- 14
- 6 359
Abonia Sojasingarayar
India
เข้าร่วมเมื่อ 23 ส.ค. 2013
👋 Hi there! Welcome to our Channel!
🚀 I share my learning and insights of industry level AI solution. Dive into the exciting world of Artificial Intelligence, where I make complex concepts simple and captivating. Whether you're curious about Natural Language Processing, Computer Vision, Machine Learning, Generative AI, or MLOps, we've got you covered!
💡 With over 7 years of hands-on experience, I'm here to share practical insights and industry-level applications in a easy and engaging way.
🎓 Let's learn, innovate, and shape the future together!
🚀 I share my learning and insights of industry level AI solution. Dive into the exciting world of Artificial Intelligence, where I make complex concepts simple and captivating. Whether you're curious about Natural Language Processing, Computer Vision, Machine Learning, Generative AI, or MLOps, we've got you covered!
💡 With over 7 years of hands-on experience, I'm here to share practical insights and industry-level applications in a easy and engaging way.
🎓 Let's learn, innovate, and shape the future together!
Anylabeling - Image Annotation Tool - ObjectDetection and Instance Segmenation #Computervision #YOLO
In this tutorial we will be seeing an anotation tool , AnyLabeling.
AnyLabeling is a powerful tool that supports a wide range of annotation tasks, including object detection, segmentation,, and more. It's designed to be user-friendly and efficient, making it an excellent choice for both beginners and experienced annotators. Lets get into it.
⭐️ Contents ⭐️
00:00 Introduction
0:56 Installation and Setup
3:30 Annotation - Object Detection
7:04 Annotation - Instance Segmentation
13:58 Import Annotations (Custom YOLO model )
___________________________________________________________________________
🔔 Get our Newsletter and Featured Articles: abonia1.github.io/newsletter/
🔗 Linkedin: www.linkedin.com/in/aboniasojasingarayar/
🔗 Find me on Github : github.com/Abonia1
🔗 Medium Articles: medium.com/@abonia
AnyLabeling is a powerful tool that supports a wide range of annotation tasks, including object detection, segmentation,, and more. It's designed to be user-friendly and efficient, making it an excellent choice for both beginners and experienced annotators. Lets get into it.
⭐️ Contents ⭐️
00:00 Introduction
0:56 Installation and Setup
3:30 Annotation - Object Detection
7:04 Annotation - Instance Segmentation
13:58 Import Annotations (Custom YOLO model )
___________________________________________________________________________
🔔 Get our Newsletter and Featured Articles: abonia1.github.io/newsletter/
🔗 Linkedin: www.linkedin.com/in/aboniasojasingarayar/
🔗 Find me on Github : github.com/Abonia1
🔗 Medium Articles: medium.com/@abonia
มุมมอง: 82
วีดีโอ
Top LLM and Deep Learning Inference Engines - Curated List
มุมมอง 14721 วันที่ผ่านมา
Inference engines like DeepSpeed, FasterTransformer, and vLLM are designed to accelerate the process of generating predictions from large language models (LLMs) by optimizing the computation and memory usage during inference. These engines are particularly useful in scenarios where the models are deployed for real-time applications, requiring fast and efficient processing of large volumes of da...
Summarization with LangChain using LLM - Stuff - Map_reduce - Refine
มุมมอง 274หลายเดือนก่อน
This tutorial focus on summarization techniques in LangChain.It covers the basic usage of document summarization techniques and provides insights into various summarization methods. Additionally, to learn more and to explore how to validate intermediate results from the output of each of these techniques. We dive into three key summarization techniques: 1. stuff 2. map_reduce 3. refine Explorin...
Deploying a Retrieval-Augmented Generation (RAG) in AWS Lambda
มุมมอง 6872 หลายเดือนก่อน
Deploying a Retrieval-Augmented Generation (RAG) model in AWS Lambda using Docker and Amazon ECR, with LangChain. The tutorial covers the necessary setup, including downloading and installing the AWS CLI, Docker, and accessing the AWS Lambda and ECR interfaces. It provides a step-by-step guide to preparing your environment, creating a Dockerfile, building and pushing your Docker image to ECR, c...
Build and Deploy LLM Application in AWS Lambda - BedRock - LangChain
มุมมอง 2.3K2 หลายเดือนก่อน
Building and deploying a Large Language Model (LLM) application on AWS Lambda, leveraging the power of BedRock and LangChain. We explore how to create a serverless LLM chain application that allows users to ask natural language questions. ⭐️ Contents ⭐️ 00:00 Introduction 1:57 Bedrock and Lambda Service 4:23 Create Lambda Function 8:04 Deployment and Test 9:17 Create and Add Custom Lambda Layer...
Run Ollama with Langchain Locally - Local LLM
มุมมอง 9003 หลายเดือนก่อน
Ollama allows you to run open-source large language models, such as Llama 2, Mistral locally. -Ollama bundles model weights, configuration, and data into a single package, defined by a Modelfile. -It optimizes setup and configuration details, including GPU usage. -For a complete list of supported models and model variants, see the Ollama model library: github.com/ollama/ollama#model-library Com...
