whoisltd Personal Knowledge Base

website·Indexed 29 days ago

https://whoisltd.me/

A personal knowledge base and 'second brain' containing learning materials on English grammar (categorized by red, yellow, and blue levels), Natural Language Processing (NLP) including RAG, and book reviews.

Tokens
30.9K
Snippets
171
Records
244
Agent score
48%

What's inside whoisltd

  1. Overview of vLLM features

    vLLM is an open-source inference engine for Large Language Models (LLMs) designed for high-throughput production environments. Key technical features include:

    • Continuous Batching: Processes multiple requests in a single GPU forward pass to increase throughput.
    • PagedAttention: Optimizes KV cache memory management to reduce fragmentation.
    • OpenAI-compatible API: Provides a drop-in replacement endpoint for easy integration.
    • Quantization Support: Supports INT8 and INT4 to reduce VRAM requirements.
  2. Overview of the Three-Body Problem novel trilogy

    The Three-Body Problem is a science fiction trilogy written by Liu Cixin. The series consists of three interconnected novels: 'The Three-Body Problem', 'The Dark Forest', and 'Death's End'.
  3. Overview of Apache Spark

    Apache Spark is an open-source cluster computing framework designed for fast data processing. It achieves high performance by performing computations in-memory (RAM), making it significantly faster than disk-based MapReduce (up to 10x faster on disk and 100x faster in RAM). Spark supports both batch and real-time data processing (via Spark Streaming), allowing it to handle millions of events per second from sources like Twitter or Facebook. It is compatible with various data sources and file formats supported by Hadoop and supports Java, Scala, Python, and R.
  4. Overview of Common NLP Tasks

    Natural Language Processing (NLP) involves several core tasks, often starting with Language Modeling as a foundation. Key tasks include:

    • Text Classification
    • Information Extraction
    • Information Retrieval
    • Question Answering
    • Machine Translation
    • Conversational Agents

    NLP is considered challenging due to human language ambiguity, heavy dependence on context, idioms, and multilingualism.

  5. Ask about the height of objects and people

    Use different phrasing depending on whether you are asking about a person or an object. For buildings, you can specifically ask about the number of stories.

    For Objects:

    • Singular: "How tall is it?", "What's the height?", "How many stories is it?" (buildings)
    • Plural: "How tall are they?", "How high are they?"

    For People:

    • Singular: "How tall are you?", "What's your height?"
    • Plural: "How tall are they?"

    Common Response Patterns:

    • Full detail: "I'm 5 feet, 4 inches tall"
    • Shorthand: "I'm 5'11""
    • Casual/Elliptical: "He's six one" (omitting 'feet' and 'inches')
  6. Ask about the weight of objects and people

    When asking about weight, use the following structures. Note that asking about a person's weight is generally considered more sensitive than asking about height.

    For Objects:

    • Singular: "How much does this weigh?", "What does this weigh?", "How heavy is this?"
    • Plural: "How much do these weigh?", "What do these weigh?", "How heavy are these?"

    For People:

    • Singular: "How much do you weigh?", "How much does he/she weigh?", "What's your weight?"
    • Plural: "How much do they weigh?", "What's their weight?"

    Usage Constraints:

    • Avoid saying "How much pounds?" or "What weight is it?"

    Common Response Patterns:

    • "I weigh 1912 pounds"
    • "He weighs about 175 pounds"
    • "It's about 10 pounds"
  7. Enterprise AI Hub Architecture Pattern

    The Enterprise AI Hub is a design pattern for building a centralized, multi-tenant AI platform. It is designed to serve multiple business use cases and different countries from a single deployment, ensuring scalability and centralized management across an organization.
  8. Examples of Present Continuous questions and answers

    Present Continuous questions are used to ask about actions happening right now. Short answers typically repeat the auxiliary verb 'to be'.
    - Question: Are you learning English right now?
      Answer: Yes, I am.
    
    - Question: Am I teaching you English right now?
      Answer: Yes, you are.
    
    - Question: What are we doing?
      Answer: We're working together.
    
    - Question: What are you doing?
      Answer: We're putting oranges into a basket.
  9. Understand the use of Nouns in English

    Nouns are words that represent people, places, things, activities, emotions, or ideas. They are categorized into two main groups: Count Nouns and Noncount Nouns.
    - That man is an actor (People)
    - That country is beautiful (Place)
    - I use pliers to fix things (Thing)
    - Everyone needs love (Emotion)
    - I have an idea! (Idea)
  10. Express dates and days of the week in English

    Use the following patterns to describe specific dates, days, and relative timeframes within a month:

    Talking about specific days:

    • Current day: "Today is the 18th", "Today's Thursday", "It's Thursday the 18th", or "It's April 18th".
    • Relative days: "Yesterday was the 17th", "Tomorrow is the 19th", "Tomorrow is Friday", or "The day after tomorrow is Saturday".
    • Month boundaries: "There are 20 days in the month of April", "April began on a Monday", or "April ends on a Tuesday".

    Talking about weeks:

    • Identifying a week: "This is the week of the 15th" (indicating the week starts on April 15).
    • Scheduling: "I'll be in Chicago on the week of the 22nd. That's next week".
    • Relative time: "Last week was not as busy" (referring to the previous week), "exactly one week ago on April [date]", "A week from now will be April 25", or "It will be May in about two weeks".
  11. Implement the Enterprise AI Hub Architecture Pattern

    The Enterprise AI Hub is a design pattern used to build a centralized, multi-tenant AI platform. This architecture allows a single deployment to serve multiple business use cases and different geographic regions (countries) efficiently. This pattern was developed based on the design of the Motul APAC AI Hub.
  12. Form questions using 'would like'

    Questions are formed by placing 'would' before the subject to ask about preferences or desires.
    - Yes/No Question: Would you like some ketchup?
    - Open Question: What would you like on your pizza?
    - Time/Quantity: What time would you like to leave?