From a225b28fc5034c9fcc3a97fd83be753a1dc4086f Mon Sep 17 00:00:00 2001 From: Deepak Date: Sat, 15 Feb 2025 04:09:28 +0530 Subject: [PATCH] updated docs --- .../meetups-talks-sessions.md | 2 +- .../projects/81-stashfin-terms.md | 2 +- docs/ai/libraries/ml-monitoring.md | 6 ++-- docs/ai/libraries/mlops-model-deployment.md | 2 +- docs/ai/llm/llm-agents.md | 30 +++++++++++++++++++ docs/book-summaries/thinking-fast-and-slow.md | 2 +- docs/cloud/others/akamai.md | 14 ++++----- .../bigquery/storage-optimizations.md | 2 +- .../nosql-databases/mongodb/vector-search.md | 3 +- docs/devops/devops-intro/high-availability.md | 2 +- docs/devops/ides/vscode-vs-code.md | 6 ++-- docs/economics/fintech-nbfc-banking-terms.md | 4 +-- .../geography/new-home-place-house.md | 2 +- .../networking-concepts/peer-to-peer.md | 2 +- 14 files changed, 56 insertions(+), 23 deletions(-) diff --git a/docs/about-deepak-sood/meetups-talks-sessions.md b/docs/about-deepak-sood/meetups-talks-sessions.md index 137a4dd0d4f..679abd43731 100644 --- a/docs/about-deepak-sood/meetups-talks-sessions.md +++ b/docs/about-deepak-sood/meetups-talks-sessions.md @@ -46,7 +46,7 @@ TFUG Ghaziabad and PyDelhi invites tech enthusiasts, innovators, and experts to - Demand forecasting using Time Series models - Running ML pipelines on Vertex AI -https://www.commudle.com/fill-form/3152 +[ML Saturday – A Celebration of AI & Community!](https://www.commudle.com/communities/tensorflow-user-group-ghaziabad/events/ml-sunday-fuel-your-weekend-with-ai) [Commudle - Connect & Learn With Software Developers](https://www.commudle.com/communities/tensorflow-user-group-ghaziabad) diff --git a/docs/about-deepak-sood/projects/81-stashfin-terms.md b/docs/about-deepak-sood/projects/81-stashfin-terms.md index ab9a1f97866..45d11fb1a5b 100644 --- a/docs/about-deepak-sood/projects/81-stashfin-terms.md +++ b/docs/about-deepak-sood/projects/81-stashfin-terms.md @@ -48,7 +48,7 @@ This fee is what we call Switching Fee in the payment ecosystem. This fee can ### Interchange Fee -An interchange fee is an amount that the issuing institutions collect from the acquiring bank. Usually, this fee is a percentage of the total transaction plus a fixed amount. And while the issuing institutions collect, assess and set this fee, they are paid to the issuing bank, who issue a particular card.  +An interchange fee is an amount that the issuing institutions collect from the acquiring bank. Usually, this fee is a percentage of the total transaction plus a fixed amount. And while the issuing institutions collect, assess and set this fee, they are paid to the issuing bank, who issue a particular card. Please note that the average interchange rate for a credit card is around 1.81% and for debit cards, it’s 0.3%. diff --git a/docs/ai/libraries/ml-monitoring.md b/docs/ai/libraries/ml-monitoring.md index 86f2e8b2c37..f60ad132b5f 100644 --- a/docs/ai/libraries/ml-monitoring.md +++ b/docs/ai/libraries/ml-monitoring.md @@ -16,7 +16,7 @@ It is not as simple as saying, "we have two additional dimensions" to consider w #### Entanglements -Any change in the input data distributions will influence the approximation of the target function, which may affect the predictions made by the model. In other words, changing anything changes everything. Therefore, any feature engineering and selection code must be carefully tested. +Any change in the input data distributions will influence the approximation of the target function, which may affect the predictions made by the model. In other words, changing anything changes everything. Therefore, any feature engineering and selection code must be carefully tested. #### Configurations @@ -69,13 +69,13 @@ At the heart of your machine learning system lies your machine learning model. F **Model drift**: Model drift is the decay of a model’s predictive power due to alterations in the real-world environment. Statistical tests should be used to detect drift, and predictive performance should be monitored to evaluate the model’s performance over time. -**Versions**: Always ensure the correct model is running in production. Version history and predictions should be tracked. +**Versions**: Always ensure the correct model is running in production. Version