【2020年版】NVIDIA Jetson Nanoで StyleGANの改良版の StyleGAN2で自然な画像を生成

【2020年版】NVIDIA Jetson Nano、Jetson Xavier NXの便利スクリプト

【2020年】Jetson Xavier NX 開発者キットが安かったので衝動買いした件、標準販売価格５万円が４万４千円！

【2020年版】NVIDIA Jetson Nanoの初期設定、最高速で動かす設定、空きメモリを増やす方法等

cd
git clone https://github.com/NVlabs/stylegan2 --depth 1
cd stylegan2

# undefined symbol _ZN10tensorflow12OpDefBuilder5InputESs
sed -i -e "s/-D_GLIBCXX_USE_CXX11_ABI=0/-D_GLIBCXX_USE_CXX11_ABI=1/g" dnnlib/tflib/custom_ops.py

cat dnnlib/tflib/custom_ops.py | grep ABI
# compile_opts += ' --compiler-options \'-fPIC -D_GLIBCXX_USE_CXX11_ABI=1\''

# Generate uncurated ffhq images (matches paper Figure 12)
python3 run_generator.py generate-images --network=gdrive:networks/stylegan2-ffhq-config-f.pkl \
  --seeds=6600-6625 --truncation-psi=0.5
# dnnlib: Finished run_generator.generate_images() in 5m 46s.
# Jetson Nano 5m 46s
# Xavier NX 3m 06s

# Generate curated ffhq images (matches paper Figure 11)
python3 run_generator.py generate-images --network=gdrive:networks/stylegan2-ffhq-config-f.pkl \
  --seeds=66,230,389,1518 --truncation-psi=1.0
# dnnlib: Finished run_generator.generate_images() in 6m 08s.
# Jetson Nano 6m 08s
# Xavier NX 1m 52s

# Generate uncurated car images
python3 run_generator.py generate-images --network=gdrive:networks/stylegan2-car-config-f.pkl \
  --seeds=6000-6025 --truncation-psi=0.5
# dnnlib: Finished run_generator.generate_images() in 4m 15s.
# Jetson Nano 4m 15s
# Xavier NX 2m 9s

# Example of style mixing (matches the corresponding video clip)
python3 run_generator.py style-mixing-example --network=gdrive:networks/stylegan2-ffhq-config-f.pkl \
  --row-seeds=85,100,75,458,1500 --col-seeds=55,821,1789,293 --truncation-psi=1.0
# メモリ不足で死亡 OOM = Out Of Memory
#   (0) Resource exhausted: OOM when allocating tensor with shape[1,128,1025,1025] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc
# [[{{node G_synthesis_1/_Run/G_synthesis/1024x1024/Conv0_up/conv2d_transpose}}]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.
# [[G_synthesis_1/_Run/saturate_cast/_1457]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.
#  (1) Resource exhausted: OOM when allocating tensor with shape[1,128,1025,1025] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc{3}# [[{{node G_synthesis_1/_Run/G_synthesis/1024x1024/Conv0_up/conv2d_transpose}}]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.
# 0 successful operations.
# 0 derived errors ignored.
# During handling of the above exception, another exception occurred:

# Xavier NXは実行できた
# Xavier NX 3m 11s


# 指定する個数を減らせば動く
# Example of style mixing (matches the corresponding video clip)
python3 run_generator.py style-mixing-example --network=gdrive:networks/stylegan2-ffhq-config-f.pkl \
  --row-seeds=85,100 --col-seeds=821 --truncation-psi=1.0
# Generating style-mixed images...
# Saving images...
# Saving image grid...
# dnnlib: Finished run_generator.style_mixing_example() in 5m 47s.
# Jetson Nano 5m 47s
# Xavier NX 2m 18s

jetson@jetson-desktop:~/stylegan2 $ python3 run_generator.py --help
2020-06-26 13:59:11.058536: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
usage: run_generator.py [-h] {generate-images,style-mixing-example} ...

StyleGAN2 generator.

