Made examples use tensorlake_function and tensorlake_router

examples/contextual_rag/workflow.py

@@ -15,7 +15,7 @@
 
 from tensorlake.functions_sdk.graph import Graph
 from tensorlake.functions_sdk.image import Image
-from tensorlake.functions_sdk.functions import indexify_function, TensorlakeCompute
+from tensorlake.functions_sdk.functions import tensorlake_function, TensorlakeCompute
 
 # TODO User set this
 contextual_retrieval_prompt = """
@@ -43,7 +43,7 @@ class ChunkContext(BaseModel):
     chunks: List[str]
     chunk_contexts: List[str]
 
-@indexify_function(image=image)
+@tensorlake_function(image=image)
 def generate_chunk_contexts(doc: str) -> ChunkContext:
     text_splitter = RecursiveCharacterTextSplitter(
         chunk_size=750,

examples/ditributed_map.py

@@ -1,5 +1,5 @@
 from pydantic import BaseModel
-from tensorlake import tensorlake_function, indexify_router, Graph
+from tensorlake import tensorlake_function, tensorlake_router, Graph
 from typing import List, Union
 
 @tensorlake_function()

examples/knowledge_graph/workflow.py

@@ -4,7 +4,7 @@
 from pydantic import BaseModel, Field
 from tensorlake import RemoteGraph
 from tensorlake.functions_sdk.graph import Graph
-from tensorlake.functions_sdk.functions import indexify_function, TensorlakeCompute
+from tensorlake.functions_sdk.functions import tensorlake_function, TensorlakeCompute
 from tensorlake.functions_sdk.image import Image
 from neo4j import GraphDatabase
 import json
@@ -187,7 +187,7 @@ def run(self, data: Tuple[List[Entity], str, Document]) -> Tuple[List[Entity], L
             logging.error(f"Error in extract_relationships: {str(e)}")
             raise
 
-@indexify_function(image=base_image)
+@tensorlake_function(image=base_image)
 def build_knowledge_graph(data: Tuple[List[Entity], List[Relationship], Document]) -> KnowledgeGraphOutput:
     try:
         entities, relationships, doc = data
@@ -199,7 +199,7 @@ def build_knowledge_graph(data: Tuple[List[Entity], List[Relationship], Document
         logging.error(f"Error in build_knowledge_graph: {str(e)}", "knowledge_graph_output --->")
         raise
 
-@indexify_function(image=neo4j_image)
+@tensorlake_function(image=neo4j_image)
 def store_in_neo4j(data: KnowledgeGraphOutput) -> bool:
     try:
         kg = data.knowledge_graph
@@ -226,7 +226,7 @@ def store_in_neo4j(data: KnowledgeGraphOutput) -> bool:
         logging.error(f"Error in store_in_neo4j: {str(e)}")
         raise
 
-@indexify_function(image=embedding_image)
+@tensorlake_function(image=embedding_image)
 def generate_embeddings(data: KnowledgeGraphOutput) -> TextChunk:
     try:
         doc = data.document
@@ -246,7 +246,7 @@ def generate_embeddings(data: KnowledgeGraphOutput) -> TextChunk:
         logging.error(f"Error in generate_embeddings: {str(e)}")
         raise
 
-@indexify_function(image=gemini_image)
+@tensorlake_function(image=gemini_image)
 def question_to_cypher(question: Question) -> CypherQueryAndQuestion:
     try:
         model = genai.GenerativeModel("gemini-pro")
@@ -274,7 +274,7 @@ def question_to_cypher(question: Question) -> CypherQueryAndQuestion:
         logging.error(f"Error in question_to_cypher: {str(e)}")
         raise
 
-@indexify_function(image=neo4j_image)
+@tensorlake_function(image=neo4j_image)
 def execute_cypher_query(data: CypherQueryAndQuestion) -> QuestionAndResult:
     cypher_query, question = data.cypher_query, data.question
     uri = os.getenv('NEO4J_URI', "bolt://localhost:7687")
@@ -294,7 +294,7 @@ def execute_cypher_query(data: CypherQueryAndQuestion) -> QuestionAndResult:
 
     return QuestionAndResult(question=question, query_result=QueryResult(result=records))
 
