feat: training

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text

app.py

@@ -1,10 +1,14 @@
+from random import shuffle
 import gradio as gr
 import zipfile
 import os
 import tempfile
 import shutil
-from infer.modules.train.preprocess import PreProcess, preprocess_trainset
+from glob import glob
+from infer.modules.train.preprocess import PreProcess
 from infer.modules.train.extract.extract_f0_rmvpe import FeatureInput
+from infer.modules.train.train import train
+from infer.lib.train.process_ckpt import extract_small_model
 from zero import zero
 
 
@@ -44,11 +48,6 @@ def preprocess(zip_file: str) -> str:
     return temp_dir, f"Preprocessed {len(audio_files)} audio files.\n{log}"
 
 
-def download_expdir(exp_dir: str) -> str:
-    shutil.make_archive(exp_dir, "zip", exp_dir)
-    return f"{exp_dir}.zip"
-
-
 @zero(duration=120)
 def extract_features(exp_dir: str) -> str:
     err = None
@@ -67,6 +66,108 @@ def extract_features(exp_dir: str) -> str:
     return log
 
 
+def write_filelist(exp_dir: str) -> None:
+    if_f0_3 = True
+    spk_id5 = 0
+    gt_wavs_dir = "%s/0_gt_wavs" % (exp_dir)
+    feature_dir = "%s/3_feature768" % (exp_dir)
+
+    if if_f0_3:
+        f0_dir = "%s/2a_f0" % (exp_dir)
+        f0nsf_dir = "%s/2b-f0nsf" % (exp_dir)
+        names = (
+            set([name.split(".")[0] for name in os.listdir(gt_wavs_dir)])
+            & set([name.split(".")[0] for name in os.listdir(feature_dir)])
+            & set([name.split(".")[0] for name in os.listdir(f0_dir)])
+            & set([name.split(".")[0] for name in os.listdir(f0nsf_dir)])
+        )
+    else:
+        names = set([name.split(".")[0] for name in os.listdir(gt_wavs_dir)]) & set(
+            [name.split(".")[0] for name in os.listdir(feature_dir)]
+        )
+    opt = []
+    for name in names:
+        if if_f0_3:
+            opt.append(
+                "%s/%s.wav|%s/%s.npy|%s/%s.wav.npy|%s/%s.wav.npy|%s"
+                % (
+                    gt_wavs_dir.replace("\\", "\\\\"),
+                    name,
+                    feature_dir.replace("\\", "\\\\"),
+                    name,
+                    f0_dir.replace("\\", "\\\\"),
+                    name,
+                    f0nsf_dir.replace("\\", "\\\\"),
+                    name,
+                    spk_id5,
+                )
+            )
+        else:
+            opt.append(
+                "%s/%s.wav|%s/%s.npy|%s"
+                % (
+                    gt_wavs_dir.replace("\\", "\\\\"),
+                    name,
+                    feature_dir.replace("\\", "\\\\"),
+                    name,
+                    spk_id5,
+                )
+            )
+    fea_dim = 768
+
+    now_dir = os.getcwd()
+    sr2 = "40k"
+    if if_f0_3:
+        for _ in range(2):
+            opt.append(
+                "%s/logs/mute/0_gt_wavs/mute%s.wav|%s/logs/mute/3_feature%s/mute.npy|%s/logs/mute/2a_f0/mute.wav.npy|%s/logs/mute/2b-f0nsf/mute.wav.npy|%s"
+                % (now_dir, sr2, now_dir, fea_dim, now_dir, now_dir, spk_id5)
+            )
+    else:
+        for _ in range(2):
+            opt.append(
+                "%s/logs/mute/0_gt_wavs/mute%s.wav|%s/logs/mute/3_feature%s/mute.npy|%s"
+                % (now_dir, sr2, now_dir, fea_dim, spk_id5)
+            )
+    shuffle(opt)
+    with open("%s/filelist.txt" % exp_dir, "w") as f:
+        f.write("\n".join(opt))
+
+
+@zero(duration=300)
+def train_model(exp_dir: str) -> str:
+    shutil.copy("config.json", exp_dir)
+    write_filelist(exp_dir)
+    train(exp_dir)
+
+    models = glob(f"{exp_dir}/G_*.pth")
+    if not models:
+        raise gr.Error("No model found")
+
+    latest_model = max(models, key=os.path.getctime)
+    return latest_model
+
+
+def download_weight(exp_dir: str) -> str:
+    models = glob(f"{exp_dir}/G_*.pth")
+    if not models:
+        raise gr.Error("No model found")
+
+    latest_model = max(models, key=os.path.getctime)
+
+    name = os.path.basename(exp_dir)
+    extract_small_model(
+        latest_model, name, "40k", True, "Model trained by ZeroGPU.", "v2"
+    )
+
+    return "assets/weights/%s.pth" % name
+
+
+def download_expdir(exp_dir: str) -> str:
+    shutil.make_archive(exp_dir, "zip", exp_dir)
+    return f"{exp_dir}.zip"
+
+
 with gr.Blocks() as app:
     with gr.Row():
         with gr.Column():
@@ -89,6 +190,20 @@ with gr.Blocks() as app:
                 label="Feature extraction output", lines=5
             )
 
+    with gr.Row():
+        with gr.Column():
+            train_btn = gr.Button(value="Train", variant="primary")
+        with gr.Column():
+            latest_model = gr.File(label="Latest model")
+
+    with gr.Row():
+        with gr.Column():
+            download_weight_btn = gr.Button(
+                value="Download latest model", variant="primary"
+            )
+        with gr.Column():
+            download_weight_output = gr.File(label="Download latest model")
+
     with gr.Row():
         with gr.Column():
             download_expdir_btn = gr.Button(
@@ -109,6 +224,18 @@ with gr.Blocks() as app:
         outputs=[extract_features_output],
     )
 
