chlorane

Is it possible to use mini vg for training? How could I modify the code to make it suitable for training? The original VG is too large. Should I use...

I tried to download the models from "Benchmarking" section in this project and put them for test. I configured MODEL.WEIGHT_DET to the path of these models. The same problem happened...

> As stated in the documentation (https://github.com/jwyang/graph-rcnn.pytorch#evaluate) you have to pass the `--inference` flag if you only want to do testing. > See this code snippet in main.py: > >...

> Can you remove the `data` directory and run stage 1 again? In fact, I carefully checked the script and commented the following line: (asr1/asr.sh, near line 615) (# data/"${dset}"/wav.scp)...

> Thanks for the catch @chlorane- regarding memory issues, you should trim the features to fit the GPU capability you have- we used a custom Python script and didn't release...

I received the following error. Does this mean I need to use Cuda 10.0? Now I'm using Cuda 11.3. [compute04] 2023-02-09 10:22:49,083 (abs_task:1527) INFO: [valid] Batch sampler: LengthBatchSampler(N-batch=2, batch_bins=60000000, sort_in_batch=descending,...

> Yes you do. > > Alternatively, you can create a custom cuda kernel for Cuda 11.3 using steps outlined in https://github.com/allenai/longformer/blob/master/scripts/cheatsheet.txt Will the script run on cuda 10.2? I...

I tried Cuda 10.0 and Python 3.7 following the set of longformer package. Now, esp-net can be installed. However, when I run how2_2000h/asr1/run.sh, the following error happens [compute04] 2023-02-10 11:59:29,289...