Chroma

Simple Wonders of RAG using Ollama, Langchain and ChromaDB

Hello, readers; I would like to extend a warm welcome to all of you for reading my blog post. This particular post is my second one, focusing on vector databases. In my previous , I covered the concepts of vectors, vector databases, and vector search. I also provided a practical demonstration of the FAISS vector store using Python. Similarly, in this article, we will delve into another popular vector store known as ChromaDB. blog post Introduction to ChromaDB In this part of the article, we will explore what ChromaDB is, its features, and how it works in real-time. What is ChromaDB? is a vector store that is open-source and is utilized for the storage and retrieval of vector embeddings. Its primary purpose is to store embeddings and associated metadata for future use by extensive language models. Furthermore, it can also be employed for semantic search engines that operate on text data. Chroma DB What are the features of ChromaDB? ChromaDB offers various storage options, such as DuckDB for standalone use or ClickHouse for scalability. It provides Python and JavaScript/TypeScript SDKs and emphasizes simplicity, speed, and analysis capabilities. ChromaDB can store vectors with additional metadata and allows for filtering during the query search on the vector database. Chroma DB provides various options for storing vector embeddings. The options include storing the vector database , where it is flushed when the RAM is refreshed. in-memory Storing it on the and loading it into memory when needed. local file system Another option is to host the database on a machine, allowing to make requests to the server for each query. server clients How does ChromaDB work? The initial step in vector databases is to create a collection, which is similar to a table in a relational database. By default, ChromaDB utilizes the model for generating embeddings. The embedding model can be customized according to the user's requirements. all-MiniLM-L6-V2 Documents can be added to the collection, and if they are in text format, ChromaDB will automatically convert them into embeddings based on the specified embedding model. It is optional to include meta information when adding a document, but a unique document ID must be provided for identification purposes. After creating and populating the collection with documents, we can begin querying the vector store to retrieve similar documents. The process of filtering the documents while querying is referred to as , and it is also available as an option in ChromaDB. meta-filtering In the next part, we will use Chroma and all-MiniLM-L6-V2 to create our own vector DB. Getting started with ChromaDB In this section, we will create a vector store, add collections, add text to the collection, and perform a query search with and without meta-filtering using in-memory ChromaDB. As the first step, we will try installing the ChromaDB package. !pip3 install chromadb Once the package is installed, try importing the required packages into your working notebook. import chromadb
import pandas as pd
import numpy as np After importing the packages, we will now have a look into our dataset, which we feed to our vector store for training. passage_data = pd.read_csv("/Users/Python_Programs/Text_Similarity/Stanford_Edited.csv")
passage_data For this tutorial, I have selected the Stanford dataset. The dataset has 37,792 rows of data and 3 columns named: Id: Unique id to represent each document in the data store. Content: The content of each document. Language: The language in which the content is written (for filtering purpose). For meta-filtering, we will have the language column in the dataset with values en_US as English and ru_RU as Russian. The data count in each of these labels is as follows: Having explored the dataset, now it is time to prepare the data in the format that the ChromaDB accepts. To add a set of documents to the Chroma index, documents - list of documents(text) metadatas - list of dictionary that has unique ids to represent the documents (ids should be in String format) documents = passage_data['Content'].tolist()
metadatas = []
ids = []

for idx in range(len(passage_data)):
    metadatas.append({"Language": passage_data['Language'][idx]})
    ids.append(str(passage_data['Id'][idx])) The dataset and input documents with metadata and IDs are prepared. We will create a client connection to ChromaDB. client = chromadb.Client() Once the client is initialized, we create a new collection to hold the vectors(data) as an index. collection = client.create_collection("Stanford_Collection") It is time to add the documents to the index. collection.add(
    documents=documents,
    metadatas=metadatas,
    ids=ids
) There are two approaches to querying the index: querying without filters and querying with filters (known as meta-filtering). We will now attempt both methods. Attempting a query without any filters. Attempting a query with filters (meta-filtering). Now, we will test both approaches. # 1. query without any filters
results = collection.query(
    query_texts=["University of Architecture"],
    n_results=3,
)
print(results) The output displays three documents that have similarities to the query document ("University of Architecture"). It includes the document IDs, distances, and metadata of these documents. Now, we will try a query with a language filter. results = collection.query(
    query_texts=["University of Architecture"],
    n_results=3,
    where={"Language": "en_US"}, # optional filter
)
print(results) The output displays three documents that are in the English language for the given query. This is how meta-filtering works. Few other functionalities Having seen the main functionalities of ChromaDB, we will now explore a few other functionalities in short. To update a document in the Chroma Index: collection.update(
    ids=["2"],
    documents=["This document contains information about medical college"],
    metadatas=[{"Language": "en_US"}],
) There is slight difference between update() and upsert() in ChromaDB. In short, if you are trying to update a document which does not exist in the Index: update() raises a exception, whereas upsert() tries to insert the document as new vector into the index. To delete a document from the Chroma Index: collection.delete(ids = ['1']) To get the count of documents present in an index, we use the following: print(collection.count()) To get the documents present in the collection, we use: print(collection.get()) To modify the name of the existing collection: collection.modify(name="new_collection_name") To delete or drop an Index in ChromeDB, we use: client.delete_collection(name="new_collection_name") Conclusion This article has presented an introduction to Chroma, a Vector Database that is open source. We have discussed the process of creating collections in Chroma and adding documents to them. Furthermore, we have examined how Chroma DB generates vector embeddings from the data. Lastly, we have gained knowledge on how to retrieve relevant information from a specific collection in the Vector Store based on a given query. I hope this blog post has aided you in exploring another vector database and understanding the concepts clearly. I look forward to seeing you on my next blog. Until then, happy learning 👍🏻.

