OPEN SOURCE

–

08 July 2024 16h46

6. Drawdata A python library that allows you to draw a 2-D dataset of any shape in a Jupyter Notebook.

MLX: Deep Learning Framework for Machine Learning on macOS

By

–

08 July 2024 16h46

5. MLX: A deeplearning framework for machine learning on macOS! You can efficiently train and deploy ML models. The repo also includes examples to run open-source models using MLX, like LLaMA, Mixtral 8x7B, LoRA, CLIP, Whisper and Stable Diffusion. https://
github.com/ml-explore/mlx

LazyPredict: Train and Evaluate All ML Models Instantly

By

–

08 July 2024 16h46

4. LazyPredict Lazy Predict allows you to train, test and evaluate all your ML models at once with just a few lines of code.

Pandas AI: Natural Language Data Analysis and Visualization

By

–

08 July 2024 16h45

2. Pandas AI Pandas AI adds Generative AI capabilities to Pandas. Now you can analyze complex data frames and plot visualizations just by using natural language.

7 Amazing Python and Machine Learning Libraries

By

–

08 July 2024 16h45

I have been coding in Python for 4 years now. Here are 7 amazing Python and Machine Learning libraries not known to many:

Jamba’s 256K Context Reveals fused_moe Kernel Issues

By

–

08 July 2024 16h44

BTW, Since Jamba supports a 256K context with high throughput, we also stumbled upon an issue where the fused_moe kernel didn’t work well in long contexts. Others seems to have had this too, according to some other open issues

MoE Token Independence Enables Efficient Long Context Processing in vLLM

By

–

08 July 2024 16h44

@avshalomm solved it by utilizing the fact that actually, there is no interaction between different tokens in the MoE block, so we can iterate over the long context in chunks. This was also merged and now fixed in vLLM

Jamba Modeling and Mamba Cache Management Updates

By

–

08 July 2024 16h44

Things in this PR:
– Jamba modeling file, supporting Jamba variations
– Mamba cache management (also benefit other Mamba-based models)
– Send more request related properties to the forward pass, for model-specific implementations (including things needed for speculative decoding)

Jamba Model Now Supported in vLLM for Efficient Serving

By

–

08 July 2024 16h44

Jamba support is now live on vLLM Due to its novel hybrid SSM-Transformer arch, Jamba didn’t work out-of-the-box in vLLM. our own @MorZusman worked together with @vllm_project to integrate Jamba for an efficient serving in vLLM