Sulaiman ShamasnainThe Deep HubRAG II: Query TransformationsNaive RAG typically splits documents into chunks, embeds them, and retrieves chunks with high semantic similarity to a user question. But…Aug 28Aug 28
Sulaiman ShamasnaLoRA DemystifiedAn important paradigm of natural language processing consists of large-scale pre-training on general domain data and adaptation to…Jul 31Jul 31
Sulaiman ShamasnaT5 Model: Fine-Tuning on a Single GPU in a Docker ContainerT5 model, stands for Text-to-Text Transfer Transformer, is a special variation of transformers developed by Google. The T5 model stands out…Jul 23Jul 23
Sulaiman ShamasnaDistributed Model TrainingDistributed model training, mainly applicable for Deep Learning, combines distributed system principles with machine learning techniques to…Jul 16Jul 16
Sulaiman ShamasnaLarge Language and Vision Assistant (LLaVA) — v1.6 vs. v1.5LLaVA (or Large Language and Vision Assistant), an open-source large multi-modal model. The current version; 1.16 proposes some…Jul 12Jul 12
Sulaiman ShamasnaFalcon 7B Fine-Tuning on GPU with LoRAFine-tuning large language models, such as Falcon 7B, can be a resource-intensive and complex task. Techniques like Low-Rank Adaptation…Jul 4Jul 4
Sulaiman ShamasnaThe Hugging Face EcosystemWhat started with Transformers has quickly grown into a whole ecosystem consisting of many libraries and tools to accelerate your NLP and…Jun 20Jun 20
Sulaiman ShamasnaRAG IV: Agentic RAG with LlamaIndexRouter Query Engine, Tool Calling, Agent Reasoning Loop, Multi-Document Agent.Jun 12Jun 12
Sulaiman ShamasnaThe Realistic Picture of a Machine Learning Model LifecycleIn an average organization today, the realistic picture of a machine learning model lifecycle involves many different people with…Jun 5Jun 5
Sulaiman ShamasnaLLMOps: Automation and Orchestration of LLMs’ WorkflowsIntroduction to LLMOps, Data Pipelines, Data Warehouse, KubeFlow, Model Deployment, Monitoring, Fine-tuning …Jun 4Jun 4