history and predictions should be tracked. #### The output To understand how the model performs, you must also understand the predictions the model outputs in the production environment. A machine learning model is put into production to solve a problem. Thus, monitoring the model’s output is a valuable way to ensure it performs according to the metrics used as KPIs. For example: -**Ground truth:** For some problems, you can acquire ground truth labels. For example, if a model is used to recommend personalized ads to users (you are predicting if a user will click the ad or not), and a user clicks to imply the ad is relevant, you can almost immediately acquire the ground truth. In such scenarios, an aggregation of a model’s predictions can be evaluated against the actual solution to determine how well the model performs. However, evaluating model predictions against ground truth labels is difficult in most machine learning use cases, and an alternative method is required. +**Ground truth:** For some problems, you can acquire ground truth labels. For example, if a model is used to recommend personalized ads to users (you are predicting if a user will click the ad or not), and a user clicks to imply the ad is relevant, you can almost immediately acquire the ground truth. In such scenarios, an aggregation of a model’s predictions can be evaluated against the actual solution to determine how well the model performs. However, evaluating model predictions against ground truth labels is difficult in most machine learning use cases, and an alternative method is required. **Prediction drift:** When it is not possible to acquire ground truth labels, predictions must be monitored. If there is a drastic change in the distribution of predictions, something has potentially gone wrong. For example, if you are using a model to predict fraudulent credit card transactions and suddenly the proportion of transactions identified as fraud shoots up, then something has changed. Perhaps input data structure has been altered, some other microservice in the system is misbehaving, or maybe there is just more fraud in the world. diff --git a/docs/ai/libraries/mlops-model-deployment.md b/docs/ai/libraries/mlops-model-deployment.md index d4a2fa8b130..da112b404ee 100755 --- a/docs/ai/libraries/mlops-model-deployment.md +++ b/docs/ai/libraries/mlops-model-deployment.md @@ -34,7 +34,7 @@ MLOps is an engineering discipline that aims to unify ML systems development (de - Deploy ML models - Iterate - Monitor, optimize and maintain the performance of the model -### Deploying models to the production system +### Deploying models to the production system There are mainly two ways of deploying an ML model: diff --git a/docs/ai/llm/llm-agents.md b/docs/ai/llm/llm-agents.md index d7ac8ee923e..c69d51636d3 100644 --- a/docs/ai/llm/llm-agents.md +++ b/docs/ai/llm/llm-agents.md @@ -26,10 +26,25 @@ To initialize a minimal agent, you need at least these two arguments: - [GitHub - huggingface/smolagents: 🤗 smolagents: a barebones library for agents. Agents write python code to call tools and orchestrate other agents.](https://github.com/huggingface/smolagents) - [smolagents](https://huggingface.co/docs/smolagents/en/index) +- [Introducing smolagents: simple agents that write actions in code.](https://huggingface.co/blog/smolagents) - [Build Multi-Agent Systems with SmolAgents - YouTube](https://www.youtube.com/watch?v=uzskhpH5fvo) - [Build AI Agents using HuggingFace's SmolAgents \| Agentic AI - YouTube](https://www.youtube.com/watch?v=VSm5-CX4QaM) - [Build AI Agents using HuggingFace's SmolAgents \| Agentic AI - YouTube](https://www.youtube.com/watch?v=VSm5-CX4QaM) - [Hugging Face's Smolagents: A Guide With Examples](https://www.datacamp.com/tutorial/smolagents) +- [SmolAgents: A Smol Library to Build Agents - YouTube](https://www.youtube.com/watch?v=icRKf_Mvmt8) + +## CrewAI + +Production-grade framework for orchestrating sophisticated AI agent systems. From simple automations to complex real-world applications, CrewAI provides precise control and deep customization. By fostering collaborative intelligence through flexible, production-ready architecture, CrewAI empowers agents to work together seamlessly, tackling complex