoyageAIEmbedder().get_embedding("The quick brown fox jumps over the lazy dog.")
+
+# Print the embeddings and their dimensions
+print(f"Embeddings: {embeddings[:5]}")
+print(f"Dimensions: {len(embeddings)}")
+
+# Example usage:
 knowledge_base = AgentKnowledge(
     vector_db=PgVector(
         db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
         table_name="voyageai_embeddings",
-        embedder=VoyageAIEmbedder(
-            api_key=os.getenv("VOYAGE_API_KEY"),
-            dimensions=1024,
-        ),
+        embedder=VoyageAIEmbedder(),
     ),
     num_documents=2,
 )
-
-# Add information to the knowledge base
-knowledge_base.load_text(
-    "This classic spaghetti carbonara combines perfectly cooked al dente pasta "
-    "with crispy pancetta and creamy eggs that create a luscious sauce."
-)
-
-# Add the knowledge base to the Agent
-agent = Agent(knowledge_base=knowledge_base)

phi/embedder/huggingface.py

@@ -1,3 +1,5 @@
+import json
+from os import getenv
 from typing import Any, Dict, List, Optional, Tuple
 
 from phi.embedder.base import Embedder
@@ -14,7 +16,7 @@ class HuggingfaceCustomEmbedder(Embedder):
     """Huggingface Custom Embedder"""
 
     model: str = "jinaai/jina-embeddings-v2-base-code"
-    api_key: Optional[str] = None
+    api_key: Optional[str] = getenv("HUGGINGFACE_API_KEY")
     client_params: Optional[Dict[str, Any]] = None
     huggingface_client: Optional[InferenceClient] = None
 
@@ -39,7 +41,9 @@ def _response(self, text: str):
     def get_embedding(self, text: str) -> List[float]:
         response = self._response(text=text)
         try:
-            return response
+            decoded_string = response.decode("utf-8")
+            return json.loads(decoded_string)
+
         except Exception as e:
             logger.warning(e)
             return []