Gists

Abstract

Gemini excels at text generation with RAG for large datasets, but smaller ones benefit from prompting or data upload. This report explores using Gemini 1.5 Flash/Pro with RAG on medium-sized, Google Spreadsheet-stored datasets for improved accuracy and effectiveness.

Introduction

Gemini’s text generation capabilities have seen significant advancements with the Retrieval-Augmented Generation (RAG). This approach excels for large datasets, where embedding data and querying the model leads to high-quality answers. However, for smaller datasets, directly including data in the prompt or an uploaded file can be more efficient. Ref

Gists

Abstract

This research explores “pseudo function calling” in Gemini API using prompt engineering with JSON schema, bypassing model dependency limitations.

Introduction

Large Language Models (LLMs) like Gemini and ChatGPT offer powerful functionalities, but their capabilities can be further extended through function calling. This feature allows the LLM to execute pre-defined functions with arguments generated based on the user’s prompt. This unlocks a wide range of applications, as demonstrated in these resources (see References).

GeminiWithFiles was updated to v2.0.2

• v2.0.2 (September 26, 2024)

  1. As the option for generationConfig, the properties response_schema and temperature were added.

You can see the detail information here https://github.com/tanaikech/GeminiWithFiles

Gists

Abstract

This report presents a method to train AI to effectively generate content from smaller, structured datasets using Python. Gemini’s token processing capabilities are leveraged to effectively utilize limited data, while techniques for interpreting CSV and JSON formats are explored.

Introduction

In the era of rapidly advancing artificial intelligence (AI), the ability to analyze and leverage large datasets is paramount. While RAG (Retrieval Augmented Generation) environments are often ideal for such tasks, there are scenarios where content generation needs to be achieved with smaller datasets.

Gists

Abstract

This report improves Gmail email labeling with Gemini API using JSON schema and leverages advancements in Gemini 1.5 Flash for faster processing.

Introduction

As Gemini continues to evolve, existing scripts utilizing its capabilities can be revisited to improve efficiency and accuracy. This includes the process of flexible labeling for Gmail emails using the Gemini API. I have previously explored this topic in two reports:

Abstract

This post introduces a Google Apps Script solution that automates the creation, sharing, and monitoring of multiple Google Spreadsheets, providing a more efficient and streamlined approach to managing user data.

Introduction

I’ve often encountered requests from clients who need to manage multiple Google Spreadsheets for various users, often by copying a template spreadsheet. In these situations, I typically propose the following approach:

Gists

Abstract

This report presents a method to optimize AI-generated scripts for processing costs using Gemini and Google Apps Script. By incorporating external knowledge from sources like StackOverflow, we demonstrate the effective generation of efficient scripts that minimize overhead while maintaining desired outcomes. This approach can be considered a dynamic pseudo-RAG technique.

Introduction

The proliferation of generative AI, exemplified by Google Gemini, has led to a surge in AI-generated scripts. This trend is evident in the growing number of questions on platforms like StackOverflow that involve AI-generated scripts. While this indicates a significant improvement in AI performance, it’s crucial to note that AI-generated scripts may not always be optimized for processing costs, especially when the prompt fails to provide sufficient context.

UtlApp was updated to v1.0.7.

• v1.0.7 (September 4, 2024)

  1. Following 3 methods were added.
  • snake_caseToCamelCase: This method is used for converting a string of the snake case to the camel case.
  • camelCaseTosnake_case: This method is used for converting a string of the camel case to the snake case.
  • createFormDataObject: This method is used for creating the form data to HTTP request from an object.

You can see the detail information here https://github.com/tanaikech/UtlApp

Gists

Abstract

A script using resumable upload with file streams is proposed to enhance file handling within the Gemini Generative AI API for Node.js. This script allows uploading from web URLs and local storage, efficiently handles large files, and offers potential reusability with other Google APIs.

Description

The @google/generative-ai library provides a powerful way to interact with the Gemini Generative AI API using Node.js. This enables developers to programmatically generate creative text formats, translate languages, write different kinds of creative content, and answer your questions in an informative way, all powered by Gemini’s advanced AI models. Ref

Gists

Abstract

This study proposes a workaround to address the Gemini API’s current inability to directly process web content from URLs. By utilizing Google Apps Script, the method extracts relevant information from a specified URL and feeds it into the API for summarization. This approach offers a solution for generating comprehensive summaries from web-based content until the API’s limitations are resolved.

Introduction

While Gemini API offers powerful text generation capabilities, it currently faces limitations in directly accessing and processing web content from URLs. When prompted to summarize an article at a specific URL like Summarize the article at the following URL. https://###, the API often returns an error message indicating its inability to retrieve the necessary information. This limitation arises from the API’s current design, which may not be equipped to handle web requests and parse HTML content.