-@indexify_function(image=gemini_image)
+@tensorlake_function(image=gemini_image)
 def generate_answer(data: QuestionAndResult) -> Answer:
     query_result, question = data.query_result, data.question
 

examples/object_detection/workflow.py

@@ -3,7 +3,7 @@
 from tensorlake.functions_sdk.data_objects import File
 from tensorlake.functions_sdk.functions import (
     TensorlakeCompute,
-    indexify_function,
+    tensorlake_function,
 )
 from pydantic import BaseModel
 from typing import List

examples/pdf_document_extraction/embedding.py

@@ -1,14 +1,14 @@
 from typing import Any, List
 
-from tensorlake.functions_sdk.functions import TensorlakeCompute, indexify_function
+from tensorlake.functions_sdk.functions import TensorlakeCompute, tensorlake_function
 from sentence_transformers import SentenceTransformer
 from common_objects import ImageWithEmbedding, TextChunk, PDFParserDoclingOutput
 from inkwell.api.document import Document
 from inkwell.api.page import PageFragmentType
 import base64
 from images import st_image
 
-@indexify_function(image=st_image)
+@tensorlake_function(image=st_image)
 def chunk_text(document: Document) -> List[TextChunk]:
     """
     Extract chunks from document
@@ -40,7 +40,7 @@ def chunk_text(document: Document) -> List[TextChunk]:
     return chunks
 
 
-@indexify_function(image=st_image)
+@tensorlake_function(image=st_image)
 def chunk_text_docling(document: PDFParserDoclingOutput) -> List[TextChunk]:
     """
     Extract chunks from documents

examples/pdf_document_extraction/workflow.py

@@ -3,10 +3,10 @@
 from tensorlake import RemoteGraph
 from tensorlake.functions_sdk.data_objects import File
 from tensorlake.functions_sdk.graph import Graph
-from tensorlake.functions_sdk.functions import indexify_function
+from tensorlake.functions_sdk.functions import tensorlake_function
 from images import http_client_image
 
-@indexify_function(image=http_client_image)
+@tensorlake_function(image=http_client_image)
 def download_pdf(url: str) -> File:
     """
     Download pdf from url

examples/pdf_structured_extraction/document_ai_api_version_workflow.py

@@ -4,7 +4,7 @@
 from decimal import Decimal
 from datetime import date
 from tensorlake.functions_sdk.data_objects import File
-from tensorlake import indexify_function, Graph, RemoteGraph, Image
+from tensorlake import tensorlake_function, Graph, RemoteGraph, Image
 from typing import Optional, List, Any
 
 
@@ -18,7 +18,7 @@
 )
 
 
-@indexify_function(image=image)
+@tensorlake_function(image=image)
 def tensorlake_document_ai_parse(file: File) -> str:
     """
     Call the inkwell parse API. TODO replace the token below for `Bearer`.
@@ -120,7 +120,7 @@ def _call_oai_client(system_prompt: str, user_prompt: str, markdown: str) -> str
     return response.choices[0].message.content
 
 
-@indexify_function(image=image)
+@tensorlake_function(image=image)
 def extract_with_oai(markdown: str) -> BillSchema:
     schema = BillSchema.model_json_schema()
 

examples/pdf_structured_extraction/workflow.py

@@ -3,7 +3,7 @@
 from datetime import date
 import tempfile
 from tensorlake.functions_sdk.data_objects import File
-from tensorlake import indexify_function, Graph, RemoteGraph, Image
+from tensorlake import tensorlake_function, Graph, RemoteGraph, Image
 from typing import Optional
 
 image = (
@@ -37,7 +37,7 @@ class Config:
             Decimal: lambda v: str(v)
         }
 
