mtmd

**Original Issue**: https://github.com/flashlight/flashlight/issues/707 ### Summary This pull request improves the performance of the decoder (`fl_asr_decode`), targeting the transformer model, by offering the following optimizations: 1. Adding support for AMP in...

**Original Issue**: https://github.com/flashlight/flashlight/issues/630 ### Summary This pull request improves the performance of flashlight by offering a new data structure: Sample. Sample transfers the data that it contains to the GPU...

Enhancing the Performance of flashlight using cudnnFind, data-loader optimization, and control flow optimization

11

**Original Issue**: https://github.com/flashlight/flashlight/issues/630 ### Summary This commit improves the performance of flashlight using the following optimizations: 1. cudnnFind is used instead of the flashlight benchmark to improve the performance. 2....

Performance Enhancement Opportunity

2

### Feature Description Flashlight ASR decoder, `fl_asr_decode` can benefit from reduced precision computing and batch processing. In addition, the positional embedding function appears to be a bottleneck for both training...