CVXPY for scikit-learn estimators https://
bit.ly/4dt9Gfi
#AI #MachineLearning #DeepLearning #LLMs #DataScience
CODE
-
CVXPY Integration for scikit-learn Estimators
By
–
-
torch.compile CUDAGraph overhead optimization reinforcement learning
By
–
there's more CPU overhead, but it practically doesn't matter IMO unless you are doing tiny reinforcement-like workloads — and even there `torch.compile` will CUDAGraph it to near-zero overhead.
-
Claude Adds Data Analysis Feature Preview
By
–
BREAKING 🚨: Claude got Data Analysis as a feature preview.
— 🚨 AI News | TestingCatalog (@testingcatalog) 24 octobre 2024
With this feature, it can execute the code to run complex data visualisations and render them via Artifacts. https://t.co/99xovpNVqO pic.twitter.com/yxeHS3BgudBREAKING : Claude got Data Analysis as a feature preview. With this feature, it can execute the code to run complex data visualisations and render them via Artifacts.
-
Two Quantization Techniques for Model Optimization Released
By
–
We used two different techniques for quantizing these models.
Quantization-Aware Training with LoRA adaptorsprioritizing accuracy.
SpinQuant, a post-training quantization method which prioritizes portability.
Both versions are available for download as part of this release. -
Managing LLM Prompts Across Cross-Functional Teams
By
–
We commonly see that teams building LLM apps are getting more and more cross functional
— Harrison Chase (@hwchase17) 24 octobre 2024
A lot of their collaboration happens on the prompts!
Having a way to properly tag prompts makes it easier to pull new versions into code, but allow them live and be managed outside https://t.co/PCRCpnOvcDWe commonly see that teams building LLM apps are getting more and more cross functional A lot of their collaboration happens on the prompts! Having a way to properly tag prompts makes it easier to pull new versions into code, but allow them live and be managed outside
-
Claude adds code execution and mathematical analysis capabilities
By
–
Claude can now write and run code.
— Anthropic (@AnthropicAI) 24 octobre 2024
We've added a new analysis tool. The tool helps Claude respond with mathematically precise and reproducible answers. You can then create interactive data visualizations with Artifacts.
Enable the feature preview: https://t.co/bJ8BjBT6zG. pic.twitter.com/Jq5xOHBmiRClaude can now write and run code. We've added a new analysis tool. The tool helps Claude respond with mathematically precise and reproducible answers. You can then create interactive data visualizations with Artifacts. Enable the feature preview: https://
claude.ai/new?fp=1. -
MobileNetV3 Image Classification Community Template
By
–
Community Template 3: Image Classification with MobileNetV3 from @jsonmathsai (
https://
github.com/SakanaAI/AI-Sc
ientist/pull/141
…) -
Sketch-RNN Community Template from Alexis D Jacq
By
–
Community Template 2: Sketch-RNN from @Alexis_D_Jacq (
https://
github.com/SakanaAI/AI-Sc
ientist/pull/143
…) -
Infectious Disease Modeling Community Template Released
By
–
Community Template 1: Infectious Disease Modeling , our first non-strictly ML template from @strayer_13 (
https://
github.com/SakanaAI/AI-Sc
ientist/pull/137
…) -
Alternative file format proposed for AI model distribution
By
–
Yep i've seen this but I feel like it'd be cleaner to do a distinct file format for this (GGUF is very LLM centric, including metadata etc)