-@indexify_function(image=image)
+@tensorlake_function(image=image)
 def parse_pdf(file: File) -> BillSchema:
     from inkwell.ocr import OCRType, OCRFactory
     from inkwell.io import read_pdf_as_images
@@ -61,7 +61,7 @@ def parse_pdf(file: File) -> BillSchema:
         print(result)
         return BillSchema.model_validate_json(result)
 
-@indexify_function(image=image)
+@tensorlake_function(image=image)
 def write_to_db(bill: BillSchema) -> None:
     from sqlmodel import Field, SQLModel, Session, create_engine
     from typing import Optional

examples/readme/distributed_map.py

@@ -1,12 +1,12 @@
 from pydantic import BaseModel
-from tensorlake import indexify_function, indexify_router, Graph
+from tensorlake import tensorlake_function, tensorlake_router, Graph
 from typing import List, Union
 
-@indexify_function()
+@tensorlake_function()
 def generate_sequence(a: int) -> List[int]:
     return [i for i in range(a)]
 
-@indexify_function()
+@tensorlake_function()
 def squared(x: int) -> int:
     return x * x
 

examples/readme/readme_example.py

@@ -1,19 +1,19 @@
 from pydantic import BaseModel
-from tensorlake import indexify_function, Graph
+from tensorlake import tensorlake_function, Graph
 from typing import List
 
 class Total(BaseModel):
     val: int = 0
 
-@indexify_function()
+@tensorlake_function()
 def generate_numbers(a: int) -> List[int]:
     return [i for i in range(a)]
 
-@indexify_function()
+@tensorlake_function()
 def square(x: int) -> int:
     return x ** 2
 
-@indexify_function(accumulate=Total)
+@tensorlake_function(accumulate=Total)
 def add(total: Total, new: int) -> Total:
     total.val += new
     return total

examples/readme/website.py

@@ -1,15 +1,15 @@
-from tensorlake import indexify_function, Graph
+from tensorlake import tensorlake_function, Graph
 from pydantic import BaseModel
 
 class Audio(BaseModel):
     file: bytes
 
-@indexify_function()
+@tensorlake_function()
 def scrape_website(url: str) -> str:
     import requests
     return requests.get(f"http://r.jina.ai/{url}").text
 
-@indexify_function()
+@tensorlake_function()
 def summarize_text(text: str) -> str:
     from openai import OpenAI
     completion = OpenAI().chat.completions.create(
@@ -21,7 +21,7 @@ def summarize_text(text: str) -> str:
     )
     return completion.choices[0].message.content
 
-@indexify_function()
+@tensorlake_function()
 def create_audio(summary: str) -> Audio:
     import elevenlabs
     from elevenlabs import save

examples/tweetsgenerator/workflow.py

@@ -5,7 +5,7 @@
 from pydantic import BaseModel, Field
 from tensorlake import RemoteGraph
 from tensorlake.functions_sdk.graph import Graph
-from tensorlake.functions_sdk.functions import indexify_function
+from tensorlake.functions_sdk.functions import tensorlake_function
 from tensorlake.functions_sdk.image import Image
 
 # Set up logging
@@ -51,7 +51,7 @@ class RankedTweets(BaseModel):
 )
 
 
-@indexify_function(image=openai_image_3_10)
+@tensorlake_function(image=openai_image_3_10)
 def generate_tweet_topics(subject: str) -> List[str]:
     """Generate topics for tweets about a given subject."""
     import openai
@@ -74,7 +74,7 @@ class Topics(BaseModel):
     topics = response.choices[0].message.parsed
     return topics.topics
 
-@indexify_function(image=openai_image_3_10)
+@tensorlake_function(image=openai_image_3_10)
 def generate_tweet(topic: str) -> str:
     """Generate a tweet about a given topic."""
     import openai
@@ -96,13 +96,13 @@ class Tweet(BaseModel):
     tweet = response.choices[0].message.parsed
     return tweet.tweet
 
