Neural-Texture-Extraction-Distribution.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Neural-Texture-Extraction-Distribution.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyNQnoX3HB6M0z+s1lyCg3j5"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","source":["# Install the environment"],"metadata":{"id":"0mA3vsKXzIXW"}},{"cell_type":"code","source":["%%bash\n","MINICONDA_INSTALLER_SCRIPT=Miniconda3-4.5.4-Linux-x86_64.sh\n","MINICONDA_PREFIX=/usr/local\n","wget https://repo.continuum.io/miniconda/$MINICONDA_INSTALLER_SCRIPT\n","chmod +x $MINICONDA_INSTALLER_SCRIPT\n","./$MINICONDA_INSTALLER_SCRIPT -b -f -p $MINICONDA_PREFIX"],"metadata":{"id":"lW_b-Gzyn20Z"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["%%bash\n","conda install --channel defaults conda python=3.6 --yes\n","conda update --channel defaults --all --yes"],"metadata":{"id":"jvSl59Q2oVLo"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["!conda --version\n","!python --version"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"4fRrUMKYt8WO","executionInfo":{"status":"ok","timestamp":1651255264196,"user_tz":-480,"elapsed":16,"user":{"displayName":"俞睿任","userId":"18417540312902318536"}},"outputId":"875285f6-be69-47ae-eda8-32e1d43d8ca6"},"execution_count":3,"outputs":[{"output_type":"stream","name":"stdout","text":["conda 4.10.3\n","Python 3.6.13 :: Anaconda, Inc.\n"]}]},{"cell_type":"code","source":["!conda install -c pytorch pytorch=1.7.1 torchvision cudatoolkit=10.2 --yes"],"metadata":{"id":"qJqKxkvgteo0"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# Clone the Repo and Download resources for the Demos"],"metadata":{"id":"Xs64JriBzVso"}},{"cell_type":"code","source":["!git clone --recursive https://github.com/RenYurui/Neural-Texture-Extraction-Distribution.git"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7Kft95rqu2jQ","executionInfo":{"status":"ok","timestamp":1651255412517,"user_tz":-480,"elapsed":3918,"user":{"displayName":"俞睿任","userId":"18417540312902318536"}},"outputId":"0c5a31f8-28b0-43ab-ae44-c5b3fe9c83e7"},"execution_count":5,"outputs":[{"output_type":"stream","name":"stdout","text":["rm: cannot remove 'Neural-Texture-Extraction-Distribution': No such file or directory\n","Cloning into 'Neural-Texture-Extraction-Distribution'...\n","remote: Enumerating objects: 105, done.\u001b[K\n","remote: Counting objects: 100% (105/105), done.\u001b[K\n","remote: Compressing objects: 100% (83/83), done.\u001b[K\n","remote: Total 105 (delta 22), reused 97 (delta 17), pack-reused 0\u001b[K\n","Receiving objects: 100% (105/105), 68.40 KiB | 2.53 MiB/s, done.\n","Resolving deltas: 100% (22/22), done.\n","Submodule 'third_part/mmdetection' (https://github.com/open-mmlab/mmdetection.git) registered for path 'third_part/mmdetection'\n","Cloning into '/content/Neural-Texture-Extraction-Distribution/third_part/mmdetection'...