TszSimLaw

File "train.py", line 1249, in main() File "train.py", line 1240, in main mp.spawn(run, args=(world_size, args), nprocs=world_size, join=True) File "/home/anaconda3/envs/kaldi/lib/python3.8/site-packages/torch/multiprocessing/spawn.py", line 199, in spawn return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')...

> Could you post more error logs? I had used gdb ; run train.py --world-size 4 --num-epochs 30 --start-epoch 1 --exp-dir pruned_transducer_stateless7_bbpe/exp --max-duration 400 error as above My torch version...

> > File "train.py", line 1249, in main() File "train.py", line 1240, in main mp.spawn(run, args=(world_size, args), nprocs=world_size, join=True) File "/home/anaconda3/envs/kaldi/lib/python3.8/site-packages/torch/multiprocessing/spawn.py", line 199, in spawn return start_processes(fn, args, nprocs, join,...

> Could you change > > ``` > File "/home/icefall/egs/tal_csasr/ASR/pruned_transducer_stateless7_bbpe/train.py", line 672, in compute_loss > y = k2.RaggedTensor(y).to(device) > ``` > > to > > ``` > print(y) > print(device)...

> Does it crash after printing? no crash. then print the log: [F] /home/runner/work/k2/k2/k2/csrc/device_guard.h:66:static int32_t k2::DeviceGuard::GetDevice() k2 compiled without CUDA support [ Stack-Trace: ] /home/anaconda3/envs/kaldi/lib/python3.8/site-packages/k2/lib/libk2_log.so(k2::internal::GetStackTrace()+0x47) [0x7ffeb6ab7077] /home/anaconda3/envs/kaldi/lib/python3.8/site-packages/_k2.cpython-38-x86_64-linux-gnu.so(+0x8439a) [0x7ffeb759e39a] /home/anaconda3/envs/kaldi/lib/python3.8/site-packages/_k2.cpython-38-x86_64-linux-gnu.so(+0x116daf) [0x7ffeb7630daf]...

> > [F] /home/runner/work/k2/k2/k2/csrc/device_guard.h:66:static int32_t k2::DeviceGuard::GetDevice() k2 compiled without CUDA support > > You are using a CPU version of k2. Please install a CUDA version. > > Please follow...