+    train_btn.click(
+        fn=train_model,
+        inputs=[exp_dir],
+        outputs=[latest_model],
+    )
+
+    download_weight_btn.click(
+        fn=download_weight,
+        inputs=[exp_dir],
+        outputs=[download_weight_output],
+    )
+
     download_expdir_btn.click(
         fn=download_expdir,
         inputs=[exp_dir],

config.json

@@ -0,0 +1,79 @@
+{
+    "data": {
+        "filter_length": 2048,
+        "hop_length": 400,
+        "max_wav_value": 32768.0,
+        "mel_fmax": null,
+        "mel_fmin": 0.0,
+        "n_mel_channels": 125,
+        "sampling_rate": 40000,
+        "win_length": 2048
+    },
+    "model": {
+        "filter_channels": 768,
+        "gin_channels": 256,
+        "hidden_channels": 192,
+        "inter_channels": 192,
+        "kernel_size": 3,
+        "n_heads": 2,
+        "n_layers": 6,
+        "p_dropout": 0,
+        "resblock": "1",
+        "resblock_dilation_sizes": [
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ]
+        ],
+        "resblock_kernel_sizes": [
+            3,
+            7,
+            11
+        ],
+        "spk_embed_dim": 109,
+        "upsample_initial_channel": 512,
+        "upsample_kernel_sizes": [
+            16,
+            16,
+            4,
+            4
+        ],
+        "upsample_rates": [
+            10,
+            10,
+            2,
+            2
+        ],
+        "use_spectral_norm": false
+    },
+    "train": {
+        "batch_size": 4,
+        "betas": [
+            0.8,
+            0.99
+        ],
+        "c_kl": 1.0,
+        "c_mel": 45,
+        "epochs": 20000,
+        "eps": 1e-09,
+        "fp16_run": false,
+        "init_lr_ratio": 1,
+        "learning_rate": 0.0001,
+        "log_interval": 200,
+        "lr_decay": 0.999875,
+        "seed": 1234,
+        "segment_size": 12800,
+        "warmup_epochs": 0
+    }
+}