\n","remote: Enumerating objects: 24305, done.        \n","remote: Total 24305 (delta 0), reused 0 (delta 0), pack-reused 24305        \n","Receiving objects: 100% (24305/24305), 37.47 MiB | 27.00 MiB/s, done.\n","Resolving deltas: 100% (17013/17013), done.\n","Submodule path 'third_part/mmdetection': checked out '0f33c08d8d46eba8165715a0995841a975badfd4'\n"]}]},{"cell_type":"code","source":["%%bash\n","cd Neural-Texture-Extraction-Distribution\n","pip install -r requirements.txt"],"metadata":{"id":"pK6x1ldWqBYt"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["%%bash\n","ls\n","cd Neural-Texture-Extraction-Distribution/scripts\n","chmod +x download_demos.sh\n","./download_demos.sh"],"metadata":{"id":"uGQnU8Njr1C5"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# Pose Transfer Demo"],"metadata":{"id":"nev0uQ5ozoIg"}},{"cell_type":"code","source":["%%bash\n","cd Neural-Texture-Extraction-Distribution\n","PATH_TO_OUTPUT=./demo_results # save results in ./Neural-Texture-Extraction-Distribution/demo_results\n","python demo.py \\\n","--config ./config/fashion_512.yaml \\\n","--which_iter 495400 \\\n","--name fashion_512 \\\n","--file_pairs ./txt_files/demo.txt \\\n","--input_dir ./demo_images \\\n","--output_dir $PATH_TO_OUTPUT"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"BOiPhcNCyd6I","executionInfo":{"status":"ok","timestamp":1651255554359,"user_tz":-480,"elapsed":106589,"user":{"displayName":"俞睿任","userId":"18417540312902318536"}},"outputId":"541cc82c-eac4-4dc4-9aaf-1cc3a826fa36"},"execution_count":8,"outputs":[{"output_type":"stream","name":"stdout","text":["net [net_G] parameter count: 42,612,402\n","net [net_D] parameter count: 26,885,697\n","Setup trainer.\n","Perceptual loss:\n","\tMode: vgg19\n","GAN mode: style_gan2\n","Perceptual loss:\n","\tMode: vgg19\n","Loss perceptual           Weight 1\n","Loss attn_rec             Weight 15\n","Loss gan                  Weight 1.5\n","Loss face                 Weight 1\n","load [net_G_ema] from result/fashion_512/epoch_00200_iteration_000495400_checkpoint.pt\n","Done with loading the checkpoint.\n"]},{"output_type":"stream","name":"stderr","text":["Downloading: \"https://download.pytorch.org/models/vgg19-dcbb9e9d.pth\" to /root/.cache/torch/hub/checkpoints/vgg19-dcbb9e9d.pth\n","\r  0%|          | 0.00/548M [00:00<?, ?B/s]\r  1%|          | 5.39M/548M [00:00<00:10, 56.5MB/s]\r  2%|▏         | 11.7M/548M [00:00<00:09, 62.1MB/s]\r  3%|▎         | 18.8M/548M [00:00<00:08, 67.8MB/s]\r  5%|▍         | 26.4M/548M [00:00<00:07, 72.4MB/s]\r  6%|▌         | 33.3M/548M [00:00<00:08, 66.2MB/s]\r  7%|▋         | 40.7M/548M [00:00<00:07, 69.8MB/s]\r  9%|▊         | 47.8M/548M [00:00<00:07, 71.1MB/s]\r 10%|█         | 55.2M/548M [00:00<00:07, 73.4MB/s]\r 11%|█▏        | 62.3M/548M [00:00<00:06, 72.8MB/s]\r 13%|█▎        | 69.3M/548M [00:01<00:07, 70.0MB/s]\r 14%|█▍        | 76.1M/548M [00:01<00:07, 67.0MB/s]\r 15%|█▌        | 82.6M/548M [00:01<00:07, 63.9MB/s]\r 16%|█▌        | 88.8M/548M [00:01<00:07, 64.3MB/s]\r 17%|█▋        | 95.0M/548M [00:01<00:07, 63.7MB/s]\r 19%|█▊        | 102M/548M [00:01<00:07, 65.0MB/s] \r 20%|█▉        | 108M/548M [00:01<00:07, 65.0MB/s]\r 21%|██        | 114M/548M [00:01<00:07, 64.5MB/s]\r 22%|██▏       | 121M/548M [00:01<00:06, 67.4MB/s]\r 23%|██▎       | 128M/548M [00:02<00:06, 65.2MB/s]\r 24%|██▍       | 134M/548M [00:02<00:07, 59.4MB/s]\r 25%|██▌       | 140M/548M [00:02<00:07, 55.9MB/s]\r 27%|██▋       | 146M/548M [00:02<00:07, 58.5MB/s]\r 28%|██▊       | 152M/548M [00:03<00:22, 18.7MB/s]\r 29%|██▊       | 157M/548M [00:03<00:17, 22.8MB/s]\r 30%|██▉       | 164M/548M [00:03<00:13, 30.2MB/s]\r 31%|███       | 170M/548M [00:03<00:11, 35.3MB/s]\r 32%|███▏      | 177M/548M [00:03<00:09, 42.6MB/s]\r 33%|███▎      | 183M/548M [00:03<00:08, 44.2MB/s]\r 35%|███▍      | 189M/548M [00:03<00:07, 48.8MB/s]\r 36%|███▌      | 196M/548M [00:03<00:06, 53.3MB/s]\r 37%|███▋      | 203M/548M [00:04<00:06, 58.5MB/s]\r 38%|███▊      | 209M/548M [00:04<00:06, 58.1MB/s]\r 39%|███▉      | 215M/548M [00:04<00:05, 59.8MB/s]\r 41%|████      | 222M/548M [00:04<00:05, 64.2MB/s]\r 42%|████▏     | 229M/548M [00:04<00:05, 58.2MB/s]\r 43%|████▎     | 236M/548M [00:04<00:05, 61.9MB/s]\r 45%|████▍     | 244M/548M [00:04<00:04, 66.6MB/s]\r 46%|████▌     | 251M/548M [00:04<00:04, 67.8MB/s]\r 47%|████▋     | 258M/548M [00:04<00:04, 69.4MB/s]\r 48%|████▊     | 265M/548M [00:05<00:04, 72.5MB/s]\r 50%|████▉     | 273M/548M [00:05<00:03, 73.3MB/s]\r 51%|█████     | 280M/548M [00:05<00:04, 68.1MB/s]\r 52%|█████▏    | 287M/548M [00:05<00:03, 70.3MB/s]\r 54%|█████▎    | 294M/548M [00:05<00:04, 66.5MB/s]\r 55%|█████▍    | 300M/548M [00:05<00:03, 67.3MB/s]\r 56%|█████▌    | 307M/548M [00:05<00:03, 66.8MB/s]\r 57%|█████▋    | 314M/548M [00:05<00:03, 67.9MB/s]\r 58%|█████▊    | 321M/548M [00:05<00:03, 69.1MB/s]\r 60%|█████▉    | 327M/548M [00:05<00:03, 69.4MB/s]\r 61%|██████    | 334M/548M [00:06<00:03, 67.2MB/s]\r 62%|██████▏   | 340M/548M [00:06<00:03, 57.6MB/s]\r 63%|██████▎   | 347M/548M [00:06<00:03, 59.7MB/s]\r 64%|██████▍   | 352M/548M [00:06<00:03, 57.6MB/s]\r 65%|██████▌   | 358M/548M [00:06<00:03, 56.6MB/s]\r 67%|██████▋   | 365M/548M [00:06<00:03, 59.4MB/s]\r 68%|██████▊   | 370M/548M [00:06<00:03, 57.0MB/s]\r 69%|██████▉   | 379M/548M [00:06<00:02, 65.3MB/s]\r 70%|███████   | 385M/548M [00:07<00:02, 60.5MB/s]\r 71%|███████▏  | 391M/548M [00:07<00:02, 61.2MB/s]\r 72%|███████▏  | 397M/548M [00:07<00:02, 59.7MB/s]\r 74%|███████▎  | 404M/548M [00:07<00:02, 62.0MB/s]\r 75%|███████▍  | 410M/548M [00:07<00:02, 65.0MB/s]\r 76%|███████▋  | 418M/548M [00:07<00:01, 70.3MB/s]\r 78%|███████▊  | 425M/548M [00:07<00:01, 71.2MB/s]\r 79%|███████▉  | 433M/548M [00:07<00:01, 74.8MB/s]\r 80%|████████  | 441M/548M [00:07<00:01, 72.1MB/s]\r 82%|████████▏ | 448M/548M [00:07<00:01, 70.9MB/s]\r 83%|████████▎ | 454M/548M [00:08<00:01, 69.9MB/s]\r 84%|████████▍ | 462M/548M [00:08<00:01, 71.2MB/s]\r 86%|████████▌ | 470M/548M [00:08<00:01, 74.9MB/s]\r 87%|████████▋ | 477M/548M [00:08<00:00, 75.7MB/s]\r 88%|████████▊ | 485M/548M [00:08<00:00, 69.0MB/s]\r 90%|████████▉ | 492M/548M [00:08<00:00, 70.1MB/s]\r 91%|█████████ | 498M/548M [00:08<00:00, 69.6MB/s]\r 92%|█████████▏| 505M/548M [00:08<00:00, 67.9MB/s]\r 93%|█████████▎| 512M/548M [00:08<00:00, 66.7MB/s]\r 95%|█████████▍| 518M/548M [00:09<00:00, 67.7MB/s]\r 96%|█████████▌| 526M/548M [00:09<00:00, 70.3MB/s]\r 97%|█████████▋| 532M/548M [00:09<00:00, 66.0MB/s]\r 98%|█████████▊| 539M/548M [00:09<00:00, 62.5MB/s]\r 99%|█████████▉| 545M/548M [00:09<00:00, 63.1MB/s]\r100%|██████████| 548M/548M [00:09<00:00, 60.2MB/s]\n","\r0it [00:00, ?it/s]\r1it [00:01,  1.41s/it]\r2it [00:02,  1.05it/s]\r3it [00:02,  1.23it/s]\r4it [00:03,  1.34it/s]\r5it [00:04,  1.35it/s]\r6it [00:04,  1.37it/s]\r7it [00:05,  1.37it/s]\r8it [00:06,  1.38it/s]\r9it [00:06,  1.40it/s]\r10it [00:07,  1.41it/s]\r11it [00:08,  1.42it/s]\r12it [00:09,  1.42it/s]\r13it [00:09,  1.38it/s]\r14it [00:10,  1.36it/s]\r15it [00:11,  1.34it/s]\r16it [00:12,  1.33it/s]\r17it [00:12,  1.37it/s]\r18it [00:13,  1.40it/s]\r19it [00:14,  1.42it/s]\r20it [00:14,  1.43it/s]\r21it [00:15,  1.43it/s]\r22it [00:16,  1.43it/s]\r23it [00:16,  1.42it/s]\r24it [00:17,  1.42it/s]\r25it [00:18,  1.41it/s]\r26it [00:19,  1.42it/s]\r27it [00:19,  1.41it/s]\r28it [00:20,  1.40it/s]\r29it [00:21,  1.38it/s]\r30it [00:21,  1.37it/s]\r31it [00:22,  1.37it/s]\r32it [00:23,  1.36it/s]\r32it [00:23,  1.37it/s]\n"]}]},{"cell_type":"markdown","source":["# Appearance Control"],"metadata":{"id":"lVUvurwVethu"}},{"cell_type":"code","source":["%%bash\n","pip install mmcv==0.5.1\n","pip install pycocotools==2.0.4\n","pip install ipykernel"],"metadata":{"id":"Df2dVJiHKkl3"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["%%bash\n","cd Neural-Texture-Extraction-Distribution/scripts\n","chmod +x insert_mmfashion2mmdetection.sh\n","./insert_mmfashion2mmdetection.sh\n","cd ../third_part/mmdetection\n","pip install -v -e ."],"metadata":{"id":"lD0cljkFLKU3"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["%%bash\n","cd Neural-Texture-Extraction-Distribution/\n","python appearance_control.py \\\n","--config ./config/fashion_512.yaml \\\n","--name fashion_512 \\\n","--which_iter 495400 \\\n","--input_dir ./demo_images \\\n","--file_pairs ./txt_files/appearance_control.txt"],"metadata":{"id":"3U4tzY7vNkmp"},"execution_count":null,"outputs":[]}]}