briansp2020

#/bin/sh
cd
export HSA_OVERRIDE_GFX_VERSION=11.0.0
export PYTORCH_ROCM_ARCH="gfx1100"
export HIP_VISIBLE_DEVICES=0
export ROCM_PATH=/opt/rocm-6.3.2
echo "export HSA_OVERRIDE_GFX_VERSION=11.0.0" | tee --append .bashrc
echo "export PYTORCH_ROCM_ARCH=\"gfx1100\"" | tee --append .bashrc
echo "export HIP_VISIBLE_DEVICES=0" | tee --append .bashrc
echo "export ROCM_PATH=/opt/rocm-6.3.2" | tee --append .bashrc

briansp2020

FROM ubuntu:22.04

RUN apt update && apt install -y curl rsync \
  && curl -O https://repo.radeon.com/amdgpu-install/6.3.2/ubuntu/jammy/amdgpu-install_6.3.60302-1_all.deb \
  && dpkg -i amdgpu-install_6.3.60302-1_all.deb \
  && DEBIAN_FRONTEND=noninteractive amdgpu-install --usecase=rocm --no-dkms --no-32 -y

RUN apt install -y git git-lfs pkg-config patchelf nano wget gpg cmake build-essential libgtest-dev \
    python3-venv python3-dev python3-wheel python3-setuptools python3-requests python3-packaging python-is-python3 \
  && apt clean

briansp2020

(pt) root@rocm:~# python tmp/quickstart.py
/root/pt/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/root/pt/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ConvNeXt_Small_Weights.IMAGENET1K_V1`. You can also use `weights=ConvNeXt_Small_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
epoch     train_loss  valid_loss  error_rate  time
0         0.110175    0.006588    0.001353    00:34
epoch     train_loss  valid_loss  error_rate  time
0         0.015370    0.002482    0.000677    00:47
Training text processing model

briansp2020

[    0.000000] Linux version 5.15.0-97-generic (buildd@lcy02-amd64-033) (gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0, GNU ld (GNU Binutils for Ubuntu) 2.38) #107-Ubuntu SMP Wed Feb 7 13:26:48 UTC 2024 (Ubuntu 5.15.0-97.107-generic 5.15.136)
[    0.000000] Command line: BOOT_IMAGE=/vmlinuz-5.15.0-97-generic root=/dev/mapper/ubuntu--vg-ubuntu--lv ro
[    0.000000] KERNEL supported cpus:
[    0.000000]   Intel GenuineIntel
[    0.000000]   AMD AuthenticAMD
[    0.000000]   Hygon HygonGenuine
[    0.000000]   Centaur CentaurHauls
[    0.000000]   zhaoxin   Shanghai  
[    0.000000] BIOS-provided physical RAM map:
[    0.000000] BIOS-e820: [mem 0x0000000000000000-0x000000000009ffff] usable

briansp2020

#/bin/sh
cd
export HSA_OVERRIDE_GFX_VERSION=11.0.0
export PYTORCH_ROCM_ARCH="gfx1100"
export HIP_VISIBLE_DEVICES=0
export ROCM_PATH=/opt/rocm
echo "export HSA_OVERRIDE_GFX_VERSION=11.0.0" | tee --append .bashrc
echo "export PYTORCH_ROCM_ARCH=\"gfx1100\"" | tee --append .bashrc
echo "export HIP_VISIBLE_DEVICES=0" | tee --append .bashrc
echo "export ROCM_PATH=/opt/rocm" | tee --append .bashrc

briansp2020

bsp2020@rocm:~$ sudo reboot
Connection to 10.0.0.200 closed by remote host.
Connection to 10.0.0.200 closed.
(base) bsp2020@Ryzen5950X:~$ ssh 10.0.0.200
Welcome to Ubuntu 22.04.3 LTS (GNU/Linux 5.15.0-88-generic x86_64)

 * Documentation:  https://help.ubuntu.com
 * Management:     https://landscape.canonical.com
 * Support:        https://ubuntu.com/advantage

briansp2020

FROM ubuntu:22.04

RUN apt update && apt install -y curl \
  && curl -O https://repo.radeon.com/amdgpu-install/5.7.1/ubuntu/jammy/amdgpu-install_5.7.50701-1_all.deb \
  && dpkg -i amdgpu-install_5.7.50701-1_all.deb \
  && DEBIAN_FRONTEND=noninteractive amdgpu-install --usecase=rocm --no-dkms --no-32 -y \
  && apt install -y git git-lfs pkg-config patchelf nano wget gpg cmake build-essential \
    python3-venv python3-dev python3-wheel python3-setuptools python3-requests python3-packaging \
  && apt clean

briansp2020

root@rocm:/root/tmp# python benchmark.py
2023-10-22 19:31:47.927753: I tensorflow/core/util/port.cc:110] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2023-10-22 19:31:47.946907: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 AVX512F AVX512_VNNI AVX512_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-10-22 19:31:48.774214: I tensorflow/compiler/xla/stream_executor/rocm/rocm_gpu_executor.cc:838] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-10-22 19:31:48.785167: I tensorflow/compiler/xla/stream_executor/rocm/rocm_gpu_execu

briansp2020

(tf) root@rocm:~/tmp# python benchmark.py
2023-10-14 15:02:22.116047: E external/local_xla/xla/stream_executor/plugin_registry.cc:93] Invalid plugin kind specified: DNN
2023-10-14 15:02:22.348480: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 AVX512F AVX512_VNNI AVX512_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-10-14 15:02:23.756833: I external/local_xla/xla/stream_executor/rocm/rocm_gpu_executor.cc:787] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-10-14 15:02:23.982269: I external/local_xla/xla/stream_executor/rocm/rocm_gpu_executor.cc:787] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2023-10-14 15:02:23.9823

briansp2020

from fastai.vision.all import *
from fastai.text.all import *
from fastai.collab import *
from fastai.tabular.all import *

path = untar_data(URLs.PETS)/'images'

def is_cat(x): return x[0].isupper()
dls = ImageDataLoaders.from_name_func(
    path, get_image_files(path), valid_pct=0.2, seed=42,