business challenges with predictable, consistent results. + +### Why CrewAI? + +The power of AI collaboration has too much to offer. CrewAI is a standalone framework, built from the ground up without dependencies on Langchain or other agent frameworks. It's designed to enable AI agents to assume roles, share goals, and operate in a cohesive unit - much like a well-oiled crew. Whether you're building a smart assistant platform, an automated customer service ensemble, or a multi-agent research team, CrewAI provides the backbone for sophisticated multi-agent interactions. + +### Links + +- [GitHub - crewAIInc/crewAI: Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.](https://github.com/crewAIInc/crewAI) +- [CrewAI](https://www.crewai.com/) ## AI Agents / Tools @@ -39,6 +54,21 @@ To initialize a minimal agent, you need at least these two arguments: ![AI Agents Landscape](../../media/Pasted%20image%2020250114143214.jpg) +## VertexAI + +- [Build an agent using playbooks  \|  Dialogflow CX  \|  Google Cloud](https://cloud.google.com/dialogflow/cx/docs/quick/build-agent-playbook) +- [Playbook-based pre-built agents  \|  Dialogflow CX  \|  Google Cloud](https://cloud.google.com/dialogflow/cx/docs/concept/playbook/prebuilt) +- [GitHub - FirebaseExtended/compass-travel-planning-sample](https://github.com/FirebaseExtended/compass-travel-planning-sample) +- [Intro to AI agents - YouTube](https://www.youtube.com/watch?v=ZZ2QUCePgYw) +- [Build and deploy generative AI agents using natural language with Vertex AI Agent Builder - YouTube](https://www.youtube.com/watch?v=GCmGxBl3RLY) +- [GitHub - kkrishnan90/vertex-ai-search-agent-builder-demo](https://github.com/kkrishnan90/vertex-ai-search-agent-builder-demo) + ## References +- [Agents \| Kaggle](https://www.kaggle.com/whitepaper-agents) +- [Building effective agents \\ Anthropic](https://www.anthropic.com/research/building-effective-agents) - [Google's Blueprint to Building Powerful Agents - YouTube](https://www.youtube.com/watch?v=Z8vTgJkwyA0) +- [Get started  \|  Genkit  \|  Firebase](https://firebase.google.com/docs/genkit/get-started) +- [oscar - Git at Google](https://go.googlesource.com/oscar/) +- [LLM Agents - Explained! - YouTube](https://www.youtube.com/watch?v=5CLNoPiMbUc) +- [smolagent-tutorial.ipynb](https://colab.research.google.com/drive/1A03Qt_B0k8U-NPjcvkyJVX_Ch-9955ul) diff --git a/docs/book-summaries/thinking-fast-and-slow.md b/docs/book-summaries/thinking-fast-and-slow.md index 9f49d78c955..92dfecdc61a 100644 --- a/docs/book-summaries/thinking-fast-and-slow.md +++ b/docs/book-summaries/thinking-fast-and-slow.md @@ -87,7 +87,7 @@ We often find ourselves in situations where we have to make quick judgments. To For the most part, these processes are very useful, but the problem is that our minds often overuse them. Applying these rules in inappropriate situations can lead to mistakes. To better understand what heuristics are and the errors that follow, we can consider two types: _the substitution heuristic_ and _the availability heuristic_ . -_Alternative heuristics_ occurs when we answer an easier question than the one actually asked. +_Alternative heuristics_ occurs when we answer an easier question than the one actually asked. For example, try this question: "A woman is running for sheriff. How successful will she be in that ministry?" We automatically replace the question we should have answered with an easier one, like, "Does she look like someone who would make a good sheriff?" This experimentation means that instead of researching a candidate's profile and policies, we are simply asking ourselves the much easier question of whether this woman fits our mental image of a candidate. good sheriff or not. diff --git a/docs/cloud/others/akamai.md b/docs/cloud/others/akamai.md index b97d7527b05..c249185f09a 100644 --- a/docs/cloud/others/akamai.md +++ b/docs/cloud/others/akamai.md @@ -2,20 +2,20 @@ ### Offload -In Akamai, offload is ==the percentage of requests that are served from edge servers without needing to reach the origin server==. It's a metric that helps improve user experience and reduce costs. +In Akamai, offload is ==the percentage of requests that are served from edge servers without needing to reach the