Skip to content

Fix quantization 1.2 document fm_cut not renamed issue. #71

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
MrWhoami merged 1 commit into from
Oct 9, 2025
Merged

Fix quantization 1.2 document fm_cut not renamed issue. #71

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/toolchain/quantization/1.2_Flow_and_Steps.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,7 +76,7 @@ In this stage, you will complete the conversion from a floating-point model to a
| 12 | outlier_factor | Parameter Usage: Used under 'mmse' mode.The factor applied on outliers. <br /> Range: 1.0 ~ 2.0 or higher. <br /> Default value: 1.0 <br /> Description: For example, if clamping data is sensitive to your model, set outlier_factor to 2 or higher. Higher outlier_factor will reduce outlier removal by increasing range. | Optional |
| 13 | quantize_mode | Parameter Usage: Need extra tuning or not. <br /> Range: "default" / "post_sigmoid" <br /> Default value: "default" <br /> Description: If a model's output nodes were ALL sigmoids and had been removed, choose "post_sigmoid" for better performance. | Optional |
| 14 | quan_config | Parameter Usage: User can pass in a dictionary to set constraints for quantization. <br /> Range: N/A <br /> Default value: None <br /> Description: Set scale, radix, min/max value for specific operator node. | Optional |
| 15 | fm_cut | Parameter Usage: Methods to search for best feature map cut. <br /> Range: "default" / "deep_search" / "performance" (not available yet) <br /> Default value: "default" <br /> Description: To improve the efficiency of NPU. | Optional |
| 15 | compiler_tiling | Parameter Usage: Methods to search for best feature map cut. <br /> Range: "default" / "deep_search" / "performance" (not available yet) <br /> Default value: "default" <br /> Description: To improve the efficiency of NPU. | Optional |
| 16 | p_output | Parameter Usage: Location to save the generated fix models. <br /> Range: N/A <br /> Default value: "/data1/kneron_flow" <br /> Description: | Optional |
| 17 | weight_bandwidth | Parameter Usage: Set the weight bandwidth. <br /> Range: <br /> Default value: None <br /> Description: | Optional |
| 18 | dma_bandwidth | Parameter Usage: Set the dma bandwidth. <br /> Range: <br /> Default value: None <br /> Description: | Optional |
Expand Down
Loading