{"id":144,"date":"2022-05-16T09:55:14","date_gmt":"2022-05-16T00:55:14","guid":{"rendered":"https:\/\/www.b64.pw\/blog\/?p=144"},"modified":"2022-05-20T08:19:25","modified_gmt":"2022-05-19T23:19:25","slug":"mmvc1-2-0-4-trainer%e3%82%92windows10%e3%81%a7%e5%8b%95%e3%81%8b%e3%81%99","status":"publish","type":"post","link":"https:\/\/www.b64.pw\/blog\/?p=144","title":{"rendered":"MMVC_Trainer 1.2.0.4\u3092Windows10\u3067\u52d5\u304b\u3059"},"content":{"rendered":"\n<p>\u6a5f\u68b0\u5b66\u7fd2\u3067\u30dc\u30a4\u30b9\u30c1\u30a7\u30f3\u30b8\u30e3\u30fc\u3092\u884c\u3046MMVC_Trainer\u3092Windows10\u3067\u52d5\u304b\u3057\u305f\u969b\u306e\u30e1\u30e2<\/p>\n\n\n\n<h2>\u6e96\u5099\u3059\u308b\u3082\u306e<\/h2>\n\n\n\n<p><a href=\"https:\/\/www.nvidia.co.jp\/Download\/index.aspx\">Nvidia\u306eGPU\u3068\u30c9\u30e9\u30a4\u30d0\u30fc<\/a><\/p>\n\n\n\n<p><a href=\"https:\/\/visualstudio.microsoft.com\/ja\/vs\/community\/\">VS2022 community<\/a><\/p>\n\n\n\n<p><a href=\"https:\/\/repo.anaconda.com\/archive\/Anaconda3-2021.05-Windows-x86_64.exe\">Anaconda3-2021.05<\/a> (python 3.8.8\u306e\u3082\u306e)<\/p>\n\n\n\n<p><a href=\"https:\/\/git-scm.com\/download\/win\">git<\/a> (source tree\u306b\u540c\u68b1\u3055\u308c\u3066\u3044\u308b\u3082\u306e\u3067\u3082\u53ef)<\/p>\n\n\n\n<h2>\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u624b\u9806<\/h2>\n\n\n\n<p><a href=\"https:\/\/github.com\/isletennos\/MMVC_Trainer\/tree\/v1.2.0.4\">MMVC\u306e\u30bd\u30fc\u30b9<\/a>\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>git clone -b v1.2.0.4 https:\/\/github.com\/isletennos\/MMVC_Trainer.git cd MMVC_Trainer \ncd MMVC_Trainer <\/code><\/pre>\n\n\n\n<p>\u203btrain_ms.py\u66f8\u304d\u63db\u3048 \u203b\u73fe\u5728\u306f\u30d0\u30b0\u30d5\u30a3\u30af\u30b9\u7248\u304c\u51fa\u3066\u3044\u308b\u306e\u3067\u66f8\u304d\u63db\u3048\u4e0d\u8981\u306e\u53ef\u80fd\u6027\u304c\u9ad8\u3044\u3067\u3059<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>\nfrom torch.nn.parallel import DistributedDataParallel as DDP\n\u2193\nfrom torch.nn.parallel import DataParallel as DDP\n\ndist.init_process_group(backend='nccl', init_method='env:\/\/', world_size=n_gpus, rank=rank)\n\u2193\ndist.init_process_group(backend='gloo', init_method='env:\/\/', world_size=n_gpus, rank=rank)\n\ntrain_loader = DataLoader(train_dataset, num_workers=os.cpu_count(), shuffle=False, pin_memory=True,\n\u2193\ntrain_loader = DataLoader(train_dataset, num_workers=4, shuffle=False, pin_memory=True,\n\neval_loader = DataLoader(eval_dataset, num_workers=os.cpu_count(), shuffle=False, pin_memory=True,\n\u2193\neval_loader = DataLoader(eval_dataset, num_workers=4, shuffle=False, pin_memory=True,\n<\/code><\/pre>\n\n\n\n<p>Anaconda3\u306e\u30b7\u30a7\u30eb\u304b\u3089\u4ee5\u4e0b\u5b9f\u884c<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>cd \\path\\to\\MMVC_Trainer\n\ncall \"C:\\Program Files\\Microsoft Visual Studio\\2022\\Community\\VC\\Auxiliary\\Build\\vcvarsall.bat\" x64\npip install pyopenjtalk\npip install Cython==0.29.21\npip install librosa==0.8.0\npip install numba\npip install numpy\npip install librosa==0.8.0\npip install phonemizer==2.2.1\npip install scipy==1.5.2\npip install tensorboard==2.3.0\npip install Unidecode==1.1.1\npip install retry\npip install tqdm\npip install pytz\npip install matplotlib\npip install torch==1.8.0+cu111 -f https:\/\/download.pytorch.org\/whl\/torch_stable.html\npip install torchvision==0.9.0\ncd monotonic_align\npython setup.py build_ext --inplace\ncd ..\n\n# \u5b66\u7fd2\u5b9f\u884c\npython create_dataset_jtalk.py -f train_config_zundamon -s 24000 -z True\npython train_ms.py -c configs\/train_config_zundamon.json -m 20220505_24000_zundamon -fg fine_model\/G_180000.pth -fd fine_model\/D_180000.pth\n\n# \u5b66\u7fd2\u518d\u958b\npython train_ms.py -c configs\/train_config_zundamon.json -m 20220505_24000_zundamon<\/code><\/pre>\n\n\n\n<h2>\u53c2\u8003<\/h2>\n\n\n\n<p>\u4ee5\u4e0b\u3092\u53c2\u8003\u306b\u3055\u305b\u3066\u3044\u305f\u3060\u304d\u307e\u3057\u305f\u3002\u3042\u308a\u304c\u3068\u3046\u3054\u3056\u3044\u307e\u3057\u305f\u3002<\/p>\n\n\n\n<p><a href=\"https:\/\/qiita.com\/yukarimazedofu\/items\/ed39d34632e412b2fc03\">https:\/\/qiita.com\/yukarimazedofu\/items\/ed39d34632e412b2fc03<\/a><\/p>\n\n\n\n<p><a href=\"https:\/\/note.com\/ydd\/n\/n00ea1394606e\">https:\/\/note.com\/ydd\/n\/n00ea1394606e<\/a><\/p>\n\n\n\n<p><a href=\"https:\/\/discord.com\/channels\/957230290487619665\/957230290487619668\">\u516c\u5f0fdiscord<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u6a5f\u68b0\u5b66\u7fd2\u3067\u30dc\u30a4\u30b9\u30c1\u30a7\u30f3\u30b8\u30e3\u30fc\u3092\u884c\u3046MMVC_Trainer\u3092 &#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/posts\/144"}],"collection":[{"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=144"}],"version-history":[{"count":9,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/posts\/144\/revisions"}],"predecessor-version":[{"id":163,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=\/wp\/v2\/posts\/144\/revisions\/163"}],"wp:attachment":[{"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=144"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=144"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.b64.pw\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=144"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}