Python script using the large model doesn't work

[vegandiet705]

I am trying to run the following python script, which uses the large model:

import torchaudio
from audiocraft.models import MusicGen
from audiocraft.data.audio import audio_write

model = MusicGen.get_pretrained('large')
model.set_generation_params(duration=8)  # generate 8 seconds.

descriptions = ['happy rock', 'energetic EDM', 'sad jazz']

wav = model.generate(descriptions)  # generates 3 samples.

for idx, one_wav in enumerate(wav):
    # Will save under {idx}.wav, with loudness normalization at -14 db LUFS.
    audio_write(f'{idx}', one_wav.cpu(), model.sample_rate, strategy="loudness")

I got the following logs:

Downloading: "https://dl.fbaipublicfiles.com/audiocraft/musicgen/v0/9b6e835c-1f0cf17b5e.th" to /home/worker/.cache/torch/hub/checkpoints/9b6e835c-1f0cf17b5e.th
100%|██████████████████████████████████████| 6.07G/6.07G [03:57<00:00, 27.4MB/s]
Killed

It's worth mentioning that the computer freezes for a while before giving the Killed output.

Regarding my hardware, I have an AMD Ryzen 5 2500U processor with Radeon Vega Mobile Gfx and Nvidia GeForce GTX GPU (the appropriate drivers are already installed).

I'm on Ubuntu 20.04 OS.

Here is my output for nvidia-smi command:

Mon Jun 12 00:45:05 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 520.61.05    Driver Version: 520.61.05    CUDA Version: 11.8     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  On   | 00000000:01:00.0 Off |                  N/A |
| N/A   38C    P8    N/A /  N/A |      9MiB /  2048MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1056      G   /usr/lib/xorg/Xorg                  4MiB |
|    0   N/A  N/A      2033      G   /usr/lib/xorg/Xorg                  4MiB |
+-----------------------------------------------------------------------------+

Here is my output to lscpu:

Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   43 bits physical, 48 bits virtual
CPU(s):                          8
On-line CPU(s) list:             0-7
Thread(s) per core:              2
Core(s) per socket:              4
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       AuthenticAMD
CPU family:                      23
Model:                           17
Model name:                      AMD Ryzen 5 2500U with Radeon Vega Mobile Gfx
Stepping:                        0
Frequency boost:                 enabled
CPU MHz:                         1600.000
CPU max MHz:                     2000,0000
CPU min MHz:                     1600,0000
BogoMIPS:                        3992.30
Virtualization:                  AMD-V
L1d cache:                       128 KiB
L1i cache:                       256 KiB
L2 cache:                        2 MiB
L3 cache:                        4 MiB
NUMA node0 CPU(s):               0-7

Thanks in advance for any help

[schen2315]

spent some time trying to figure out why i was getting Killed too. It's most likely an out of memory error. Right after you run you can tail your kern logs:

sudo tail /var/log/kern.log
#  Out of memory: Killed process 1004 ...