infer-web.py

@@ -0,0 +1,1265 @@
+import os
+import sys
+from dotenv import load_dotenv
+
+now_dir = os.getcwd()
+sys.path.append(now_dir)
+load_dotenv()
+from infer.modules.vc.modules import VC
+from infer.modules.uvr5.modules import uvr
+from infer.lib.train.process_ckpt import (
+    change_info,
+    extract_small_model,
+    merge,
+    show_info,
+)
+from i18n.i18n import I18nAuto
+from configs.config import Config
+from sklearn.cluster import MiniBatchKMeans
+import torch, platform
+import numpy as np
+import gradio as gr
+import faiss
+import fairseq
+import pathlib
+import json
+from time import sleep
+from subprocess import Popen
+from random import shuffle
+import warnings
+import traceback
+import threading
+import shutil
+import logging
+
+
+logging.getLogger("numba").setLevel(logging.WARNING)
+logging.getLogger("httpx").setLevel(logging.WARNING)
+
+logger = logging.getLogger(__name__)
+
+tmp = os.path.join(now_dir, "TEMP")
+shutil.rmtree(tmp, ignore_errors=True)
+shutil.rmtree("%s/runtime/Lib/site-packages/infer_pack" % (now_dir), ignore_errors=True)
+shutil.rmtree("%s/runtime/Lib/site-packages/uvr5_pack" % (now_dir), ignore_errors=True)
+os.makedirs(tmp, exist_ok=True)
+os.makedirs(os.path.join(now_dir, "logs"), exist_ok=True)
+os.makedirs(os.path.join(now_dir, "assets/weights"), exist_ok=True)
+os.environ["TEMP"] = tmp
+warnings.filterwarnings("ignore")
+torch.manual_seed(114514)
+
+
+config = Config()
+vc = VC(config)
+
+
+if config.dml == True:
+
+    def forward_dml(ctx, x, scale):
+        ctx.scale = scale
+        res = x.clone().detach()
+        return res
+
+    fairseq.modules.grad_multiply.GradMultiply.forward = forward_dml
+i18n = I18nAuto()
+logger.info(i18n)
+# 判断是否有能用来训练和加速推理的N卡
+ngpu = torch.cuda.device_count()
+gpu_infos = []
+mem = []
+if_gpu_ok = False
+
+if torch.cuda.is_available() or ngpu != 0:
+    for i in range(ngpu):
+        gpu_name = torch.cuda.get_device_name(i)
+        if any(
+            value in gpu_name.upper()
+            for value in [
+                "10",
+                "16",
+                "20",
+                "30",
+                "40",
+                "A2",
+                "A3",
+                "A4",
+                "P4",
+                "A50",
+                "500",
+                "A60",
+                "70",
+                "80",
+                "90",
+                "M4",
+                "T4",
+                "TITAN",
+                "4060",
+                "L",
+                "6000",
+            ]
+        ):
+            # A10#A100#V100#A40#P40#M40#K80#A4500
+            if_gpu_ok = True  # 至少有一张能用的N卡
+            gpu_infos.append("%s\t%s" % (i, gpu_name))
+            mem.append(
+                int(
+                    torch.cuda.get_device_properties(i).total_memory
+                    / 1024
+                    / 1024
+                    / 1024
+                    + 0.4
+                )
+            )
+if if_gpu_ok and len(gpu_infos) > 0:
+    gpu_info = "\n".join(gpu_infos)
+    default_batch_size = min(mem) // 2
+else:
+    gpu_info = i18n("很遗憾您这没有能用的显卡来支持您训练")
+    default_batch_size = 1
+gpus = "-".join([i[0] for i in gpu_infos])
+
+
+class ToolButton(gr.Button, gr.components.FormComponent):
+    """Small button with single emoji as text, fits inside gradio forms"""
+
+    def __init__(self, **kwargs):
+        super().__init__(variant="tool", **kwargs)
+
+    def get_block_name(self):
+        return "button"
+
+
+weight_root = os.getenv("weight_root")
+weight_uvr5_root = os.getenv("weight_uvr5_root")
+index_root = os.getenv("index_root")
+outside_index_root = os.getenv("outside_index_root")
+
+names = []
+for name in os.listdir(weight_root):
+    if name.endswith(".pth"):
+        names.append(name)
+index_paths = []
+
+
+def lookup_indices(index_root):
+    global index_paths
+    for root, dirs, files in os.walk(index_root, topdown=False):
+        for name in files:
+            if name.endswith(".index") and "trained" not in name:
+                index_paths.append("%s/%s" % (root, name))
+
+
+lookup_indices(index_root)
+lookup_indices(outside_index_root)
+uvr5_names = []
+for name in os.listdir(weight_uvr5_root):
+    if name.endswith(".pth") or "onnx" in name:
+        uvr5_names.append(name.replace(".pth", ""))
+
+
+def change_choices():
+    names = []
+    for name in os.listdir(weight_root):
+        if name.endswith(".pth"):
+            names.append(name)
+    index_paths = []
+    for root, dirs, files in os.walk(index_root, topdown=False):
+        for name in files:
+            if name.endswith(".index") and "trained" not in name:
+                index_paths.append("%s/%s" % (root, name))
+    return {"choices": sorted(names), "__type__": "update"}, {
+        "choices": sorted(index_paths),
+        "__type__": "update",
+    }
+
+
+def clean():
+    return {"value": "", "__type__": "update"}
+
+
+def export_onnx(ModelPath, ExportedPath):
+    from infer.modules.onnx.export import export_onnx as eo
+
+    eo(ModelPath, ExportedPath)
+
+
+sr_dict = {
+    "32k": 32000,
+    "40k": 40000,
+    "48k": 48000,
+}
+
+
+def if_done(done, p):
+    while 1:
+        if p.poll() is None:
+            sleep(0.5)
+        else:
+            break
+    done[0] = True
+
+
+def if_done_multi(done, ps):
+    while 1:
+        # poll==None代表进程未结束
+        # 只要有一个进程未结束都不停
+        flag = 1
+        for p in ps:
+            if p.poll() is None:
+                flag = 0
+                sleep(0.5)
+                break
+        if flag == 1:
+            break
+    done[0] = True
+
+
+def preprocess_dataset(trainset_dir, exp_dir, sr, n_p):
+    sr = sr_dict[sr]
+    os.makedirs("%s/logs/%s" % (now_dir, exp_dir), exist_ok=True)
+    f = open("%s/logs/%s/preprocess.log" % (now_dir, exp_dir), "w")
+    f.close()
+    cmd = '"%s" infer/modules/train/preprocess.py "%s" %s %s "%s/logs/%s" %s %.1f' % (
+        config.python_cmd,
+        trainset_dir,
+        sr,
+        n_p,
+        now_dir,
+        exp_dir,
+        config.noparallel,
+        config.preprocess_per,
+    )
+    logger.info("Execute: " + cmd)
+    # , stdin=PIPE, stdout=PIPE,stderr=PIPE,cwd=now_dir
+    p = Popen(cmd, shell=True)
+    # 煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+    done = [False]
+    threading.Thread(
+        target=if_done,
+        args=(
+            done,
+            p,
+        ),
+    ).start()
+    while 1:
+        with open("%s/logs/%s/preprocess.log" % (now_dir, exp_dir), "r") as f:
+            yield (f.read())
+        sleep(1)
+        if done[0]:
+            break
+    with open("%s/logs/%s/preprocess.log" % (now_dir, exp_dir), "r") as f:
+        log = f.read()
+    logger.info(log)
+    yield log
+
+
+# but2.click(extract_f0,[gpus6,np7,f0method8,if_f0_3,trainset_dir4],[info2])
+def extract_f0_feature(gpus, n_p, f0method, if_f0, exp_dir, version19, gpus_rmvpe):
+    gpus = gpus.split("-")
+    os.makedirs("%s/logs/%s" % (now_dir, exp_dir), exist_ok=True)
+    f = open("%s/logs/%s/extract_f0_feature.log" % (now_dir, exp_dir), "w")
+    f.close()
+    if if_f0:
+        if f0method != "rmvpe_gpu":
+            cmd = (
+                '"%s" infer/modules/train/extract/extract_f0_print.py "%s/logs/%s" %s %s'
+                % (
+                    config.python_cmd,
+                    now_dir,
+                    exp_dir,
+                    n_p,
+                    f0method,
+                )
+            )
+            logger.info("Execute: " + cmd)
+            p = Popen(
+                cmd, shell=True, cwd=now_dir
+            )  # , stdin=PIPE, stdout=PIPE,stderr=PIPE
+            # 煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+            done = [False]
+            threading.Thread(
+                target=if_done,
+                args=(
+                    done,
+                    p,
+                ),
+            ).start()
+        else:
+            if gpus_rmvpe != "-":
+                gpus_rmvpe = gpus_rmvpe.split("-")
+                leng = len(gpus_rmvpe)
+                ps = []
+                for idx, n_g in enumerate(gpus_rmvpe):
+                    cmd = (
+                        '"%s" infer/modules/train/extract/extract_f0_rmvpe.py %s %s %s "%s/logs/%s" %s '
+                        % (
+                            config.python_cmd,
+                            leng,
+                            idx,
+                            n_g,
+                            now_dir,
+                            exp_dir,
+                            config.is_half,
+                        )
+                    )
+                    logger.info("Execute: " + cmd)
+                    p = Popen(
+                        cmd, shell=True, cwd=now_dir
+                    )  # , shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE, cwd=now_dir
+                    ps.append(p)
+                # 煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+                done = [False]
