optimum-benchmark
Performance comparer
A tool for comparing and optimizing the performance of various machine learning frameworks and models on different hardware platforms.
🏋️ A unified multi-backend utility for benchmarking Transformers, Timm, PEFT, Diffusers and Sentence-Transformers with full support of Optimum's hardware optimizations & quantization schemes.
274 stars
5 watching
50 forks
Language: Python
last commit: 4 months ago
Linked from 1 awesome list
benchmarkneural-compressoronnxruntimeopenvinopytorchtensorrt-llmtext-generation-inference
Related projects:
Repository | Description | Stars |
---|---|---|
| A PyTorch quantization backend for models. | 847 |
| Provides benchmarking functions for mathematical optimization algorithms | 67 |
| An evaluation framework for machine learning models and datasets, providing standardized metrics and tools for comparing model performance. | 2,063 |
| Automated performance testing and analysis tool for distributed systems | 25 |
| An automated toolbox for performing worst-case analyses of first-order optimization methods in convex and nonconvex optimization. | 53 |
| A collection of benchmark problems for hyperparameter optimization | 140 |
| A package for optimizing hyperparameters and meta-learning using gradient-based methods in TensorFlow. | 188 |
| Comparative benchmarks of various machine learning algorithms | 169 |
| A reinforcement learning-based framework for optimizing hyperparameters in distributed machine learning environments. | 15 |
| An open-source benchmarking framework for evaluating cross-style visual capability of large multimodal models | 84 |
| A benchmark for evaluating machine learning algorithms' performance on large datasets | 1,874 |
| A performance benchmarking and testing framework for .NET applications. | 534 |
| An authoritative source of real-world benchmarks for Python implementations. | 877 |
| Automated machine learning tool for tabular data pipelines | 343 |
| A tool for comparing the performance of different programming languages or code snippets in a controlled and measurable way | 18 |