Pinned Loading

1. Large-Scale-Training-Inference-and-Continuous-Deployment-Platform Large-Scale-Training-Inference-and-Continuous-Deployment-Platform Public

   Forked from theomthakur/ece-gy-9183-group19

   Jupyter Notebook

2. CUDA-performance-optimization-project CUDA-performance-optimization-project Public

   GPU optimization techniques for ML: memory coalescing, tiling, and convolution implementations

   Cuda

3. edge-ml-optimization edge-ml-optimization Public

   Post-training quantization (PTQ) for edge deployment. INT8 implementation with per-channel scaling, activation calibration, and bias correction for 4x compression with <1% accuracy loss.

   Jupyter Notebook

4. multi-gpu-training multi-gpu-training Public

   PyTorch DDP implementation exploring gradient synchronization and scaling dynamics on multi-GPU clusters. Benchmarks communication bottlenecks and large-batch convergence strategies.

   Python

5. Low-Rank-RoBERTa Low-Rank-RoBERTa Public

   Repository for Low-Rank RoBERTa: Achieving High Accuracy Under 1M Parameters

   Jupyter Notebook

6. speculative-moe speculative-moe Public

   Python