+                threading.Thread(
+                    target=if_done_multi,  #
+                    args=(
+                        done,
+                        ps,
+                    ),
+                ).start()
+            else:
+                cmd = (
+                    config.python_cmd
+                    + ' infer/modules/train/extract/extract_f0_rmvpe_dml.py "%s/logs/%s" '
+                    % (
+                        now_dir,
+                        exp_dir,
+                    )
+                )
+                logger.info("Execute: " + cmd)
+                p = Popen(
+                    cmd, shell=True, cwd=now_dir
+                )  # , shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE, cwd=now_dir
+                p.wait()
+                done = [True]
+        while 1:
+            with open(
+                "%s/logs/%s/extract_f0_feature.log" % (now_dir, exp_dir), "r"
+            ) as f:
+                yield (f.read())
+            sleep(1)
+            if done[0]:
+                break
+        with open("%s/logs/%s/extract_f0_feature.log" % (now_dir, exp_dir), "r") as f:
+            log = f.read()
+        logger.info(log)
+        yield log
+    # 对不同part分别开多进程
+    """
+    n_part=int(sys.argv[1])
+    i_part=int(sys.argv[2])
+    i_gpu=sys.argv[3]
+    exp_dir=sys.argv[4]
+    os.environ["CUDA_VISIBLE_DEVICES"]=str(i_gpu)
+    """
+    leng = len(gpus)
+    ps = []
+    for idx, n_g in enumerate(gpus):
+        cmd = (
+            '"%s" infer/modules/train/extract_feature_print.py %s %s %s %s "%s/logs/%s" %s %s'
+            % (
+                config.python_cmd,
+                config.device,
+                leng,
+                idx,
+                n_g,
+                now_dir,
+                exp_dir,
+                version19,
+                config.is_half,
+            )
+        )
+        logger.info("Execute: " + cmd)
+        p = Popen(
+            cmd, shell=True, cwd=now_dir
+        )  # , shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE, cwd=now_dir
+        ps.append(p)
+    # 煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+    done = [False]
+    threading.Thread(
+        target=if_done_multi,
+        args=(
+            done,
+            ps,
+        ),
+    ).start()
+    while 1:
+        with open("%s/logs/%s/extract_f0_feature.log" % (now_dir, exp_dir), "r") as f:
+            yield (f.read())
+        sleep(1)
+        if done[0]:
+            break
+    with open("%s/logs/%s/extract_f0_feature.log" % (now_dir, exp_dir), "r") as f:
+        log = f.read()
+    logger.info(log)
+    yield log
+
+
+def get_pretrained_models(path_str, f0_str, sr2):
+    if_pretrained_generator_exist = os.access(
+        "assets/pretrained%s/%sG%s.pth" % (path_str, f0_str, sr2), os.F_OK
+    )
+    if_pretrained_discriminator_exist = os.access(
+        "assets/pretrained%s/%sD%s.pth" % (path_str, f0_str, sr2), os.F_OK
+    )
+    if not if_pretrained_generator_exist:
+        logger.warning(
+            "assets/pretrained%s/%sG%s.pth not exist, will not use pretrained model",
+            path_str,
+            f0_str,
+            sr2,
+        )
+    if not if_pretrained_discriminator_exist:
+        logger.warning(
+            "assets/pretrained%s/%sD%s.pth not exist, will not use pretrained model",
+            path_str,
+            f0_str,
+            sr2,
+        )
+    return (
+        (
+            "assets/pretrained%s/%sG%s.pth" % (path_str, f0_str, sr2)
+            if if_pretrained_generator_exist
+            else ""
+        ),
+        (
+            "assets/pretrained%s/%sD%s.pth" % (path_str, f0_str, sr2)
+            if if_pretrained_discriminator_exist
+            else ""
+        ),
+    )
+
+
+def change_sr2(sr2, if_f0_3, version19):
+    path_str = "" if version19 == "v1" else "_v2"
+    f0_str = "f0" if if_f0_3 else ""
+    return get_pretrained_models(path_str, f0_str, sr2)
+
+
+def change_version19(sr2, if_f0_3, version19):
+    path_str = "" if version19 == "v1" else "_v2"
+    if sr2 == "32k" and version19 == "v1":
+        sr2 = "40k"
+    to_return_sr2 = (
+        {"choices": ["40k", "48k"], "__type__": "update", "value": sr2}
+        if version19 == "v1"
+        else {"choices": ["40k", "48k", "32k"], "__type__": "update", "value": sr2}
+    )
+    f0_str = "f0" if if_f0_3 else ""
+    return (
+        *get_pretrained_models(path_str, f0_str, sr2),
+        to_return_sr2,
+    )
+
+
+def change_f0(if_f0_3, sr2, version19):  # f0method8,pretrained_G14,pretrained_D15
+    path_str = "" if version19 == "v1" else "_v2"
+    return (
+        {"visible": if_f0_3, "__type__": "update"},
+        {"visible": if_f0_3, "__type__": "update"},
+        *get_pretrained_models(path_str, "f0" if if_f0_3 == True else "", sr2),
+    )
+
+
+# but3.click(click_train,[exp_dir1,sr2,if_f0_3,save_epoch10,total_epoch11,batch_size12,if_save_latest13,pretrained_G14,pretrained_D15,gpus16])
+def click_train(
+    exp_dir1,
+    sr2,
+    if_f0_3,
+    spk_id5,
+    save_epoch10,
+    total_epoch11,
+    batch_size12,
+    if_save_latest13,
+    pretrained_G14,
+    pretrained_D15,
+    gpus16,
+    if_cache_gpu17,
+    if_save_every_weights18,
+    version19,
+):
+    # 生成filelist
+    exp_dir = "%s/logs/%s" % (now_dir, exp_dir1)
+    os.makedirs(exp_dir, exist_ok=True)
+    gt_wavs_dir = "%s/0_gt_wavs" % (exp_dir)
+    feature_dir = (
+        "%s/3_feature256" % (exp_dir)
+        if version19 == "v1"
+        else "%s/3_feature768" % (exp_dir)
+    )
+    if if_f0_3:
+        f0_dir = "%s/2a_f0" % (exp_dir)
+        f0nsf_dir = "%s/2b-f0nsf" % (exp_dir)
+        names = (
+            set([name.split(".")[0] for name in os.listdir(gt_wavs_dir)])
+            & set([name.split(".")[0] for name in os.listdir(feature_dir)])
+            & set([name.split(".")[0] for name in os.listdir(f0_dir)])
+            & set([name.split(".")[0] for name in os.listdir(f0nsf_dir)])
+        )
+    else:
+        names = set([name.split(".")[0] for name in os.listdir(gt_wavs_dir)]) & set(
+            [name.split(".")[0] for name in os.listdir(feature_dir)]
+        )
+    opt = []
+    for name in names:
+        if if_f0_3:
+            opt.append(
+                "%s/%s.wav|%s/%s.npy|%s/%s.wav.npy|%s/%s.wav.npy|%s"
+                % (
+                    gt_wavs_dir.replace("\\", "\\\\"),
+                    name,
+                    feature_dir.replace("\\", "\\\\"),
+                    name,
+                    f0_dir.replace("\\", "\\\\"),
+                    name,
+                    f0nsf_dir.replace("\\", "\\\\"),
+                    name,
+                    spk_id5,
+                )
+            )
+        else:
+            opt.append(
+                "%s/%s.wav|%s/%s.npy|%s"
+                % (
+                    gt_wavs_dir.replace("\\", "\\\\"),
+                    name,
+                    feature_dir.replace("\\", "\\\\"),
+                    name,
+                    spk_id5,
+                )
+            )
+    fea_dim = 256 if version19 == "v1" else 768
+    if if_f0_3:
+        for _ in range(2):
+            opt.append(
+                "%s/logs/mute/0_gt_wavs/mute%s.wav|%s/logs/mute/3_feature%s/mute.npy|%s/logs/mute/2a_f0/mute.wav.npy|%s/logs/mute/2b-f0nsf/mute.wav.npy|%s"
+                % (now_dir, sr2, now_dir, fea_dim, now_dir, now_dir, spk_id5)
+            )
+    else:
+        for _ in range(2):
+            opt.append(
+                "%s/logs/mute/0_gt_wavs/mute%s.wav|%s/logs/mute/3_feature%s/mute.npy|%s"
+                % (now_dir, sr2, now_dir, fea_dim, spk_id5)
+            )
+    shuffle(opt)
+    with open("%s/filelist.txt" % exp_dir, "w") as f:
+        f.write("\n".join(opt))
+    logger.debug("Write filelist done")
+    # 生成config#无需生成config
+    # cmd = python_cmd + " train_nsf_sim_cache_sid_load_pretrain.py -e mi-test -sr 40k -f0 1 -bs 4 -g 0 -te 10 -se 5 -pg pretrained/f0G40k.pth -pd pretrained/f0D40k.pth -l 1 -c 0"
+    logger.info("Use gpus: %s", str(gpus16))
+    if pretrained_G14 == "":
+        logger.info("No pretrained Generator")
+    if pretrained_D15 == "":
+        logger.info("No pretrained Discriminator")
+    if version19 == "v1" or sr2 == "40k":
+        config_path = "v1/%s.json" % sr2
+    else:
+        config_path = "v2/%s.json" % sr2
+    config_save_path = os.path.join(exp_dir, "config.json")
+    if not pathlib.Path(config_save_path).exists():
+        with open(config_save_path, "w", encoding="utf-8") as f:
+            json.dump(
+                config.json_config[config_path],
+                f,
+                ensure_ascii=False,
+                indent=4,
+                sort_keys=True,
+            )
+            f.write("\n")
+    if gpus16:
+        cmd = (
+            '"%s" infer/modules/train/train.py -e "%s" -sr %s -f0 %s -bs %s -g %s -te %s -se %s %s %s -l %s -c %s -sw %s -v %s'