Vector Databases: Getting Started With ChromaDB and More

ভেক্টর ডেটাবেস: ChromaDB এবং আরও অনেক কিছু দিয়ে শুরু করা

Vektordatenbanken: Erste Schritte mit ChromaDB und mehr

Bases de datos vectoriales: introducción a ChromaDB y más

Bases de données vectorielles : premiers pas avec ChromaDB et plus

वेक्टर डेटाबेस: ChromaDB और अन्य के साथ शुरुआत करना

ベクター データベース: ChromaDB の概要など

벡터 데이터베이스: ChromaDB 시작하기 등

Bancos de dados vetoriais: primeiros passos com ChromaDB e muito mais

Векторные базы данных: начало работы с ChromaDB и многое другое

Vektör Veritabanları: ChromaDB'ye Başlarken ve Daha Fazlası

Cơ sở dữ liệu vectơ: Bắt đầu với ChromaDB và hơn thế nữa

矢量数据库：ChromaDB 等入门

HackerNoon is proud to partner with and for the , taking place in December at the Estufa Fria in Lisbon, Portugal. The venue will be turned into a canvas for a seamless blend of physical and digital art, showcasing how nature, art, and technology can work hand in hand. muti Artsies Collective ChromaFlora Exhibition Immersive artistry in the green Organized by and , the ChromaFlora Exhibition will unite over 70 global and local artists with 50+ digital and 20+ physical artworks, each contributing to the overarching theme of nature, art and technology. Blending the digital with the physical, all artworks will also be available as NFTs minted on Polygon through the marketplace . muti Artsies Collective Bueno.art muti is a cultural association and DAO based in Portugal, supporting emerging artists of all art forms. The physical artworks include pieces from rising visual artists such as , , , and . Daphne Marzia and The Spaceshipproject Arta Raituma Carlotta Premazzi Katherine Shapiro On the digital side, international artists have been invited to participate through an open call on . The sister-led Artsies Collective is curating 50+ artworks, which will be exhibited during the week in the greenhouse. The Hug The Estufa, a haven of lush greenery, provides the perfect backdrop for the exhibition, creating an environment that enhances the connection between art and nature. Attendees will have the opportunity to explore a curated selection of artworks whilst also enjoying the natural beauty of the urban jungle of Lisbon. This innovative showcase invites you to embark on a transformative journey, where the lines between nature and technology blur into enchanting hybrids. "ChromaFlora" stands as a testament to the symbiotic relationship between organic materials and cutting-edge technology. The exhibition takes pride in celebrating the inherent beauty of nature while embracing the limitless possibilities that technology offers to artistic expression. This harmonious blend creates a unique space where the natural and the digital coalesce in a mesmerizing display of creativity. Technology with a positive impact To enrich the visitor experience, on the weekend of the 15th and 16th of December visitors can join talks and workshops held by artists and projects that work in the intersection of sustainability, art, and tech. Experts and artists will share insights into how these seemingly disparate elements can coexist harmoniously, inspiring attendees to explore the possibilities of art in the digital age. The talks and workshops aim to spark dialogue and foster a deeper understanding of the relationship between the natural world and the technological advancements shaping our future. The well-curated program will give a glimpse of how technology can be used in a meaningful way, from the use of NFTs to nature-backed economy ( / ) and vertical farming ( ). Projects like , and will introduce their work and visitors will get the chance to meet the physical artists and the process behind their creation. Oasa TDF Raíz Zeitls Muro Dvue X2 Digital artworks & artists Eco Enigma by Nataliya Grimberg When nature weaves its tendrils through the circuits of our creations, a symphony of sustainability is born. In this fusion, the ethereal dance of green and gear, we uncover an enchanting truth: that the harmony of our world lies in the embrace of leaf and line, root and router. 'EcoEnigma' is a testament to the magic that sparks when the pulse of the earth meets the hum of progress, inspiring us to envision a future where technology blooms from the soil of our respect for the natural world. Lumiflora Blooms by DOLORES KRAJNAK Photograph of a flower was used as part of the original prompt. The verbal prompt was revised several times. Variations were generated until the image and colors were achieved. Program "ChromaFlora" will be open to the public from Tuesday, December 12, to Sunday, December 17, 2023, with daily opening hours from 9 am to 5 pm. Admission is priced at 3.25€, providing access to a week-long exploration of the captivating fusion between nature and technology at the estufa. 11.12.: Exclusive Vernissage 12.-17.12.: Exhibition open to public 13.12.: Web3Wednesday with Chilibangs 15.&16.12.: Workshops & Talks from 12-17:00 16.12: Afterparty For more info visit https://chromaflora.xyz/ Facebook Event: https://www.facebook.com/events/666284942276984

ChromaFlora - a Hybrid Exhibition in the Estufa Fria in Lisbon: Where art meets technology in nature

