Transformer Inference on Spartan-7 XC7S15 FPGA
3.77RMSE
Resource-aware Mixed-precision Quantization
Updated 1mo ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| 3.77 | 5.78 | 31 | 39 | 70 | 0.405 | ||
| 4.1 | 11.92 | 31 | 37 | 68 | 0.811 | ||
| 4.19 | 8.79 | 31 | 37 | 68 | 0.598 |
Resource-aware Mixed-precision Quantization
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| 3.77 | 5.78 | 31 | 39 | 70 | 0.405 | ||
| 4.1 | 11.92 | 31 | 37 | 68 | 0.811 | ||
| 4.19 | 8.79 | 31 | 37 | 68 | 0.598 |