Run 'python run_generator.py <subcommand> --help' for subcommand help.

positional arguments:
  {generate-images,style-mixing-example}
                        Sub-commands
    generate-images     Generate images
    style-mixing-example
                        Generate style mixing video

optional arguments:
  -h, --help            show this help message and exit

examples:

  # Generate ffhq uncurated images (matches paper Figure 12)
  python run_generator.py generate-images --network=gdrive:networks/stylegan2-ffhq-config-f.pkl --seeds=6600-6625 --truncation-psi=0.5

  # Generate ffhq curated images (matches paper Figure 11)
  python run_generator.py generate-images --network=gdrive:networks/stylegan2-ffhq-config-f.pkl --seeds=66,230,389,1518 --truncation-psi=1.0

  # Generate uncurated car images (matches paper Figure 12)
  python run_generator.py generate-images --network=gdrive:networks/stylegan2-car-config-f.pkl --seeds=6000-6025 --truncation-psi=0.5

  # Generate style mixing example (matches style mixing video clip)
  python run_generator.py style-mixing-example --network=gdrive:networks/stylegan2-ffhq-config-f.pkl --row-seeds=85,100,75,458,1500 --col-seeds=55,821,1789,293 --truncation-psi=1.0

jetson@jetson-desktop:~/stylegan2 $ python3 run_generator.py style-mixing-example --help
2020-06-26 13:59:39.619071: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
usage: run_generator.py style-mixing-example [-h] --network NETWORK_PKL
                                             --row-seeds ROW_SEEDS --col-seeds
                                             COL_SEEDS
                                             [--col-styles COL_STYLES]
                                             [--truncation-psi TRUNCATION_PSI]
                                             [--result-dir DIR]

optional arguments:
  -h, --help            show this help message and exit
  --network NETWORK_PKL
                        Network pickle filename
  --row-seeds ROW_SEEDS
                        Random seeds to use for image rows
  --col-seeds COL_SEEDS
                        Random seeds to use for image columns
  --col-styles COL_STYLES
                        Style layer range (default: 0-6)
  --truncation-psi TRUNCATION_PSI
                        Truncation psi (default: 0.5)
  --result-dir DIR      Root directory for run results (default: results)

2020-06-26 13:06:36.312246: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:950] ARM64 does not support NUMA - returning NUMA node zero
2020-06-26 13:06:36.312363: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1325] Created TensorFlow device (/device:GPU:0 with 1128 MB memory) -> physical GPU (device: 0, name: NVIDIA Tegra X1, pci bus id: 0000:00:00.0, compute capability: 5.3)
Loading... Failed!
Traceback (most recent call last):
  File "run_generator.py", line 168, in <module>
    main()
  File "run_generator.py", line 163, in main
    dnnlib.submit_run(sc, func_name_map[subcmd], **kwargs)
  File "/home/jetson/stylegan2/dnnlib/submission/submit.py", line 343, in submit_run
    return farm.submit(submit_config, host_run_dir)
  File "/home/jetson/stylegan2/dnnlib/submission/internal/local.py", line 22, in submit
    return run_wrapper(submit_config)
  File "/home/jetson/stylegan2/dnnlib/submission/submit.py", line 280, in run_wrapper
    run_func_obj(**submit_config.run_func_kwargs)
  File "/home/jetson/stylegan2/run_generator.py", line 21, in generate_images
    _G, _D, Gs = pretrained_networks.load_networks(network_pkl)
  File "/home/jetson/stylegan2/pretrained_networks.py", line 76, in load_networks
    G, D, Gs = pickle.load(stream, encoding='latin1')
  File "/home/jetson/stylegan2/dnnlib/tflib/network.py", line 297, in __setstate__
    self._init_graph()
  File "/home/jetson/stylegan2/dnnlib/tflib/network.py", line 154, in _init_graph
    out_expr = self._build_func(*self.input_templates, **build_kwargs)
  File "<string>", line 491, in G_synthesis_stylegan2
  File "<string>", line 455, in layer
  File "<string>", line 99, in modulated_conv2d_layer
  File "<string>", line 68, in apply_bias_act
  File "/home/jetson/stylegan2/dnnlib/tflib/ops/fused_bias_act.py", line 68, in fused_bias_act
    return impl_dict[impl](x=x, b=b, axis=axis, act=act, alpha=alpha, gain=gain)
  File "/home/jetson/stylegan2/dnnlib/tflib/ops/fused_bias_act.py", line 122, in _fused_bias_act_cuda
    cuda_kernel = _get_plugin().fused_bias_act
  File "/home/jetson/stylegan2/dnnlib/tflib/ops/fused_bias_act.py", line 16, in _get_plugin
    return custom_ops.get_plugin(os.path.splitext(__file__)[0] + '.cu')
  File "/home/jetson/stylegan2/dnnlib/tflib/custom_ops.py", line 156, in get_plugin
    plugin = tf.load_op_library(bin_file)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/load_library.py", line 61, in load_op_library
    lib_handle = py_tf.TF_LoadLibrary(library_filename)
tensorflow.python.framework.errors_impl.NotFoundError: /home/jetson/stylegan2/dnnlib/tflib/_cudacache/fused_bias_act_526f7217aed2933663fcf967d868580a.so: undefined symbol: _ZN10tensorflow12OpDefBuilder5InputESs

