feat: AI-OCR for PDFs #455
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
statefb
reviewed
Jul 19, 2024
| ``` | ||
|
|
||
| ## Launch local server | ||
|
|
||
| ```sh | ||
| pip install poetry --no-cache-dir |
There was a problem hiding this comment.
ドキュメント修正ありがとうございますmm（ちょっとした細かいドキュメント修正、何気に結構助かります）
| const setUp = async (dbConfig) => { | ||
| const client = new Client(dbConfig); | ||
| // Aurora Serverless may be down, so retry until it connect | ||
| async function connectWithRetry(maxRetries = 5, retryDelay = 60000) { |
There was a problem hiding this comment.
memo: aurora serverless v2はcapacity zeroまでscale in しないはず。別の原因？
| platform: Platform.LINUX_AMD64, | ||
| file: "Dockerfile", | ||
| exclude: [ |
| platform: Platform.LINUX_AMD64, | ||
| file: "lambda.Dockerfile", | ||
| cmd: ["app.sqs_consumer.handler"], | ||
| exclude: [ |
| @@ -102,6 +104,7 @@ export class BedrockChatStack extends cdk.Stack { | |||
| exclude: [ | |||
| "**/node_modules/**", | |||
| "**/dist/**", | |||
| "**/dev-dist/**", | |||
|
Memo: When the bedrock knowledge base retrieve api supports detail reference chunk, this PR could be a good reference. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Issue #, if available:
Using AI for PDF OCR
Description of changes:
PDFs contain various types of images, graphs, charts, designs, objects, etc., and these PDFs are easy for people to understand. However, typical OCR tools can't understand relationships between objects. Tabular tables are single-column strings, so it's difficult to infer table relationships from strings. The AI-OCR feature converts PDFs to images one by one and supports OCR using Claude3's multimodal features. As a result, structured markdown text can be retrieved and used as RAG knowledge.
The DB schema has changed. Add “Metadata” as JSON. In this PR, images converted from PDFs are stored in S3 buckets.
It is placed in the source image URL and placed in the original pdfurl of the metadata.parentSource.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.