Skip to content

Commit 972fe61

Browse files
authored
Update quatization file name (#78)
1 parent ffc43ad commit 972fe61

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

Guides/MAX78000_Workflow_Guide.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -43,7 +43,7 @@ Quantization is the process of converting floating point weights from a checkpoi
4343

4444
Before working in the synthesis directory, be sure to run _deactivate_ followed by _source bin/activate_ in order to configure the python environment for synthesis.
4545

46-
Continuing with our mnist example, copy the _best.pth.tar_ file from the logs/\<time/date\> directory into the trained directory and rename it to _ai85-mnist-qat8.pth.tar_ (you may need to first delete the pre-existing file of this name). This will allow you to use the existing mnist quantize script, _scripts/quantize\_mnist.sh_ located in the _scripts_ directory.
46+
Continuing with our mnist example, copy the _qat_best.pth.tar_ file from the logs/\<time/date\> directory into the trained directory and rename it to _ai85-mnist-qat8.pth.tar_ (you may need to first delete the pre-existing file of this name). This will allow you to use the existing mnist quantize script, _scripts/quantize\_mnist.sh_ located in the _scripts_ directory.
4747

4848
quantize\_mnist.sh:
4949

0 commit comments

Comments
 (0)