Flatland: A Romance of Many Dimensions by Edwin Abbott Abbott is part of the HackerNoon Books Series. You can jump to any chapter in this book . Of the Suppression of the Chromatic Sedition here Of the Suppression of the Chromatic Sedition The agitation for the Universal Colour Bill continued for three years; and up to the last moment of that period it seemed as though Anarchy were destined to triumph. A whole army of Polygons, who turned out to fight as private soldiers, was utterly annihilated by a superior force of Isosceles Triangles—the Squares and Pentagons meanwhile remaining neutral. Worse than all, some of the ablest Circles fell a prey to conjugal fury. Infuriated by political animosity, the wives in many a noble household wearied their lords with prayers to give up their opposition to the Colour Bill; and some, finding their entreaties fruitless, fell on and slaughtered their innocent children and husband, perishing themselves in the act of carnage. It is recorded that during that triennial agitation no less than twenty-three Circles perished in domestic discord. Great indeed was the peril. It seemed as though the Priests had no choice between submission and extermination; when suddenly the course of events was completely changed by one of those picturesque incidents which Statesmen ought never to neglect, often to anticipate, and sometimes perhaps to originate, because of the absurdly disproportionate power with which they appeal to the sympathies of the populace. It happened that an Isosceles of a low type, with a brain little if at all above four degrees—accidentally dabbling in the colours of some Tradesman whose shop he had plundered—painted himself, or caused himself to be painted (for the story varies) with the twelve colours of a Dodecagon. Going into the Market Place he accosted in a feigned voice a maiden, the orphan daughter of a noble Polygon, whose affection in former days he had sought in vain; and by a series of deceptions—aided, on the one side, by a string of lucky accidents too long to relate, and on the other, by an almost inconceivable fatuity and neglect of ordinary precautions on the part of the relations of the bride—he succeeded in consummating the marriage. The unhappy girl committed suicide on discovering the fraud to which she had been subjected. When the news of this catastrophe spread from State to State the minds of the Women were violently agitated. Sympathy with the miserable victim and anticipations of similar deceptions for themselves, their sisters, and their daughters, made them now regard the Colour Bill in an entirely new aspect. Not a few openly avowed themselves converted to antagonism; the rest needed only a slight stimulus to make a similar avowal. Seizing this favourable opportunity, the Circles hastily convened an extraordinary Assembly of the States; and besides the usual guard of Convicts, they secured the attendance of a large number of reactionary Women. Amidst an unprecedented concourse, the Chief Circle of those days—by name Pantocyclus—arose to find himself hissed and hooted by a hundred and twenty thousand Isosceles. But he secured silence by declaring that henceforth the Circles would enter on a policy of Concession; yielding to the wishes of the majority, they would accept the Colour Bill. The uproar being at once converted to applause, he invited Chromatistes, the leader of the Sedition, into the centre of the hall, to receive in the name of his followers the submission of the Hierarchy. Then followed a speech, a masterpiece of rhetoric, which occupied nearly a day in the delivery, and to which no summary can do justice. With a grave appearance of impartiality he declared that as they were now finally committing themselves to Reform or Innovation, it was desirable that they should take one last view of the perimeter of the whole subject, its defects as well as its advantages. Gradually introducing the mention of the dangers to the Tradesmen, the Professional Classes and the Gentlemen, he silenced the rising murmurs of the Isosceles by reminding them that, in spite of all these defects, he was willing to accept the Bill if it was approved by the majority. But it was manifest that all, except the Isosceles, were moved by his words and were either neutral or averse to the Bill. Turning now to the Workmen he asserted that their interests must not be neglected, and that, if they intended to accept the Colour Bill, they ought at least to do so with full view of the consequences. Many of them, he said, were on the point of being admitted to the class of the Regular Triangles; others anticipated for their children a distinction they could not hope for themselves. That honourable ambition would now have to be sacrificed. With the universal adoption of Colour, all distinctions would cease; Regularity would be confused with Irregularity; development would give place to retrogression; the Workman would in a few generations be degraded to the level of the Military, or even the Convict Class; political power would be in the hands of the greatest number, that is to say the Criminal Classes, who were already more numerous than the Workmen, and would soon out-number all the other Classes put together when the usual Compensative Laws of Nature were violated. A subdued murmur of assent ran through the ranks of the Artisans, and Chromatistes, in alarm, attempted to step forward and address them. But he found himself encompassed with guards and forced to remain silent while the Chief Circle in a few impassioned words made a final appeal to the Women, exclaiming that, if the Colour Bill passed, no marriage would henceforth be safe, no woman's honour secure; fraud, deception, hypocrisy would pervade every household; domestic bliss would share the fate of the Constitution and pass to speedy perdition. "Sooner than this," he cried, "Come death." At these words, which were the preconcerted signal for action, the Isosceles Convicts fell on and transfixed the wretched Chromatistes; the Regular Classes, opening their ranks, made way for a band of Women who, under direction of the Circles, moved, back foremost, invisibly and unerringly upon the unconscious soldiers; the Artisans, imitating the example of their betters, also opened their ranks. Meantime bands of Convicts occupied every entrance with an impenetrable phalanx. The battle, or rather carnage, was of short duration. Under the skillful generalship of the Circles almost every Woman's charge was fatal and very many extracted their sting uninjured, ready for a second slaughter. But no second blow was needed; the rabble of the Isosceles did the rest of the business for themselves. Surprised, leader-less, attacked in front by invisible foes, and finding egress cut off by the Convicts behind them, they at once—after their manner—lost all presence of mind, and raised the cry of "treachery". This sealed their fate. Every Isosceles now saw and felt a foe in every other. In half an hour not one of that vast multitude was living; and the fragments of seven score thousand of the Criminal Class slain by one another's angles attested the triumph of Order. The Circles delayed not to push their victory to the uttermost. The Working Men they spared but decimated. The Militia of the Equilaterals was at once called out; and every Triangle suspected of Irregularity on reasonable grounds, was destroyed by Court Martial, without the formality of exact measurement by the Social Board. The homes of the Military and Artisan classes were inspected in a course of visitations extending through upwards of a year; and during that period every town, village, and hamlet was systematically purged of that excess of the lower orders which had been brought about by the neglect to pay the tribute of Criminals to the Schools and University, and by the violation of the other natural Laws of the Constitution of Flatland. Thus the balance of classes was again restored. Needless to say that henceforth the use of Colour was abolished, and its possession prohibited. Even the utterance of any word denoting Colour, except by the Circles or by qualified scientific teachers, was punished by a severe penalty. Only at our University in some of the very highest and most esoteric classes—which I myself have never been privileged to attend—it is understood that the sparing use of Colour is still sanctioned for the purpose of illustrating some of the deeper problems of mathematics. But of this I can only speak from hearsay. Elsewhere in Flatland, Colour is now non-existent. The art of making it is known to only one living person, the Chief Circle for the time being; and by him it is handed down on his death-bed to none but his Successor. One manufactory alone produces it; and, lest the secret should be betrayed, the Workmen are annually consumed, and fresh ones introduced. So great is the terror with which even now our Aristocracy looks back to the far-distant days of the agitation for the Universal Colour Bill. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Edwin Abbott Abbott (1995). Flatland: A Romance of Many Dimensions. Urbana, Illinois: Project Gutenberg. Retrieved https://www.gutenberg.org/cache/epub/201/pg201-images.html This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at , located at . www.gutenberg.org https://www.gutenberg.org/policy/license.html