-@indexify_function(image=base_image_3_10, accumulate=Tweets)
+@tensorlake_function(image=base_image_3_10, accumulate=Tweets)
 def accumulate_tweets(acc: Tweets, tweet: str) -> Tweets:
     """Accumulate generated tweets."""
     acc.tweets.append(tweet)
     return acc
 
-@indexify_function(image=openai_image_3_10)
+@tensorlake_function(image=openai_image_3_10)
 def score_and_rank_tweets(tweets: Tweets) -> RankedTweets:
     """Score and rank the accumulated tweets."""
     import openai

examples/video_summarization/workflow.py

@@ -7,7 +7,7 @@
 from tensorlake.functions_sdk.data_objects import File
 from tensorlake.functions_sdk.graph import Graph
 from tensorlake.functions_sdk.image import Image
-from tensorlake.functions_sdk.functions import indexify_function, indexify_router
+from tensorlake.functions_sdk.functions import tensorlake_function, tensorlake_router
 
 # Set up logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
@@ -58,7 +58,7 @@ class Summary(BaseModel):
 )
 
 # Indexify Functions
-@indexify_function(image=yt_downloader_image)
+@tensorlake_function(image=yt_downloader_image)
 def download_youtube_video(url: YoutubeURL) -> List[File]:
     """Download the YouTube video from the URL."""
     from pytubefix import YouTube
@@ -68,14 +68,14 @@ def download_youtube_video(url: YoutubeURL) -> List[File]:
     logging.info("Video downloaded")
     return [File(data=content, mime_type="video/mp4")]
 
-@indexify_function(image=audio_image)
+@tensorlake_function(image=audio_image)
 def extract_audio_from_video(file: File) -> File:
     """Extract the audio from the video."""
     from pydub import AudioSegment
     audio = AudioSegment.from_file(file.data)
     return File(data=audio.export("audio.wav", format="wav").read(), mime_type="audio/wav")
 
-@indexify_function(image=transcribe_image)
+@tensorlake_function(image=transcribe_image)
 def transcribe_audio(file: File) -> Transcription:
     """Transcribe audio and diarize speakers."""
     from faster_whisper import WhisperModel
@@ -84,7 +84,7 @@ def transcribe_audio(file: File) -> Transcription:
     audio_segments = [SpeechSegment(text=segment.text, start_ts=segment.start, end_ts=segment.end) for segment in segments]
     return Transcription(segments=audio_segments)
 
-@indexify_function(image=llama_cpp_image)
+@tensorlake_function(image=llama_cpp_image)
 def classify_meeting_intent(speech: Transcription) -> Transcription:
     try:
         """Classify the intent of the audio."""
@@ -125,7 +125,7 @@ def classify_meeting_intent(speech: Transcription) -> Transcription:
         speech.classification = SpeechClassification(classification="unknown", confidence=0.5)
         return speech
 
-@indexify_function(image=llama_cpp_image)
+@tensorlake_function(image=llama_cpp_image)
 def summarize_job_interview(speech: Transcription) -> Summary:
     """Summarize the job interview."""
     from llama_cpp import Llama
@@ -151,7 +151,7 @@ def summarize_job_interview(speech: Transcription) -> Summary:
     output = model(prompt=prompt, max_tokens=30000, stop=["\n"])
     return Summary(summary=output["choices"][0]["text"])
 
-@indexify_function(image=llama_cpp_image)
+@tensorlake_function(image=llama_cpp_image)
 def summarize_sales_call(speech: Transcription) -> Summary:
     """Summarize the sales call."""
     from llama_cpp import Llama
@@ -177,7 +177,7 @@ def summarize_sales_call(speech: Transcription) -> Summary:
     output = model(prompt=prompt, max_tokens=30000, stop=["\n"])
     return Summary(summary=output["choices"][0]["text"])
 
-@indexify_router(image=base_image)
+@tensorlake_router(image=base_image)
 def route_transcription_to_summarizer(speech: Transcription) -> List[Union[summarize_job_interview, summarize_sales_call]]:
     """Route the transcription to the appropriate summarizer based on the classification."""
     if speech.classification.classification == "job-interview":