origin server==. It's a metric that helps improve user experience and reduce costs. #### How it works -- Offloading content to edge servers makes it closer to the end user, which speeds up delivery. -- Offloading content reduces the load on the origin server. -- Offloading content can minimize data egress costs. +- Offloading content to edge servers makes it closer to the end user, which speeds up delivery. +- Offloading content reduces the load on the origin server. +- Offloading content can minimize data egress costs. #### How to measure offload -- The offload metric is calculated as (edge-origin)/edge x 100. -- You can view offload numbers in the Control Center under COMMON SERVICES > Traffic reports. +- The offload metric is calculated as (edge-origin)/edge x 100. +- You can view offload numbers in the Control Center under COMMON SERVICES > Traffic reports. #### How to increase offload -- Enable caching for as much content as possible. +- Enable caching for as much content as possible. - Choose a long enough time to live (TTL) for content so that end users don't receive stale content. diff --git a/docs/databases/data-warehouses/bigquery/storage-optimizations.md b/docs/databases/data-warehouses/bigquery/storage-optimizations.md index 4b3cbd984bd..63d69a1860c 100644 --- a/docs/databases/data-warehouses/bigquery/storage-optimizations.md +++ b/docs/databases/data-warehouses/bigquery/storage-optimizations.md @@ -290,7 +290,7 @@ Auto Deleting Tables that are not frequently used in a project with table Prefix Why? - This script would be useful in environments where there are many temporary or ephemeral tables that are not needed after a certain period. -- It helps in managing and cleaning up the dataset by removing old or unused tables, potentially reducing costs and improving manageability. +- It helps in managing and cleaning up the dataset by removing old or unused tables, potentially reducing costs and improving manageability. ```python from google.cloud import bigquery diff --git a/docs/databases/nosql-databases/mongodb/vector-search.md b/docs/databases/nosql-databases/mongodb/vector-search.md index e7adf13a104..81764b6f377 100644 --- a/docs/databases/nosql-databases/mongodb/vector-search.md +++ b/docs/databases/nosql-databases/mongodb/vector-search.md @@ -51,4 +51,5 @@ Yes, Atlas Vector Search can query any kind of data that can be turned into an e [MongoDB Atlas Vector Search | MongoDB](https://www.mongodb.com/products/platform/atlas-vector-search) [google-research/scann at master · google-research/google-research · GitHub](https://github.com/google-research/google-research/tree/master/scann) -- ScaNN (Scalable Nearest Neighbors) is a method for efficient vector similarity search at scale. \ No newline at end of file + +- ScaNN (Scalable Nearest Neighbors) is a method for efficient vector similarity search at scale. diff --git a/docs/devops/devops-intro/high-availability.md b/docs/devops/devops-intro/high-availability.md index be03fed97e0..e3e0144c0f9 100644 --- a/docs/devops/devops-intro/high-availability.md +++ b/docs/devops/devops-intro/high-availability.md @@ -39,7 +39,7 @@ To achieve "4 nines" availability and beyond, we must consider: 1. System designs - designing for failure using: 1. Redundancy 2. Tradeoffs -2. System operations and maintenance - key principles are: +2. System operations and maintenance - key principles are: 1. Change management 2. Capacity management 3. Automated detection and troubleshooting diff --git a/docs/devops/ides/vscode-vs-code.md b/docs/devops/ides/vscode-vs-code.md index 8b942ff450c..f727ccd7e0f 100755 --- a/docs/devops/ides/vscode-vs-code.md +++ b/docs/devops/ides/vscode-vs-code.md @@ -15,8 +15,10 @@ - https://vscode.dev - https://github.com/conwnet/github1s - Gitpod - https://www.freecodecamp.org/news/exampro-cloud-developer-environment-certification-gitpod-course - -codeanywhere +- [Project IDX](https://idx.dev/) - Google + - [Project IDX  \|  Google for Developers](https://developers.google.com/idx) + - Project IDX is an AI-assisted workspace for full-stack, multiplatform app development in the cloud. With support for a broad range of frameworks, languages, and services, alongside integrations with your favorite Google products, IDX