Of the Suppression of the Chromatic Sedition

Hello, readers; I'm thrilled to have you join me as we explore an exciting trend in the industry known as Retrieval Augmented Generation (RAG). This cutting-edge technique is transforming how we handle customer queries, offering superior accuracy compared to traditional chatbots. With RAG, our chatbots can tap into recent data, providing more relevant responses—a capability many pre-trained Language Models (LLMs) lack. An Example to start with Let's dive into a common scenario: picture yourself managing an online retail giant like Walmart or Amazon. Your customers frequently inquire about products and other stuff. However, given the ever-changing nature of your inventory, constantly re-training a chatbot becomes impractical over time. That's where RAG heads up. By leveraging your product catalog, RAG swiftly retrieves relevant information from your data store or the knowledge base. It then generates a response with this information. This approach not only reduces training costs but also ensures seamless adaptation to dynamic data in modern businesses. Large Language Models (LLM’s) Before delving into the central idea of RAG, let's first examine a more familiar and straightforward concept: Large Language Model (LLM). Natural Language Processing (NLP) specialized models, such as LLMs are excelling at generating text by leveraging the extensive data they've been exposed to during training. These models, including ChatGPT, utilize transformer architectures to grasp the context and meaning of the input data. A famous example of a chatbot using an LLM model is the ChatGPT3.5 and GPT-4 models. In terms of LLMs, we encounter two main issues. One is the scarcity of knowledge about a specific domain that was not presented during training of LLM. The second is the generation of inaccurate and unreliable responses, often referred to as “hallucinations” resulting from insufficient exposure to recent data. To tackle these problem, an alternative approach is to fine-tune the LLM on domain-specific information. However, we will now explore a more straightforward solution involving Retrieval Augmented Generation (RAG). Retrieval Augmented Generation (RAG) Retrieval Augmented Generation (RAG) is an NLP technique that combines retrieval and generative AI models. It leverages the strengths of both retrieval models, which are good at retrieving accurate information from various data sources, and generative models, which excel at producing unique and human-like responses. By combining these capabilities, RAG aims to provide accurate and human-like responses while overcoming the limitations of each model type.  In a RAG-based AI system, a retrieval model is used to find relevant information from existing data sources while the generative model takes the retrieved information, synthesizes all the data, and shapes it into a coherent and contextually appropriate response. Components of a RAG system The primary components of the RAG include: Indexing documents: In this step, data from various sources are collected, preprocessed, and broken into chunks before being indexed into a vector database. We have previously discussed vector databases in a previous blog.
Retrieval: Once the documents are indexed, during the inference time, accurate documents are retrieved from the vector database using the similarity score between the input query and document embeddings.
Generation: After retrieving the accurate data, it is passed to a Generative AI to generate a unique and human-like response. The RAG process utilises the advantages of the other two models to deliver a more accurate answer. These valuable resources can help construct a Retrieval Augmented Generation model. Vector databases like LangChain FAISS, ChromaDB, Weaviate DB, Pinecone, Qdrant DB, and Milvus; and generative models such as GPT, Llama2, HuggingFace, an open-source platform with numerous pre-trained models for NLP, computer vision, and more. How RAG is being used today? Retrieval Augmented Generation can enhance business customer support by creating advanced chatbots that deliver swift and precise responses, resulting in increased customer satisfaction. Furthermore, this technology can generate informative knowledge base articles and help documents for your business by combining generative capabilities with domain-specific data retrieval. Resources Here are some other resources you may find helpful: IBM research blog on RAG
Medium blog with RAG implementation
Nvidia’s blog on RAG In conclusion, Retrieval Augmented Generation (RAG) marks a significant advancement in natural language processing by integrating retrieval and generative AI models. It effectively addresses the limitations of traditional chatbots and large language models, providing accurate and contextually relevant responses. Through its capability to index, retrieve, and generate information, RAG greatly enhances customer support and knowledge systems. This technology offers businesses a valuable tool for improving interactions, promising enhanced customer experiences in the digital realm. Its adaptability to dynamic data and specific queries positions RAG as a promising asset for driving business success and customer satisfaction through efficient dialogue systems. Wishing you an enjoyable and fruitful learning journey! Hello, readers; I'm thrilled to have you join me as we explore an exciting trend in the industry known as Retrieval Augmented Generation (RAG). This cutting-edge technique is transforming how we handle customer queries, offering superior accuracy compared to traditional chatbots. With RAG, our chatbots can tap into recent data, providing more relevant responses—a capability many pre-trained Language Models (LLMs) lack. An Example to start with Let's dive into a common scenario: picture yourself managing an online retail giant like Walmart or Amazon. Your customers frequently inquire about products and other stuff. However, given the ever-changing nature of your inventory, constantly