+            % (
+                config.python_cmd,
+                exp_dir1,
+                sr2,
+                1 if if_f0_3 else 0,
+                batch_size12,
+                gpus16,
+                total_epoch11,
+                save_epoch10,
+                "-pg %s" % pretrained_G14 if pretrained_G14 != "" else "",
+                "-pd %s" % pretrained_D15 if pretrained_D15 != "" else "",
+                1 if if_save_latest13 == i18n("是") else 0,
+                1 if if_cache_gpu17 == i18n("是") else 0,
+                1 if if_save_every_weights18 == i18n("是") else 0,
+                version19,
+            )
+        )
+    else:
+        cmd = (
+            '"%s" infer/modules/train/train.py -e "%s" -sr %s -f0 %s -bs %s -te %s -se %s %s %s -l %s -c %s -sw %s -v %s'
+            % (
+                config.python_cmd,
+                exp_dir1,
+                sr2,
+                1 if if_f0_3 else 0,
+                batch_size12,
+                total_epoch11,
+                save_epoch10,
+                "-pg %s" % pretrained_G14 if pretrained_G14 != "" else "",
+                "-pd %s" % pretrained_D15 if pretrained_D15 != "" else "",
+                1 if if_save_latest13 == i18n("是") else 0,
+                1 if if_cache_gpu17 == i18n("是") else 0,
+                1 if if_save_every_weights18 == i18n("是") else 0,
+                version19,
+            )
+        )
+    logger.info("Execute: " + cmd)
+    p = Popen(cmd, shell=True, cwd=now_dir)
+    p.wait()
+    return "训练结束, 您可查看控制台训练日志或实验文件夹下的train.log"
+
+
+# but4.click(train_index, [exp_dir1], info3)
+def train_index(exp_dir1, version19):
+    # exp_dir = "%s/logs/%s" % (now_dir, exp_dir1)
+    exp_dir = "logs/%s" % (exp_dir1)
+    os.makedirs(exp_dir, exist_ok=True)
+    feature_dir = (
+        "%s/3_feature256" % (exp_dir)
+        if version19 == "v1"
+        else "%s/3_feature768" % (exp_dir)
+    )
+    if not os.path.exists(feature_dir):
+        return "请先进行特征提取!"
+    listdir_res = list(os.listdir(feature_dir))
+    if len(listdir_res) == 0:
+        return "请先进行特征提取！"
+    infos = []
+    npys = []
+    for name in sorted(listdir_res):
+        phone = np.load("%s/%s" % (feature_dir, name))
+        npys.append(phone)
+    big_npy = np.concatenate(npys, 0)
+    big_npy_idx = np.arange(big_npy.shape[0])
+    np.random.shuffle(big_npy_idx)
+    big_npy = big_npy[big_npy_idx]
+    if big_npy.shape[0] > 2e5:
+        infos.append("Trying doing kmeans %s shape to 10k centers." % big_npy.shape[0])
+        yield "\n".join(infos)
+        try:
+            big_npy = (
+                MiniBatchKMeans(
+                    n_clusters=10000,
+                    verbose=True,
+                    batch_size=256 * config.n_cpu,
+                    compute_labels=False,
+                    init="random",
+                )
+                .fit(big_npy)
+                .cluster_centers_
+            )
+        except:
+            info = traceback.format_exc()
+            logger.info(info)
+            infos.append(info)
+            yield "\n".join(infos)
+
+    np.save("%s/total_fea.npy" % exp_dir, big_npy)
+    n_ivf = min(int(16 * np.sqrt(big_npy.shape[0])), big_npy.shape[0] // 39)
+    infos.append("%s,%s" % (big_npy.shape, n_ivf))
+    yield "\n".join(infos)
+    index = faiss.index_factory(256 if version19 == "v1" else 768, "IVF%s,Flat" % n_ivf)
+    # index = faiss.index_factory(256if version19=="v1"else 768, "IVF%s,PQ128x4fs,RFlat"%n_ivf)
+    infos.append("training")
+    yield "\n".join(infos)
+    index_ivf = faiss.extract_index_ivf(index)  #
+    index_ivf.nprobe = 1
+    index.train(big_npy)
+    faiss.write_index(
+        index,
+        "%s/trained_IVF%s_Flat_nprobe_%s_%s_%s.index"
+        % (exp_dir, n_ivf, index_ivf.nprobe, exp_dir1, version19),
+    )
+    infos.append("adding")
+    yield "\n".join(infos)
+    batch_size_add = 8192
+    for i in range(0, big_npy.shape[0], batch_size_add):
+        index.add(big_npy[i : i + batch_size_add])
+    faiss.write_index(
+        index,
+        "%s/added_IVF%s_Flat_nprobe_%s_%s_%s.index"
+        % (exp_dir, n_ivf, index_ivf.nprobe, exp_dir1, version19),
+    )
+    infos.append(
+        "成功构建索引 added_IVF%s_Flat_nprobe_%s_%s_%s.index"
+        % (n_ivf, index_ivf.nprobe, exp_dir1, version19)
+    )
+    try:
+        link = os.link if platform.system() == "Windows" else os.symlink
+        link(
+            "%s/added_IVF%s_Flat_nprobe_%s_%s_%s.index"
+            % (exp_dir, n_ivf, index_ivf.nprobe, exp_dir1, version19),
+            "%s/%s_IVF%s_Flat_nprobe_%s_%s_%s.index"
+            % (
+                outside_index_root,
+                exp_dir1,
+                n_ivf,
+                index_ivf.nprobe,
+                exp_dir1,
+                version19,
+            ),
+        )
+        infos.append("链接索引到外部-%s" % (outside_index_root))
+    except:
+        infos.append("链接索引到外部-%s失败" % (outside_index_root))
+
+    # faiss.write_index(index, '%s/added_IVF%s_Flat_FastScan_%s.index'%(exp_dir,n_ivf,version19))
+    # infos.append("成功构建索引，added_IVF%s_Flat_FastScan_%s.index"%(n_ivf,version19))
+    yield "\n".join(infos)
+
+
+# but5.click(train1key, [exp_dir1, sr2, if_f0_3, trainset_dir4, spk_id5, gpus6, np7, f0method8, save_epoch10, total_epoch11, batch_size12, if_save_latest13, pretrained_G14, pretrained_D15, gpus16, if_cache_gpu17], info3)
+def train1key(
+    exp_dir1,
+    sr2,
+    if_f0_3,
+    trainset_dir4,
+    spk_id5,
+    np7,
+    f0method8,
+    save_epoch10,
+    total_epoch11,
+    batch_size12,
+    if_save_latest13,
+    pretrained_G14,
+    pretrained_D15,
+    gpus16,
+    if_cache_gpu17,
+    if_save_every_weights18,
+    version19,
+    gpus_rmvpe,
+):
+    infos = []
+
+    def get_info_str(strr):
+        infos.append(strr)
+        return "\n".join(infos)
+
+    # step1:处理数据
+    yield get_info_str(i18n("step1:正在处理数据"))
+    [get_info_str(_) for _ in preprocess_dataset(trainset_dir4, exp_dir1, sr2, np7)]
+
+    # step2a:提取音高
+    yield get_info_str(i18n("step2:正在提取音高&正在提取特征"))
+    [
+        get_info_str(_)
+        for _ in extract_f0_feature(
+            gpus16, np7, f0method8, if_f0_3, exp_dir1, version19, gpus_rmvpe
+        )
+    ]
+
+    # step3a:训练模型
+    yield get_info_str(i18n("step3a:正在训练模型"))
+    click_train(
+        exp_dir1,
+        sr2,
+        if_f0_3,
+        spk_id5,
+        save_epoch10,
+        total_epoch11,
+        batch_size12,
+        if_save_latest13,
+        pretrained_G14,
+        pretrained_D15,
+        gpus16,
+        if_cache_gpu17,
+        if_save_every_weights18,
+        version19,
+    )
+    yield get_info_str(
+        i18n("训练结束, 您可查看控制台训练日志或实验文件夹下的train.log")
+    )
+
+    # step3b:训练索引
+    [get_info_str(_) for _ in train_index(exp_dir1, version19)]
+    yield get_info_str(i18n("全流程结束！"))
+
+
+#                    ckpt_path2.change(change_info_,[ckpt_path2],[sr__,if_f0__])
+def change_info_(ckpt_path):
+    if not os.path.exists(ckpt_path.replace(os.path.basename(ckpt_path), "train.log")):
+        return {"__type__": "update"}, {"__type__": "update"}, {"__type__": "update"}
+    try:
+        with open(
+            ckpt_path.replace(os.path.basename(ckpt_path), "train.log"), "r"
+        ) as f:
+            info = eval(f.read().strip("\n").split("\n")[0].split("\t")[-1])
+            sr, f0 = info["sample_rate"], info["if_f0"]
+            version = "v2" if ("version" in info and info["version"] == "v2") else "v1"
+            return sr, str(f0), version
+    except:
+        traceback.print_exc()
+        return {"__type__": "update"}, {"__type__": "update"}, {"__type__": "update"}
+
+
+F0GPUVisible = config.dml == False
+
+
+def change_f0_method(f0method8):
+    if f0method8 == "rmvpe_gpu":
+        visible = F0GPUVisible
+    else:
+        visible = False
+    return {"visible": visible, "__type__": "update"}
+
+
+with gr.Blocks(title="RVC WebUI") as app:
+    gr.Markdown("## RVC WebUI")
+    gr.Markdown(
+        value=i18n(
+            "本软件以MIT协议开源, 作者不对软件具备任何控制力, 使用软件者、传播软件导出的声音者自负全责. <br>如不认可该条款, 则不能使用或引用软件包内任何代码和文件. 详见根目录<b>LICENSE</b>."
+        )
+    )
+    with gr.Tabs():
+        with gr.TabItem(i18n("训练")):
+            gr.Markdown(
+                value=i18n(
+                    "step1: 填写实验配置. 实验数据放在logs下, 每个实验一个文件夹, 需手工输入实验名路径, 内含实验配置, 日志, 训练得到的模型文件. "
+                )
+            )
+            with gr.Row():
+                exp_dir1 = gr.Textbox(label=i18n("输入实验名"), value="mi-test")
+                sr2 = gr.Radio(
+                    label=i18n("目标采样率"),
+                    choices=["40k", "48k"],
+                    value="40k",
+                    interactive=True,
+                )
+                if_f0_3 = gr.Radio(
+                    label=i18n("模型是否带音高指导(唱歌一定要, 语音可以不要)"),
+                    choices=[True, False],
+                    value=True,
+                    interactive=True,
+                )