LLMLingua - Prompt Compression for LLM Use Cases 🔥
มุมมอง 2663 หลายเดือนก่อน
Large language models (LLMs) have demonstrated remarkable capabilities and have been applied across various fields. Advancements in technologies such as Chain-of-Thought (CoT), In-Context Learning (ICL), and Retrieval-Augmented Generation (RAG) have led to increasingly lengthy prompts for LLMs, sometimes exceeding tens of thousands of tokens. Longer prompts, however, can result in 1) increased ...
What is RAG (Retrieval-Augmented Generation)?
มุมมอง 1794 หลายเดือนก่อน
Retrieval-Augmented Generation (RAG) is the process of optimizing the output of a large language model, so it references an knowledge base outside of its training data sources before generating a response. Large Language Models (LLMs) are trained on vast volumes of data and use billions of parameters to generate original output for tasks like answering questions, translating languages, and comp...
BERTScore Explained in 5 minutes
มุมมอง 7514 หลายเดือนก่อน
📈 BERTScore - Evaluating Text Generation with BERT 📊 BERTScore leverages the pre-trained BERT model to score the semantic similarity between two pieces of text. It uses cosine similarity over the pooled output, cls token, and average of all tokens. This approach provides a more accurate representation of human judgment compared to traditional methods like ROUGE, METEOR and CIDEr. 🎯 BERTScore ca...
Must read LLM and AI Research Papers of 2023 🔥
มุมมอง 1924 หลายเดือนก่อน
🔥 LLM and AI Research Papers of 2023 🔥 1. LLaMA: Open and Efficient Foundation Language Models - lnkd.in/g5v5qBdn 2. GPT-4 Technical Report - lnkd.in/ggwpxSdT 3. PaLM 2 Technical Report - lnkd.in/g4f5ZHgd 4. Sparks of Artificial General Intelligence: Early experiments with GPT-4 -lnkd.in/gUJTSaZy 5. PaLM-E: An Embodied Multimodal Language Model - lnkd.in/gcRKm9db 6. QLoRA: Efficient Finetuning ...
Which version of BERT is it used in BERTScore ?
As we are using lang= "en" so it uses roberta-large. We can also customize it using the model_type param of BERScorer class For default model for other languages,find it here: github.com/Tiiiger/bert_score/blob/master/bert_score/utils.py
hey great content. please continue to do more videos and real time projects. Thanks
Glad it helped. Sure I am already on it.
Awesome mam , very easy to understand
Hi Abonia, thanks for sharing. I am facing this error . can you please tell how to resolve it "errorMessage": "Unable to import module 'lambda_function': No module named 'langchain_community'",
Hello, You are most welcome. You must prepare your ZIP file with all the necessary packages. You can refer to the instructions starting at the 09:04
Hi Abonia, thanks for the thorough guide, but i'm abit confused with the lambda_layer.zip file, why did you have to create it through docker? is there an easier way to provide the dependencies in a zip file without going through docker? Thanks in advance!
Hi Humayoun Khan, Yes we can but Docker facilitates the inclusion of the runtime interface client for Python, making the image compatible with AWS Lambda. Also it ensures a consistent and reproducible environment for Lambda function's dependencies. This is crucial for avoiding discrepancies between development, testing, and production environments. Hope this helps.
hello! Thanks for the video. I am from Brazil. What would you recommend for large documents, averaging 150 pages? I tried map-reduce, but the inference time was 40 minutes. Are there any tips for these very long documents?
Thanks for you kind words and glad this helped. Implement a strategy that combines semantic chunking with K-means clustering to address the model’s contextual limitations. By employing efficient clustering techniques, we can extract key passages effectively, thereby reducing the overhead associated with processing large volumes of text. This approach not only significantly lowers costs by minimizing the number of tokens processed but also mitigates the recency and primacy effects inherent in LLMs, ensuring a balanced consideration of all text segments.
Did you use OpenAI API key here?
Here we use open-source Mixtral from ollama.But, yes we can use OpenAI models as well.
Very informatics nd Your voice very clear dr
Glad it helped!
Can we use openai and chromadb on aws??
Yes we can! In the below tutorial I have demonstrated how we can create and deploy lambda layer via container for larger dependencies : th-cam.com/video/gicsb9p7uj4/w-d-xo.htmlsi=F_X7-6YCAb0Kz3Jc
@@AboniaSojasingarayar yes but can this be done without eks or containers?
Yes! You can try it by creating a custom lambda layer.If you face issue try to use only the required libraries and remove any unnecessary dependencies from your zip file.Hope this helps.
In the abstractive summarization use-case, usually a lot of focus is given to the LLMs being used and its performance. Limitations of LLM including context length and ways to overcome this issue are often overlooked. Its important to make sure that our application is scalable when dealing with large document sizes. Thank you for this great and insightful video.