nano dnnlib/tflib/custom_ops.py
            compile_opts += ' --compiler-options \'-fPIC -D_GLIBCXX_USE_CXX11_ABI=0\''
　を
            compile_opts += ' --compiler-options \'-fPIC -D_GLIBCXX_USE_CXX11_ABI=1\''
　に変更する。

# -D_GLIBCXX_USE_CXX11_ABI=1

【2020年版】Jetson Nanoで StyleGANを動かして可愛い美少女のアニメ顔を大量生産する方法

【2020年版】NVIDIA Jetson Nano、Jetson Xavier NXの便利スクリプト

【2020年】Jetson Xavier NX 開発者キットが安かったので衝動買いした件、標準販売価格５万円が４万４千円！

【2020年版】NVIDIA Jetson用に最新の CMake 3.17.3をビルドしてインストールする方法

【2020年版】NVIDIA JetPack 4.4 DP Developer Previewで OpenPoseのビルドエラーの対策方法

Jetson Nano、Jetson Xavier NX 開発者キット TensorFlow on Jetson Platform

NVIDIA Jetson Nano 開発者キットを買ってみた。メモリ容量 4GB LPDDR4 RAM

【2020年版】NVIDIA Jetson Nanoの初期設定、最高速で動かす設定、空きメモリを増やす方法等

【2020年版】NVIDIA Jetson Nano JetPackのバージョン情報まとめ、JetPack 4.4は仕様変更の影響が大きい

【2020年版】NVIDIA Jetson Nano対応の FFmpegをビルドする手順

【2020年版】NVIDIA Jetson Nano対応の OpenPoseをビルドする手順

NVIDIA Jetson Nanoで OpenCV 3をビルドしてインストールする方法、NVCaffe等の OpenCV 4未対応を動かす

NVIDIA Jetson Nanoで Visual Studio Code Open Sourceをビルドして実行する

【2020年版】NVIDIA Jetson Nanoで TensorFlowの StyleGANを動かして、顔画像を生成

【2020年版】Jetson Nanoで StyleGANを動かして可愛い美少女のアニメ顔を大量生産する方法

【メモリ8GB】Raspberry Pi 4 Model B 8GBを KSYで最安値で購入。ベンチマークレビュー

【Vulkan】Raspberry Pi 4 Model Bで Vulkanドライバをビルドして 3Dグラフィックのデモを動かす

PIP機能付きの 4K対応の 4入力 1出力の HDMIセレクターを買ってみた、HDMI機器が複数有る場合に便利

EDID保持機能付きの 4K対応の 4入力 2出力の マトリックス切り替え HDMIセレクター、液晶画面 2台と使用で最強