re-training a chatbot becomes impractical over time. That's where RAG heads up. By leveraging your product catalog, RAG swiftly retrieves relevant information from your data store or the knowledge base. It then generates a response with this information. This approach not only reduces training costs but also ensures seamless adaptation to dynamic data in modern businesses. Let's dive into a common scenario: Large Language Models (LLM’s) Before delving into the central idea of RAG, let's first examine a more familiar and straightforward concept: Large Language Model (LLM). Natural Language Processing (NLP) specialized models, such as LLMs are excelling at generating text by leveraging the extensive data they've been exposed to during training. These models, including ChatGPT, utilize transformer architectures to grasp the context and meaning of the input data. A famous example of a chatbot using an LLM model is the ChatGPT3.5 and GPT-4 models. In terms of LLMs, we encounter two main issues. One is the scarcity of knowledge about a specific domain that was not presented during training of LLM. The second is the generation of inaccurate and unreliable responses, often referred to as “ hallucinations” resulting from insufficient exposure to recent data. To tackle these problem, an alternative approach is to fine-tune the LLM on domain-specific information. However, we will now explore a more straightforward solution involving Retrieval Augmented Generation (RAG). hallucinations” Retrieval Augmented Generation (RAG) Retrieval Augmented Generation (RAG) is an NLP technique that combines retrieval and generative AI models. It leverages the strengths of both retrieval models, which are good at retrieving accurate information from various data sources, and generative models, which excel at producing unique and human-like responses. By combining these capabilities, RAG aims to provide accurate and human-like responses while overcoming the limitations of each model type.  In a RAG-based AI system, a retrieval model is used to find relevant information from existing data sources while the generative model takes the retrieved information, synthesizes all the data, and shapes it into a coherent and contextually appropriate response. Components of a RAG system The primary components of the RAG include: Indexing documents: In this step, data from various sources are collected, preprocessed, and broken into chunks before being indexed into a vector database. We have previously discussed vector databases in a previous blog. Retrieval: Once the documents are indexed, during the inference time, accurate documents are retrieved from the vector database using the similarity score between the input query and document embeddings. Generation: After retrieving the accurate data, it is passed to a Generative AI to generate a unique and human-like response. The RAG process utilises the advantages of the other two models to deliver a more accurate answer. Indexing documents: In this step, data from various sources are collected, preprocessed, and broken into chunks before being indexed into a vector database. We have previously discussed vector databases in a previous blog . Indexing documents: previous blog Retrieval: Once the documents are indexed, during the inference time, accurate documents are retrieved from the vector database using the similarity score between the input query and document embeddings. Retrieval: Generation: After retrieving the accurate data, it is passed to a Generative AI to generate a unique and human-like response. The RAG process utilises the advantages of the other two models to deliver a more accurate answer. Generation: These valuable resources can help construct a Retrieval Augmented Generation model. Vector databases like LangChain FAISS, ChromaDB, Weaviate DB, Pinecone, Qdrant DB, and Milvus; and generative models such as GPT, Llama2, HuggingFace, an open-source platform with numerous pre-trained models for NLP, computer vision, and more. These valuable resources can help construct a Retrieval Augmented Generation model. Vector databases like LangChain FAISS, ChromaDB, Weaviate DB, Pinecone, Qdrant DB , and Milvus ; and generative models such as GPT, Llama2, HuggingFace, an open-source platform with numerous pre-trained models for NLP, computer vision , and more. LangChain FAISS, ChromaDB, Weaviate DB, Pinecone, Qdrant DB Milvus GPT, Llama2, HuggingFace, an open-source platform with numerous pre-trained models for NLP, computer vision How RAG is being used today? Retrieval Augmented Generation can enhance business customer support by creating advanced chatbots that deliver swift and precise responses, resulting in increased customer satisfaction. Furthermore, this technology can generate informative knowledge base articles and help documents for your business by combining generative capabilities with domain-specific data retrieval. Resources Here are some other resources you may find helpful: IBM research blog on RAG Medium blog with RAG implementation Nvidia’s blog on RAG IBM research blog on RAG IBM research blog on RAG Medium blog with RAG implementation Medium blog with RAG implementation Nvidia’s blog on RAG Nvidia’s blog on RAG In conclusion, Retrieval Augmented Generation (RAG) marks a significant advancement in natural language processing by integrating retrieval and generative AI models. It effectively addresses the limitations of traditional chatbots and large language models, providing accurate and contextually relevant responses. Through its capability to index, retrieve, and generate information, RAG greatly enhances customer support and knowledge systems. This technology offers businesses a valuable tool for improving interactions, promising enhanced customer experiences in the digital realm. Its adaptability to dynamic data and specific queries positions RAG as a promising asset for driving business success and customer satisfaction through efficient dialogue systems. Wishing you an enjoyable and fruitful learning journey!