Thank you Vijay Gandhi, for your insightful comment! You've raised an excellent point about the importance of considering the limitations of LLMs in the context of abstractive summarization, especially regarding their context length and scalability issues when dealing with large documents. Indeed, one of the significant challenges in using LLMs for abstractive summarization is their inherent limitation in processing long texts due to the maximum token limit imposed by these models. This constraint can be particularly problematic when summarizing lengthy documents or articles, where the full context might not fit within the model's capacity.
Really useful info mam , keep up the good work
It's my pleasure.
👍👍
Very Informative thanks for uploading
Glad it helped!
Nice video
Thanks Akshay. Glad it helped!
Nyccc Mam 😍
Glad it helped 😊
Very nice video, learnt a lot
Thank you! Glad it helped🤓
Please do more on AWS Bedrock to develop on RAG applications......your explanation is simple and effective.......stay motivated and upload more videos about LLM
Thanks for your kind words! Sure I will do it.
Yes same thing i want @@AboniaSojasingarayar
Here the tutorial link to Deploying a Retrieval-Augmented Generation (RAG) in AWS : th-cam.com/video/gicsb9p7uj4/w-d-xo.html
👍👍
😊😊
Very informative
Glad it was helpful!
Awesome , thanks
🎯𝐋𝐋𝐌 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐅𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤𝐬🎯 ✅ 𝐀𝐥𝐩𝐚 is a system for training and serving large-scale neural networks. Scaling neural networks to hundreds of billions of parameters has enabled dramatic breakthroughs such as GPT-3, but training and serving these large-scale neural networks require complicated distributed system techniques. Alpa aims to automate large-scale distributed training and serving with just a few lines of code. 📌Alpa: github.com/alpa-projects/alpa 📌Serving OPT-175B, BLOOM-176B and CodeGen-16B using Alpa: alpa.ai/tutorials/opt_serving.html ✅ 𝐃𝐞𝐞𝐩𝐒𝐩𝐞𝐞𝐝 is an easy-to-use deep learning optimization software suite that enables unprecedented scale and speed for DL Training and Inference. 📌Megatron-LM GPT2 tutorial: www.deepspeed.ai/tutorials/megatron/ 📌DeepSpeed: github.com/microsoft/DeepSpeed ✅𝐌𝐞𝐠𝐚𝐭𝐫𝐨𝐧-𝐋𝐌 / Megatron is a large, powerful transformer developed by the Applied Deep Learning Research team at NVIDIA. Below repository is for ongoing research on training large transformer language models at scale. Developing efficient, model-parallel (tensor, sequence, and pipeline), and multi-node pre-training of transformer based models such as GPT, BERT, and T5 using mixed precision. 📌pretrain_gpt3_175B.sh: github.com/NVIDIA/Megatron-LM/blob/main/examples/pretrain_gpt3_175B.sh 📌Megatron-LM: github.com/NVIDIA/Megatron-LM ✅ 𝐂𝐨𝐥𝐨𝐬𝐬𝐚𝐥-𝐀𝐈 provides a collection of parallel components for you. It aim to support us to write our distributed deep learning models just like how we write our model on our laptop. It provide user-friendly tools to kickstart distributed training and inference in a few lines. 📌Colossal-AI: colossalai.org/ 📌Open source solution replicates ChatGPT training process.Ready to go with only 1.6GB GPU memory and gives you 7.73 times faster training: www.hpc-ai.tech/blog/colossal-ai-chatgpt ✅ 𝐁𝐌𝐓𝐫𝐚𝐢𝐧 is an efficient large model training toolkit that can be used to train large models with tens of billions of parameters. It can train models in a distributed manner while keeping the code as simple as stand-alone training. 📌BMTrain: github.com/OpenBMB/BMTrain ✅ 𝐌𝐞𝐬𝐡 𝐓𝐞𝐧𝐬𝐨𝐫𝐅𝐥𝐨𝐰 (mtf) is a language for distributed deep learning, capable of specifying a broad class of distributed tensor computations. The purpose of Mesh TensorFlow is to formalize and implement distribution strategies for your computation graph over your hardware/processors. For example: "Split the batch over rows of processors and split the units in the hidden layer across columns of processors." Mesh TensorFlow is implemented as a layer over TensorFlow. 📌Mesh TensorFlow: github.com/tensorflow/mesh Please let me know in the comment section if there are any frameworks in your curated list 👇
Good dr.....❤
Thank you 🙂
TOP 3 Repo to start LLM your Learning in 2024 1. Awesome-LLM: github.com/Hannibal046/Awesome-LLM 2.Awsome-LLMOps: github.com/tensorchord/awesome-llmops 3.awsome-llm: github.com/KennethanCeyer/awesome-llm
thanks!
Glad it helped!
👍👍👍
😕👍
all the best
👍👍👍