streamlines your development workflow so you can build and ship apps across platforms with speed, ease, and quality. +- codeanywhere ## Extensions diff --git a/docs/economics/fintech-nbfc-banking-terms.md b/docs/economics/fintech-nbfc-banking-terms.md index 070a01760e0..a1b5b15d048 100755 --- a/docs/economics/fintech-nbfc-banking-terms.md +++ b/docs/economics/fintech-nbfc-banking-terms.md @@ -125,9 +125,9 @@ In the Fixed-Rate method, the interest is calculated on the entire loan amount. ## Covenant -In finance, a covenant is ==a promise or agreement between a borrower and lender that limits the borrower's actions, and ensures the borrower's financial ability to repay the loan==. Covenants are also known as debt covenants or banking covenants. +In finance, a covenant is ==a promise or agreement between a borrower and lender that limits the borrower's actions, and ensures the borrower's financial ability to repay the loan==. Covenants are also known as debt covenants or banking covenants. -Covenants can be financial, information, ownership, affirmative, negative, or positive. Examples of financial covenants include: +Covenants can be financial, information, ownership, affirmative, negative, or positive. Examples of financial covenants include: - **Financial ratios -** The borrower agrees to maintain a certain financial ratio, such as the interest coverage ratio, debt-to-equity ratio, or working capital ratio - **Restrictive covenants -** The borrower agrees to not take certain actions, such as issuing dividends, merging with another company, or purchasing or selling fixed assets without lender approval diff --git a/docs/knowledge/geography/new-home-place-house.md b/docs/knowledge/geography/new-home-place-house.md index aea012033fe..71acc29c757 100755 --- a/docs/knowledge/geography/new-home-place-house.md +++ b/docs/knowledge/geography/new-home-place-house.md @@ -123,7 +123,7 @@ https://zolostays.com/blog/cities-with-best-weather-in-india - Near beach & independent house with open areas and no humidity and no huge temperature variance (in a different country, with low income tax and high HDI, happiness index, low crime, low inequality, good education, no racism) - https://www.thehindu.com/real-estate/the-occupancy-certificate-why-it-is-now-more-important-than-ever/article19294876.ece - Natural Disaster area - earthquake, tsunami, hurricane, flood, drought, etc - - Global Seismic Hazard Map + - Global Seismic Hazard Map - **Don't buy a house more than 50 lakhs in India (1 crore house in america gives you a condo)** - In $300,000 (2.5 crore) you can get a 5 bedroom condo - Buy home from bank auctions - [auction property: Is it safe to buy auction property? Six things a buyer need to keep in mind - The Economic Times](https://economictimes.indiatimes.com/wealth/real-estate/is-it-safe-to-buy-auction-property-six-things-a-buyer-need-to-keep-in-mind/articleshow/103044435.cms) diff --git a/docs/networking/networking-concepts/peer-to-peer.md b/docs/networking/networking-concepts/peer-to-peer.md index c7d284a7949..e2eeba40820 100644 --- a/docs/networking/networking-concepts/peer-to-peer.md +++ b/docs/networking/networking-concepts/peer-to-peer.md @@ -42,7 +42,7 @@ Here is a general overview of how a P2P network operates. ### Node Initialization -When a new node in a P2P network boots up, it doesn’t know anything about the network, because there is no central server. Usually, developers provide a list of trusted nodes written directly into the code of the P2P client application that can be used for initial peer discovery. These trusted nodes could be centralized servers or peers depending upon the P2P application. +When a new node in a P2P network boots up, it doesn’t know anything about the network, because there is no central server. Usually, developers provide a list of trusted nodes written directly into the code of the P2P client application that can be used for initial peer discovery. These trusted nodes could be centralized servers or peers depending upon the P2P application. A node is usually identified by the following node triple: IP address, Port number, and node ID. The node ID should be unique with no collisions between peers. There are several methods to achieve this: