ظاهرن یکی از برتری های فعلی docling نسبت به MarkItDown اینه که داکلینگ توی LlamaIndex و LangChain هم Integrate شده 🔗

یکی از کتاب های مناسب برای مباحث پایه ریاضی مورد نیاز در علوم کامپیوتر

@ai_python

https://courses.csail.mit.edu/6.042/spring18/mcs.pdf


➕ Linear algebra
➕ Series
➕ Logic
➕ Probability
➕ Number theory
➕ Graphs

@ai_python

هر ری اکتی ، ری اکت نیست 🤪

@ai_python

https://cloud.google.com/bigquery/docs/vector-index-text-search-tutorial

Perform semantic search and retrieval-augmented generation

👉 This tutorial guides you through the end-to-end process of creating and using text embeddings, including using vector indexes to improve search performance.

This tutorial covers the following tasks:

🖥 Creating a BigQuery ML remote model over a Vertex AI embedding model.

🖥 Using the remote model with the ML.GENERATE_EMBEDDING function to generate embeddings from text in a BigQuery table.

🖥 Creating a vector index to index the embeddings.

🖥 Using the VECTOR_SEARCH function with the embeddings to search for similar text.

🖥 Perform retrieval-augmented generation (RAG) by generating text with the ML.GENERATE_TEXT function, and using vector search results to augment the prompt input and improve results.

This tutorial uses the BigQuery public table patents-public-data.google_patents_research.publications.

@ai_python

لاوبل اینا هم از فایرکراول استفاده می کنن 🥹

یک پایپ لاین بسیار مناسب برای چانک کردن فایل های صوتی طولانی به طوری که برای کار با مدل های Whisper مربوط به Groq مناسب باشن :

@ai_python

https://github.com/groq/groq-api-cookbook/blob/main/tutorials/audio-chunking/audio_chunking_tutorial.ipynb


Groq is great for processing long audio files thanks to its fast inference speeds and even hours of audio that we process into chunks can be transcribed in a matter of minutes. As such, we'll use Whisper Large V3 powered by Groq and learn how to:

@ai_python

👂 Preprocess audio files for optimal transcription
👂 Split audio files into manageable chunks
👂 Implement a smart overlap for our chunks
👂 Transcribe our chunks using Whisper Large V3
👂 Merge our results while properly handling overlaps
👂 Save our transcriptions in multiple formats for further handling

مدل TimesFM یکی از مدل های پیش بینی Time Series هست که توسط گوگل توسعه داده شده است. این مدل در Hugging Face نیز در دسترس قرار دارد :

@ai_python

https://huggingface.co/google/timesfm-2.0-500m-pytorch

مقاله های مرتبط :

🔗 https://arxiv.org/abs/2310.10688

🔗 https://arxiv.org/abs/2402.02592