+                version19 = gr.Radio(
+                    label=i18n("版本"),
+                    choices=["v1", "v2"],
+                    value="v2",
+                    interactive=True,
+                    visible=True,
+                )
+                np7 = gr.Slider(
+                    minimum=0,
+                    maximum=config.n_cpu,
+                    step=1,
+                    label=i18n("提取音高和处理数据使用的CPU进程数"),
+                    value=int(np.ceil(config.n_cpu / 1.5)),
+                    interactive=True,
+                )
+            with gr.Group():  # 暂时单人的, 后面支持最多4人的#数据处理
+                gr.Markdown(
+                    value=i18n(
+                        "step2a: 自动遍历训练文件夹下所有可解码成音频的文件并进行切片归一化, 在实验目录下生成2个wav文件夹; 暂时只支持单人训练. "
+                    )
+                )
+                with gr.Row():
+                    trainset_dir4 = gr.Textbox(
+                        label=i18n("输入训练文件夹路径"),
+                        value=i18n("E:\\语音音频+标注\\米津玄师\\src"),
+                    )
+                    spk_id5 = gr.Slider(
+                        minimum=0,
+                        maximum=4,
+                        step=1,
+                        label=i18n("请指定说话人id"),
+                        value=0,
+                        interactive=True,
+                    )
+                    but1 = gr.Button(i18n("处理数据"), variant="primary")
+                    info1 = gr.Textbox(label=i18n("输出信息"), value="")
+                    but1.click(
+                        preprocess_dataset,
+                        [trainset_dir4, exp_dir1, sr2, np7],
+                        [info1],
+                        api_name="train_preprocess",
+                    )
+            with gr.Group():
+                gr.Markdown(
+                    value=i18n(
+                        "step2b: 使用CPU提取音高(如果模型带音高), 使用GPU提取特征(选择卡号)"
+                    )
+                )
+                with gr.Row():
+                    with gr.Column():
+                        gpus6 = gr.Textbox(
+                            label=i18n(
+                                "以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"
+                            ),
+                            value=gpus,
+                            interactive=True,
+                            visible=F0GPUVisible,
+                        )
+                        gpu_info9 = gr.Textbox(
+                            label=i18n("显卡信息"), value=gpu_info, visible=F0GPUVisible
+                        )
+                    with gr.Column():
+                        f0method8 = gr.Radio(
+                            label=i18n(
+                                "选择音高提取算法:输入歌声可用pm提速,高质量语音但CPU差可用dio提速,harvest质量更好但慢,rmvpe效果最好且微吃CPU/GPU"
+                            ),
+                            choices=["pm", "harvest", "dio", "rmvpe", "rmvpe_gpu"],
+                            value="rmvpe_gpu",
+                            interactive=True,
+                        )
+                        gpus_rmvpe = gr.Textbox(
+                            label=i18n(
+                                "rmvpe卡号配置：以-分隔输入使用的不同进程卡号,例如0-0-1使用在卡0上跑2个进程并在卡1上跑1个进程"
+                            ),
+                            value="%s-%s" % (gpus, gpus),
+                            interactive=True,
+                            visible=F0GPUVisible,
+                        )
+                    but2 = gr.Button(i18n("特征提取"), variant="primary")
+                    info2 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    f0method8.change(
+                        fn=change_f0_method,
+                        inputs=[f0method8],
+                        outputs=[gpus_rmvpe],
+                    )
+                    but2.click(
+                        extract_f0_feature,
+                        [
+                            gpus6,
+                            np7,
+                            f0method8,
+                            if_f0_3,
+                            exp_dir1,
+                            version19,
+                            gpus_rmvpe,
+                        ],
+                        [info2],
+                        api_name="train_extract_f0_feature",
+                    )
+            with gr.Group():
+                gr.Markdown(value=i18n("step3: 填写训练设置, 开始训练模型和索引"))
+                with gr.Row():
+                    save_epoch10 = gr.Slider(
+                        minimum=1,
+                        maximum=50,
+                        step=1,
+                        label=i18n("保存频率save_every_epoch"),
+                        value=5,
+                        interactive=True,
+                    )
+                    total_epoch11 = gr.Slider(
+                        minimum=2,
+                        maximum=1000,
+                        step=1,
+                        label=i18n("总训练轮数total_epoch"),
+                        value=20,
+                        interactive=True,
+                    )
+                    batch_size12 = gr.Slider(
+                        minimum=1,
+                        maximum=40,
+                        step=1,
+                        label=i18n("每张显卡的batch_size"),
+                        value=default_batch_size,
+                        interactive=True,
+                    )
+                    if_save_latest13 = gr.Radio(
+                        label=i18n("是否仅保存最新的ckpt文件以节省硬盘空间"),
+                        choices=[i18n("是"), i18n("否")],
+                        value=i18n("否"),
+                        interactive=True,
+                    )
+                    if_cache_gpu17 = gr.Radio(
+                        label=i18n(
+                            "是否缓存所有训练集至显存. 10min以下小数据可缓存以加速训练, 大数据缓存会炸显存也加不了多少速"
+                        ),
+                        choices=[i18n("是"), i18n("否")],
+                        value=i18n("否"),
+                        interactive=True,
+                    )
+                    if_save_every_weights18 = gr.Radio(
+                        label=i18n(
+                            "是否在每次保存时间点将最终小模型保存至weights文件夹"
+                        ),
+                        choices=[i18n("是"), i18n("否")],
+                        value=i18n("否"),
+                        interactive=True,
+                    )
+                with gr.Row():
+                    pretrained_G14 = gr.Textbox(
+                        label=i18n("加载预训练底模G路径"),
+                        value="assets/pretrained_v2/f0G40k.pth",
+                        interactive=True,
+                    )
+                    pretrained_D15 = gr.Textbox(
+                        label=i18n("加载预训练底模D路径"),
+                        value="assets/pretrained_v2/f0D40k.pth",
+                        interactive=True,
+                    )
+                    sr2.change(
+                        change_sr2,
+                        [sr2, if_f0_3, version19],
+                        [pretrained_G14, pretrained_D15],
+                    )
+                    version19.change(
+                        change_version19,
+                        [sr2, if_f0_3, version19],
+                        [pretrained_G14, pretrained_D15, sr2],
+                    )
+                    if_f0_3.change(
+                        change_f0,
+                        [if_f0_3, sr2, version19],
+                        [f0method8, gpus_rmvpe, pretrained_G14, pretrained_D15],
+                    )
+                    gpus16 = gr.Textbox(
+                        label=i18n(
+                            "以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"
+                        ),
+                        value=gpus,
+                        interactive=True,
+                    )
+                    but3 = gr.Button(i18n("训练模型"), variant="primary")
+                    but4 = gr.Button(i18n("训练特征索引"), variant="primary")
+                    but5 = gr.Button(i18n("一键训练"), variant="primary")
+                    info3 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=10)
+                    but3.click(
+                        click_train,
+                        [
+                            exp_dir1,
+                            sr2,
+                            if_f0_3,
+                            spk_id5,
+                            save_epoch10,
+                            total_epoch11,
+                            batch_size12,
+                            if_save_latest13,
+                            pretrained_G14,
+                            pretrained_D15,
+                            gpus16,
+                            if_cache_gpu17,
+                            if_save_every_weights18,
+                            version19,
+                        ],
+                        info3,
+                        api_name="train_start",
+                    )
+                    but4.click(train_index, [exp_dir1, version19], info3)
