Pinned Repositories
chain_constant_memory
Example of backprop which uses constant memory
imperative
imperative programming in TensorFlow
kfac_pytorch
memory_probe_ops
TensorFlow kernels for probing memory
memory_util
TensorFlow util for building memory usage timeline from LOG_MEMORY messages
notebooks
stuff
Stuff I uploaded to share online or to access from a different machine
tensorflow-community-wheels
Place to upload links to TensorFlow wheels
yaroslavvb's Repositories
yaroslavvb/tensorflow-community-wheels
Place to upload links to TensorFlow wheels
yaroslavvb/stuff
Stuff I uploaded to share online or to access from a different machine
yaroslavvb/kfac_pytorch
yaroslavvb/memory_util
TensorFlow util for building memory usage timeline from LOG_MEMORY messages
yaroslavvb/chain_constant_memory
Example of backprop which uses constant memory
yaroslavvb/max_align_bytes_op
TensorFlow op to return EIGEN_MAX_ALIGN_BYTES
yaroslavvb/tf_build
TensorFlow build on AWS
yaroslavvb/bflm
bflm
yaroslavvb/newton
yaroslavvb/cluster
train on AWS
yaroslavvb/dominic_densenet
yaroslavvb/early-eager
yaroslavvb/einograd
yaroslavvb/kaczmarz
yaroslavvb/2024
yaroslavvb/asdl
ASDL: Automatic Second-order Differentiation Library for PyTorch
yaroslavvb/dawn-bench-entries
DAWNBench: An End-to-End Deep Learning Benchmark and Competition
yaroslavvb/fbpcp
FBPCP (Facebook Private Computation Platform) is a secure, privacy safe and scalable architecture to deploy MPC (Multi Party Computation) applications in a distributed way on virtual private clouds. FBPCF (Facebook Private Computation Framework) is for scaling MPC computation up via threading, while FBPCP is for scaling MPC computation out via Private Scaling architecture.
yaroslavvb/fbpcs
FBPCS (Facebook Private Computation Solutions) leverages secure multi-party computation (MPC) to output aggregated data without making unencrypted, readable data available to the other party or any third parties. Facebook provides impression & opportunity data, and the advertiser provides conversion / outcome data. Both parties have dedicated cloud computing instances living on separate Virtual Private Clouds (VPCs) that are connected to allow network communication. The FBPMP products that have been implemented are Private Lift and Private Attribution. It’s expected that more products will be implemented and added to the Private Measurement suite.
yaroslavvb/machine-learning-systems-design
A booklet on machine learning systems design with exercises
yaroslavvb/Megatron-LM
Ongoing research training transformer language models at scale, including: BERT
yaroslavvb/models
Models built with TensorFlow
yaroslavvb/pytorch-pretrained-BERT
📖The Big-&-Extending-Repository-of-Transformers: Pretrained PyTorch models for Google's BERT, OpenAI GPT & GPT-2, Google/CMU Transformer-XL.
yaroslavvb/ray
A high-performance distributed execution engine
yaroslavvb/retro
Official repo to On the Generalization Ability of Retrieval-Enhanced Transformers
yaroslavvb/SPADE
Semantic Image Synthesis with SPADE
yaroslavvb/studio
Studio: Simplify and expedite model building process
yaroslavvb/tensorflow
Computation using data flow graphs for scalable machine learning
yaroslavvb/tensorpack
A Neural Net Training Interface on TensorFlow
yaroslavvb/transformer-xl