-
When modelling a large area it can be necessary to coarsen the grid beyond the default 100 m resolution dictated by the REGIS dataset. The method to_model_ds uses resample.ds_to_structured_grid, using…
-
When I use VCTK-DEMAND dataset to test, I found the sample_rate of wavs in the VCTK-DEMAND/test/ is 48000, but evaluation.py
> assert sr == 16000
so I add lines to resample,
> def evaluatio…
-
Image size is currently roughly 6gb. Which isn't good. As we aren't actually inferencing any ml model locally.
Most of the image size is due to torch and torchaudio installing 4-5 GBs of nvidia packa…
-
**Dear Mr.Zhong, may I ask some questions aboult the code.**
1. Why resample the dataset in the preprocessing stage, the uk-dale dataset original were sampled by 6 second, and I don't know why resamp…
-
Great, thanks for your example.
Did you try it with sample rate 44.1 using resampler?
Or maybe you know simple and working way to resample from 44.1 to 16 kHz ?
AFAIK, 44.1 is the only one sample…
-
As previously discussed with @ryentes I'm also posting the code to compute the Resasmpled Individual Reliability Index. This should provide a more reliable way to assess the individual reliability com…
-
Hello,
I want to convert the 3d 1*1*1 flair image of the same patient to 3d 1*1*5 flair. And I want it to be in the transversal plane. I opened applyxfm in the user interface in fsl.
I put the fla…
-
In resample step, i find log file every burst very large about 31.4Gb. And rsmp files no data. I can't find some bugs.
-
-
Hi, in the resample kernel part,when generating hann window to truncated sinc function,i confused why creat a 4 /times zeros hann window and then select the odd part. Is there any difference to diret…