+                    but5.click(
+                        train1key,
+                        [
+                            exp_dir1,
+                            sr2,
+                            if_f0_3,
+                            trainset_dir4,
+                            spk_id5,
+                            np7,
+                            f0method8,
+                            save_epoch10,
+                            total_epoch11,
+                            batch_size12,
+                            if_save_latest13,
+                            pretrained_G14,
+                            pretrained_D15,
+                            gpus16,
+                            if_cache_gpu17,
+                            if_save_every_weights18,
+                            version19,
+                            gpus_rmvpe,
+                        ],
+                        info3,
+                        api_name="train_start_all",
+                    )
+
+        with gr.TabItem(i18n("ckpt处理")):
+            with gr.Group():
+                gr.Markdown(value=i18n("模型融合, 可用于测试音色融合"))
+                with gr.Row():
+                    ckpt_a = gr.Textbox(
+                        label=i18n("A模型路径"), value="", interactive=True
+                    )
+                    ckpt_b = gr.Textbox(
+                        label=i18n("B模型路径"), value="", interactive=True
+                    )
+                    alpha_a = gr.Slider(
+                        minimum=0,
+                        maximum=1,
+                        label=i18n("A模型权重"),
+                        value=0.5,
+                        interactive=True,
+                    )
+                with gr.Row():
+                    sr_ = gr.Radio(
+                        label=i18n("目标采样率"),
+                        choices=["40k", "48k"],
+                        value="40k",
+                        interactive=True,
+                    )
+                    if_f0_ = gr.Radio(
+                        label=i18n("模型是否带音高指导"),
+                        choices=[i18n("是"), i18n("否")],
+                        value=i18n("是"),
+                        interactive=True,
+                    )
+                    info__ = gr.Textbox(
+                        label=i18n("要置入的模型信息"),
+                        value="",
+                        max_lines=8,
+                        interactive=True,
+                    )
+                    name_to_save0 = gr.Textbox(
+                        label=i18n("保存的模型名不带后缀"),
+                        value="",
+                        max_lines=1,
+                        interactive=True,
+                    )
+                    version_2 = gr.Radio(
+                        label=i18n("模型版本型号"),
+                        choices=["v1", "v2"],
+                        value="v1",
+                        interactive=True,
+                    )
+                with gr.Row():
+                    but6 = gr.Button(i18n("融合"), variant="primary")
+                    info4 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                but6.click(
+                    merge,
+                    [
+                        ckpt_a,
+                        ckpt_b,
+                        alpha_a,
+                        sr_,
+                        if_f0_,
+                        info__,
+                        name_to_save0,
+                        version_2,
+                    ],
+                    info4,
+                    api_name="ckpt_merge",
+                )  # def merge(path1,path2,alpha1,sr,f0,info):
+            with gr.Group():
+                gr.Markdown(
+                    value=i18n("修改模型信息(仅支持weights文件夹下提取的小模型文件)")
+                )
+                with gr.Row():
+                    ckpt_path0 = gr.Textbox(
+                        label=i18n("模型路径"), value="", interactive=True
+                    )
+                    info_ = gr.Textbox(
+                        label=i18n("要改的模型信息"),
+                        value="",
+                        max_lines=8,
+                        interactive=True,
+                    )
+                    name_to_save1 = gr.Textbox(
+                        label=i18n("保存的文件名, 默认空为和源文件同名"),
+                        value="",
+                        max_lines=8,
+                        interactive=True,
+                    )
+                with gr.Row():
+                    but7 = gr.Button(i18n("修改"), variant="primary")
+                    info5 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                but7.click(
+                    change_info,
+                    [ckpt_path0, info_, name_to_save1],
+                    info5,
+                    api_name="ckpt_modify",
+                )
+            with gr.Group():
+                gr.Markdown(
+                    value=i18n("查看模型信息(仅支持weights文件夹下提取的小模型文件)")
+                )
+                with gr.Row():
+                    ckpt_path1 = gr.Textbox(
+                        label=i18n("模型路径"), value="", interactive=True
+                    )
+                    but8 = gr.Button(i18n("查看"), variant="primary")
+                    info6 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                but8.click(show_info, [ckpt_path1], info6, api_name="ckpt_show")
+            with gr.Group():
+                gr.Markdown(
+                    value=i18n(
+                        "模型提取(输入logs文件夹下大文件模型路径),适用于训一半不想训了模型没有自动提取保存小文件模型,或者想测试中间模型的情况"
+                    )
+                )
+                with gr.Row():
+                    ckpt_path2 = gr.Textbox(
+                        label=i18n("模型路径"),
+                        value="E:\\codes\\py39\\logs\\mi-test_f0_48k\\G_23333.pth",
+                        interactive=True,
+                    )
+                    save_name = gr.Textbox(
+                        label=i18n("保存名"), value="", interactive=True
+                    )
+                    sr__ = gr.Radio(
+                        label=i18n("目标采样率"),
+                        choices=["32k", "40k", "48k"],
+                        value="40k",
+                        interactive=True,
+                    )
+                    if_f0__ = gr.Radio(
+                        label=i18n("模型是否带音高指导,1是0否"),
+                        choices=["1", "0"],
+                        value="1",
+                        interactive=True,
+                    )
+                    version_1 = gr.Radio(
+                        label=i18n("模型版本型号"),
+                        choices=["v1", "v2"],
+                        value="v2",
+                        interactive=True,
+                    )
+                    info___ = gr.Textbox(
+                        label=i18n("要置入的模型信息"),
+                        value="",
+                        max_lines=8,
+                        interactive=True,
+                    )
+                    but9 = gr.Button(i18n("提取"), variant="primary")
+                    info7 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    ckpt_path2.change(
+                        change_info_, [ckpt_path2], [sr__, if_f0__, version_1]
+                    )
+                but9.click(
+                    extract_small_model,
+                    [ckpt_path2, save_name, sr__, if_f0__, info___, version_1],
+                    info7,
+                    api_name="ckpt_extract",
+                )
+
+        with gr.TabItem(i18n("Onnx导出")):
+            with gr.Row():
+                ckpt_dir = gr.Textbox(
+                    label=i18n("RVC模型路径"), value="", interactive=True
+                )
+            with gr.Row():
+                onnx_dir = gr.Textbox(
+                    label=i18n("Onnx输出路径"), value="", interactive=True
+                )
+            with gr.Row():
+                infoOnnx = gr.Label(label="info")
+            with gr.Row():
+                butOnnx = gr.Button(i18n("导出Onnx模型"), variant="primary")
+            butOnnx.click(
+                export_onnx, [ckpt_dir, onnx_dir], infoOnnx, api_name="export_onnx"
+            )
+
+        tab_faq = i18n("常见问题解答")
+        with gr.TabItem(tab_faq):
+            try:
+                if tab_faq == "常见问题解答":
+                    with open("docs/cn/faq.md", "r", encoding="utf8") as f:
+                        info = f.read()
+                else:
+                    with open("docs/en/faq_en.md", "r", encoding="utf8") as f:
+                        info = f.read()
+                gr.Markdown(value=info)
+            except:
+                gr.Markdown(traceback.format_exc())
+
+    if config.iscolab:
+        app.queue(concurrency_count=511, max_size=1022).launch(share=True)
+    else:
+        app.queue(concurrency_count=511, max_size=1022).launch(
+            server_name="0.0.0.0",
+            inbrowser=not config.noautoopen,
+            server_port=config.listen_port,
+            quiet=True,
+        )