Revolutionising Chatbots: The Rise of Retrieval Augmented Generation (RAG)

**SAN FRANSCISCO, United States, April 4th, 2024/GamingWire/--**Cloudborn, the latest release from Antler Interactive, successfully demoed at the Game Developers Conference in San Francisco last week. Featured in both the Inworld and BGA booths, the first-of-its-kind Web3, RPG, MMO, AI, PC-based game proved to be more than just a lengthy combination of acronyms. Constantin Berthelier, Narrative Designer at Antler Interactive, led the demonstrations and game tests for attendees at this year’s show. “We had testers of all sorts, some big names from big studios and publishers. Others were devs mostly curious about our title,” he explained. Attendees appeared most curious about Cloudborn’s use of AI-backed Non-Playable Characters (NPCs), something rival titles are racing to implement. With the first mover’s advantage, the team from Antler Interactive was able to drum up significant interest in their AI-NPC character, Zero. Constantin recounted, “I had anticipated that most GDC attendees were privy to the recent updates on the AI scene and would be somewhat blasé when interacting with our characters. But no, it did provide the wow effect! Most users were overwhelmed that they had the liberty to tell Zero all they wanted,” he said, then adding “Zero remained relevant and on character no matter what the testers would tell him.” Cloudborn is a Web3 gaming title, meaning various game elements utilize blockchain technology to give more control and self-ownership to gamers. This is presumed to open doors to many opportunities in a rapidly evolving and competitive technology space. Constantin shared that “The executives of big publishers, like NetEase, Square, Enix, and Amazon, all thought this [the web3 aspect] would be interesting, and even more so when I explained what it could allow.” Cloudborn encompasses a unique mix of AI and web3 technologies, aiming to improve the gaming experience. Using advanced, gameplay-focused AI - developed by their partners, Inworld - Cloudborn introduces a world that evolves as players make their moves, making environments and characters change in real-time, leading to a more immersive and personal game. Oliwer Svensson from Antler Interactive notes, “This goes beyond simple data; it's about the unpredictable ways AI shapes your character.” The use of blockchain means players truly own their characters, items, and achievements. Those testing the new title at the Game Developers Conference described it as “beautiful”, “unique”, and “fresh”, while also remarking on the “well-made UI”. One attendee is quoted as saying that Cloudborn has designed “the best UI possible to contain the AI prompts and feedback”. Attending the conference has proven to be a catalyst for Antler Interactive’s developmental charge, as they look to make Cloudborn a market leader. During the event, they opened early partnership discussions with several game producers, developers, artists, narrative designers, technologists, Web3 experts, and more, taking away considerable connections to support their growth. What was most interesting for the Cloudborn team was that while there was a noticeably big web3 presence at GDC there were no direct rivals to their title at the event. “There was one other Web3 game, but offering a very different concept. AI companies were present too, but they weren’t exploring narrative in the same way we are, in creating dynamic NPC personalities. One gaming project was looking at narrative development and AI but on a non-live model”. Cloudborn left the event feeling satisfied by the feedback and the knowledge that they have a unique proposition of a Web3 game that pushes ownership, experience, AI NPCs, and advanced trading. About Anter Interactive Antler Interactive is a Swedish gaming studio renowned for its innovative approach to gaming. Focusing on cutting-edge technologies, Antler Interactive strives to create immersive and involving experiences that captivate players worldwide. Antler, a multi-award-winning gaming studio, won the Binance Launchpool project of the year for My Neighbor Alice. Antler Interactive is a subsidiary of ChromaWay. More information can be found at antlerinteractive.com and cloudborn.game. Contact PR Manager Ramsey Shallal GELO ramsey@gelocubed.com This story was distributed as a release by Gamingwire under HackerNoon’s Business Blogging Program. Learn more about the program here. **SAN FRANSCISCO, United States, April 4th, 2024/GamingWire/--**Cloudborn, the latest release from Antler Interactive, successfully demoed at the Game Developers Conference in San Francisco last week. Featured in both the Inworld and BGA booths, the first-of-its-kind Web3, RPG, MMO, AI, PC-based game proved to be more than just a lengthy combination of acronyms. Constantin Berthelier, Narrative Designer at Antler Interactive, led the demonstrations and game tests for attendees at this year’s show. “We had testers of all sorts, some big names from big studios and publishers. Others were devs mostly curious about our title,” he explained. “We had testers of all sorts, some big names from big studios and publishers. Others were devs mostly curious about our title,” he explained. Attendees appeared most curious about Cloudborn’s use of AI-backed Non-Playable Characters (NPCs), something rival titles are racing to implement. With the first mover’s advantage, the team from Antler Interactive was able to drum up significant interest in their AI-NPC character, Zero. Constantin recounted, “I had anticipated that most GDC attendees were privy to the recent updates on the AI scene and would be somewhat blasé when interacting with our characters. But no, it did provide the wow effect! Most users were overwhelmed that they had the liberty to tell Zero all they wanted,” he said, then adding “Zero remained relevant and on character no matter what the testers would tell him.” Constantin recounted, “I had anticipated that most GDC attendees were privy to the recent updates on the AI scene and would be somewhat blasé when interacting with our characters. But no, it did provide the wow effect! Most users were overwhelmed that they had the liberty to tell Zero all they wanted,” he said, then adding “Zero remained relevant and on character no matter what the testers would tell him.” Cloudborn is a Web3 gaming title, meaning various game elements utilize blockchain technology to give more control and self-ownership to gamers. This is presumed to open doors to many opportunities in a rapidly evolving and competitive technology space. Constantin shared that “The executives of big publishers, like NetEase, Square, Enix, and Amazon, all thought this [the web3 aspect] would be interesting, and even more so when I explained what it could allow.” Cloudborn encompasses a unique mix of AI and web3 technologies, aiming to improve the gaming experience. Using advanced, gameplay-focused AI - developed by their partners, Inworld - Cloudborn introduces a world that evolves as players make their moves, making environments and characters change in real-time, leading to a more immersive and personal game. Oliwer Svensson from Antler Interactive notes, “This goes beyond simple data; it's about the unpredictable ways AI shapes your character.” The use of blockchain means players truly own their characters, items, and achievements. Inworld Inworld Those testing the new title at the Game Developers Conference described it as “beautiful”, “unique”, and “fresh”, while also remarking on the “well-made UI”. One attendee is quoted as saying that Cloudborn has designed “the best UI possible to contain the AI prompts and feedback”. Attending the conference has proven to be a catalyst for Antler Interactive’s developmental charge, as they look to make Cloudborn a market leader. During the event, they opened early partnership discussions with several game producers, developers, artists, narrative designers, technologists, Web3 experts, and more, taking away considerable connections to support their growth. What was most interesting for the Cloudborn team was that while there was a noticeably big web3 presence at GDC there were no direct rivals to their title at the event. “There was one other Web3 game, but offering a very different concept. AI companies were present too, but they weren’t exploring narrative in the same way we are, in creating dynamic NPC personalities. One gaming project was looking at narrative development and AI but on a non-live model”. Cloudborn left the event feeling satisfied by the feedback and the knowledge that they have a unique proposition of a Web3 game that pushes ownership, experience, AI NPCs, and advanced trading. About Anter Interactive Antler Interactive is a Swedish gaming studio renowned for its innovative approach to gaming. Focusing on cutting-edge technologies, Antler Interactive strives to create immersive and involving experiences that captivate players worldwide. Antler, a multi-award-winning gaming studio, won the Binance Launchpool project of the year for My Neighbor Alice. Antler Interactive is a subsidiary of ChromaWay. More information can be found at antlerinteractive.com and cloudborn.game. Contact PR Manager Ramsey Shallal GELO ramsey@gelocubed.com This story was distributed as a release by Gamingwire under HackerNoon’s Business Blogging Program. Learn more about the program here. This story was distributed as a release by Gamingwire under HackerNoon’s Business Blogging Program. Learn more about the program here . here here

Cloudborn Demo Takes GDC By Storm With Many Wowed By Gameplay

Hey everyone! and just like you, I’ve been fascinated with the recent progress of artificial intelligence. Realizing that I needed to stay abreast with all the developments happening, I decided to embark on a personal journey of learning, thus was born! With this series, I will be learning about LLMs and share ideas, experiments, opinions, trends & learnings through my blog posts. You can follow along the journey on HackerNoon or my personal website . In today’s article, we’ll be looking at how to develop a RAG using Microsoft’s Semantic Kernel. , I’m Nataraj 100 days of AI here here Retrieval augmented generation (RAG) is one of the most common application that is being developed using different LLMs. We have previously explored how to develop a RAG using . In this post we will create a RAG using . langchain Microsoft’s Semantic Kernel To follow along, you will need the Open AI API. Step 1: Initialize Semantic Kernel The first step is to initialize semantic kernel and tell the kernel that we want to use Open AI’s chat completion and Open AI’s embedding model which we will later use to create embeddings. We will also tell the kernel that we want to use a memory store which will be Chroma DB in our case. Note that we are also instructing the Kernel that this memory store needs to be persistent. kernel = sk.Kernel()

kernel.add_text_completion_service("openai", OpenAIChatCompletion("gpt-4",api_key))

kernel.add_text_embedding_generation_service("openai-embedding", OpenAITextEmbedding("text-embedding-ada-002", api_key))

