also shout out to @clifapt for running BM25 via ES on LoCo first: https://
x.com/clifapt/status
/1755303087873429667
… my implementation seems to be slightly less performant than ElasticSearch; not sure why. maybe because i'm using a subword tokenizer, or suboptimal hyperparams?
OPEN SOURCE
-
BM25 Search Implementation Performance Comparison with ElasticSearch
By
–
-
Fast GPU-Enabled BM25 Implementation in PyTorch Achieves SOTA
By
–
implemented a fast, GPU-enabled BM25 in pytorch! BM25 is a simple search algorithm from the 70s that works as well as neural networks for most search problems; for all the advances we've made in neural text retrieval, it's still around got near SOTA on stanford LoCO benchmark
-
BM25_PT: Matrix Multiplication for Efficient Search Scoring
By
–
my library is here, bm25_pt: http://
github.com/jxmorris12/bm2
5_pt
… the key insight is that you can reduce the BM25 scoring to a single matrix multiplication. everything in the big fraction on the right side here can be stored in a big matrix of scores then, bm25(q) = bag(q) @ scores.T -
Mistral-7B KVCache Optimization for XLA Compilation
By
–
hey @fchollet . I think this is tricky to do in general (like a no free lunch way).
I think the Mistral-7B implementation KerasNLP writes out a KVCache (for compiler optimization) and specifically writes code in a way that is needed for XLA Compilation to work well.
Do let me -
LangChain Founders Share Building Story at ClickHouse Meetup
By
–
Want to hear more about the story of how we built LangChain? Come join us at the ClickHouse meetup in SF on March 4! Note: Signup required before this Friday, March 1st. https://
meetup.com/clickhouse-sil
icon-valley-meetup-group/events/299058486/
… -
Building JSON Agents with Ollama LangChain Neo4j
By
–
JSON-based Agents With Ollama & LangChain Learn to implement a Mixtral agent that interacts with a graph database Neo4j through a semantic layer This work by @tb_tomaz is great for a few reasons: – Shows how to build an agent with an OSS model
– Shows how to build and use a -
GPT-Fast Integrates Gemma with Optimized Token Performance
By
–
+/-12 lines of code in gpt-fast integrated Gemma from @GoogleDeepMind ; and shows:
* 234 tokens / sec on V100 in int8
* 144 tokens / sec on V100 in float precision https://
github.com/pytorch-labs/g
pt-fast/commit/ef055fc12188eaf80d8ba948ad743ee5583d0f3c
… -
Whisperkit revealed from Argmax ex-Apple team
By
–
Spoiler: it is Whisperkit from the ex-Apple folks at Argmax https://
takeargmax.com/blog/hello-wor
ld
… -
GTR Model Embeddings Bug Fix and Post-Processing Correction
By
–
unfortunately that's due to a bug I made when training the model; the gtr embeddings i use are missing a last post-processing step. someone trained a fixed model, more info here https://
github.com/jxmorris12/vec
2text/issues/28
… -
Scikit-LLM: Integrating Large Language Models with Scikit-Learn
By
–
Introducing Scikit-LLM: Scikit-Learn Meets Large Language Models! You can seamlessly integrate powerful language models like ChatGPT into scikit-learn for enhanced text analysis tasks. Here's what you can do with Scikit-LLM: 1. Zero-Shot Text Classification
2. Few-Shot Text