# chrome db
kernel.register_memory_store(memory_store=ChromaMemoryStore(persist_directory='mymemories2'))

print("Made two new services attached to the kernel and made a Chroma memory store that's persistent.") Step 2 – Create Embeddings In this example, we are creating a RAG that can answer questions on a SWOT analysis that we created for a pizza business. So in order to do that we take the SWOT analysis and get the embeddings for them and store the corresponding embeddings in a collection called “SWOT” in the persistent datastore that we created in the last step. strength_questions = ["What unique recipes or ingredients does the pizza shop use?","What are the skills and experience of the staff?","Does the pizza shop have a strong reputation in the local area?","Are there any unique features of the shop or its location that attract customers?", "Does the pizza shop have a strong reputation in the local area?", "Are there any unique features of the shop or its location that attract customers?"]
weakness_questions = ["What are the operational challenges of the pizza shop? (e.g., slow service, high staff turnover)","Are there financial constraints that limit growth or improvements?","Are there any gaps in the product offering?","Are there customer complaints or negative reviews that need to be addressed?"]
opportunities_questions = ["Is there potential for new products or services (e.g., catering, delivery)?","Are there under-served customer segments or market areas?","Can new technologies or systems enhance the business operations?","Are there partnerships or local events that can be leveraged for marketing?"]
threats_questions = ["Who are the major competitors and what are they offering?","Are there potential negative impacts due to changes in the local area (e.g., construction, closure of nearby businesses)?","Are there economic or industry trends that could impact the business negatively (e.g., increased ingredient costs)?","Is there any risk due to changes in regulations or legislation (e.g., health and safety, employment)?"]

strengths = [ "Unique garlic pizza recipe that wins top awards","Owner trained in Sicily at some of the best pizzerias","Strong local reputation","Prime location on university campus" ]
weaknesses = [ "High staff turnover","Floods in the area damaged the seating areas that are in need of repair","Absence of popular calzones from menu","Negative reviews from younger demographic for lack of hip ingredients" ]
opportunities = [ "Untapped catering potential","Growing local tech startup community","Unexplored online presence and order capabilities","Upcoming annual food fair" ]
threats = [ "Competition from cheaper pizza businesses nearby","There's nearby street construction that will impact foot traffic","Rising cost of cheese will increase the cost of pizzas","No immediate local regulatory changes but it's election season" ]

print("✅ SWOT analysis for the pizza shop is resident in native memory")

memoryCollectionName = "SWOT"

# lets put these in memory / vector store
async def run_storeinmemory_async():
    for i in range(len(strengths)):
        await kernel.memory.save_information_async(memoryCollectionName, id=f"strength-{i}", text=f"Internal business strength (S in SWOT) that makes customers happy and satisfied Q&A: Q: {strength_questions[i]} A: {strengths[i]}")
    for i in range(len(weaknesses)):
        await kernel.memory.save_information_async(memoryCollectionName, id=f"weakness-{i}", text=f"Internal business weakness (W in SWOT) that makes customers unhappy and dissatisfied Q&A: Q: {weakness_questions[i]} A: {weaknesses[i]}")
    for i in range(len(opportunities)):
        await kernel.memory.save_information_async(memoryCollectionName, id=f"opportunity-{i}", text=f"External opportunity (O in SWOT) for the business to gain entirely new customers Q&A: Q: {opportunities_questions[i]} A: {opportunities[i]}")
    for i in range(len(threats)):
        await kernel.memory.save_information_async(memoryCollectionName, id=f"threat-{i}", text=f"External threat (T in SWOT) to the business that impacts its survival Q&A: Q: {threats_questions[i]} A: {threats[i]}")

asyncio.run(run_storeinmemory_async())

print("😶‍🌫️ Embeddings for SWOT have been generated and stored in vector db") Step 3 – Ask your question Now that we have the embeddings of our data stored in the chrome vector store, we can ask questions related to the pizza business and get an answer. #ask questions on swot
potential_question = "What are the easiest ways to make more money?"
counter = 0

async def run_askquestions_async():
    memories = await kernel.memory.search_async(memoryCollectionName, potential_question, limit=5, min_relevance_score=0.5)
    display(f"### ❓ Potential question: {potential_question}")
    for memory in memories:
        if counter == 0:
            related_memory = memory.text
        counter += 1
        print(f"  > 🧲 Similarity result {counter}:\n  >> ID: {memory.id}\n  Text: {memory.text}  Relevance: {memory.relevance}\n")

asyncio.run(run_askquestions_async()) This is a grossly simplified version of how a RAG can be created using Semantic Kernel. The most popular choice of framework to build right now using LLMs is langchain and we have previously seen how to build a RAG using langchain. Although Langchain is more popular as we see more and more companies building tools, there will be more sophisticated tools out there and I have found that Semantic Kernel has a few special features that make it stand out. That’s it for Day 18 of 100 Days of AI. I write a newsletter called Above Average where I talk about the second order insights behind everything that is happening in big tech. If you are in tech and don’t want to be average, . subscribe to it Follow me on , or for latest updates on 100 days of AI or . If you are in tech you might be interested in joining my community of tech professionals . Twitter LinkedIn HackerNoon bookmark this page here

100 Days of AI, Day 18: Developing a RAG using Microsoft Semantic Kernel

AI এর 100 দিন, 18 দিন: মাইক্রোসফ্ট শব্দার্থিক কার্নেল ব্যবহার করে একটি RAG তৈরি করা

100 Tage KI, Tag 18: Entwicklung eines RAG mit Microsoft Semantic Kernel

100 días de IA, día 18: Desarrollo de un RAG utilizando el kernel semántico de Microsoft

100 jours d'IA, jour 18 : développement d'un RAG à l'aide du noyau sémantique Microsoft

AI के 100 दिन, दिन 18: Microsoft Semantic Kernel का उपयोग करके RAG विकसित करना

100日間のAI、18日目: Microsoft Semantic Kernelを使用したRAGの開発

100일간의 AI, 18일차: Microsoft Semantic Kernel을 사용하여 RAG 개발

100 dias de IA, dia 18: desenvolvendo um RAG usando o kernel semântico da Microsoft

100 дней искусственного интеллекта, день 18. Разработка RAG с использованием семантического ядра Microsoft

100 Gün Yapay Zeka, 18. Gün: Microsoft Semantic Kernel kullanarak bir RAG geliştirme

100 ngày AI, Ngày 18: Phát triển RAG bằng Microsoft Semantic Kernel

100 天 AI，第 18 天：使用 Microsoft Semantic Kernel 开发 RAG

STOCKHOLM, Sweden, March 14th, 2024/Chainwire/-- Antler Interactive is bringing their latest creation, Cloudborn - A Web3 PC RPG with a sprinkle of MMO - into the limelight at GDC 2024. The turn-based battler set in a fantasy sky island realm is poised to take immersive gaming experiences to new heights by intertwining cutting-edge technology to create enchanting gameplay. The game births a seamless blend of strategic gameplay similar to Raid Shadow Legends and the accessibility of the Zelda franchise to create a turn-based RPG like no other. But that is where similarities end. Cloudborn is at the forefront of dynamic gameplay experiences, melding web3 technology and AI to create responsive worlds and gameplay, ultimately owned by the players. With tradeable NFT characters that develop in attributes and personality from the direct actions of the player, Cloudborn is taking immersive and player-owned experiences to the next level. Antler Interactive, the creators of My Neighbor Alice, describes the game as one where players strategize in turn-based battles and players roam an ever-evolving world shaped by their actions. The beauty of the game is in how it utilizes the next-generation tech stack to create dynamic worlds and unique characters, whose attributes and personalities change according to players’ direct actions, creating a truly unrivaled gaming experience. Which then are ultimately owned by the player, giving them full autonomy over their experiences. users can check out the and be impressed for themselves. Cloudborn game trailer Cloudborn’s main features: Dynamic World Building: The Cloudborn verse adapts and changes at the whim of the player, as every action has true consequences on the game's environment, all possible through advanced AI integrations Evolving Characters: playable and non-playable characters evolve based on players’ interactions. These changes stick with the characters even after trading with other players via NFT metadata. Characters you inherit from others will remember their past, for better or worse. Any bonds or rivalries get passed on even after trading. True Ownership: harnessing web3 technology, Cloudborn is giving players true ownership and autonomy over their experiences Strategic Turn-Based Battles: Plan your battles with your army of character cards. Your character lineup and positioning matter on the battlefield. Free Travel & Airship Base Management: The world of Miyunu is yours to explore. Take your customizable airship and explore the floating islands, upgrading your base as you go. Cloudborn will be accessible through the ownership of their in-game digital assets, which can be bought or leased, and will run on the Chromia blockchain. Taking place on fantasy islands in the clouds, Cloudborn is a story of old. According to legend, over a millennia ago a universe-shattering war between 2 godlike powers, leaving behind a fragmented world of sky-islands corrupted the remnants of the war. Players are left to traverse through the skies on their mission to unravel the mystery of the conflict and explore beyond the known.The Cloudborn verse is powered by a unique blend of AI and web3 technology, that has the sole purpose of elevating the gaming experience. Through AI, Cloudborn boasts a dynamic world-building experience, by which the gaming environments and characters react and develop in real-time based on player actions, giving rise to a deeper, unique gaming experience. As Oliwer Svensson, Chief Creative Director of Antler Interactive, says “It's not just binary data, but rather it's unpredictable characteristics that are created by AI that molds your character”. Gameplay is further enhanced by blockchain technology, as characters, items, and progression are in full control of the player. What makes Cloudborn truly special is how it elegantly intertwines the 2 technologies in a harmonious symphony by which gameplay is not only personalized and immersive but carries intent. Lenny Pettersson, CEO of Antler Interactive puts it beautifully, as he believes in using this new technology “as a paintbrush to create great and unique things. ” Whether you are picking up Cloudborn for a casual play-through, or to get into the nitty-gritty of strategic gameplay and theory crafting, you are sure to be greeted with a gaming experience like no other - quite literally. Cloudborn will be demoed next week at GDC 2024, from the 20th to the 22nd of March at the Inworld Booth (2-6 pm, March 20th), and the BGA Booths (all day, March 22nd). Players can access more information on the game through their , , and channels. Twitter Telegram Discord About Anter Interactive Antler Interactive is a Swedish gaming studio renowned for its innovative approach to gaming. Focusing on cutting-edge technologies, Antler Interactive strives to create immersive and involving experiences that captivate players worldwide. Antler, a multi-award-winning gaming studio, won the Binance Launchpool project of the year for My Neighbor Alice. Antler Interactive is a subsidiary of ChromaWay. More information can be found at and . antlerinteractive.com cloudborn.game Contact PR Manager Ramsey Shallal Antler Interactive ramsey@gelocubed.com This story was distributed as a release by Chainwire under HackerNoon’s Business Blogging Program. Learn more about the program here.

Antler Interactive To Showcase Their Latest Creation, Cloudborn, At GDC

CHROMA

chroma.co

-- employees

Since n.d.

COMPANY RANKING

CHROMA

EVERGREEN INDEX #3380

Representing this company? Click here to claim and customize this page!

Chroma WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

CHROMA

chroma.co

-- employees

Since n.d.

COMPANY RANKING

CHROMA

EVERGREEN INDEX #3380

RELATED COMPANIESRANK

STORIES ABOUT CHROMA

Representing this company? Click here to claim and customize this page!

Chroma WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!