infer/modules/train/train.py

@@ -11,9 +11,9 @@ import datetime
 
 from infer.lib.train import utils
 
-hps = utils.get_hparams()
-os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",")
-n_gpus = len(hps.gpus.split("-"))
+# hps = utils.get_hparams()
+# os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",")
+# n_gpus = len(hps.gpus.split("-"))
 from random import randint, shuffle
 
 import torch
@@ -54,18 +54,18 @@ from infer.lib.train.data_utils import (
     TextAudioLoaderMultiNSFsid,
 )
 
-if hps.version == "v1":
-    from infer.lib.infer_pack.models import MultiPeriodDiscriminator
-    from infer.lib.infer_pack.models import SynthesizerTrnMs256NSFsid as RVC_Model_f0
-    from infer.lib.infer_pack.models import (
-        SynthesizerTrnMs256NSFsid_nono as RVC_Model_nof0,
-    )
-else:
-    from infer.lib.infer_pack.models import (
-        SynthesizerTrnMs768NSFsid as RVC_Model_f0,
-        SynthesizerTrnMs768NSFsid_nono as RVC_Model_nof0,
-        MultiPeriodDiscriminatorV2 as MultiPeriodDiscriminator,
-    )
+# if hps.version == "v1":
+#     from infer.lib.infer_pack.models import MultiPeriodDiscriminator
+#     from infer.lib.infer_pack.models import SynthesizerTrnMs256NSFsid as RVC_Model_f0
+#     from infer.lib.infer_pack.models import (
+#         SynthesizerTrnMs256NSFsid_nono as RVC_Model_nof0,
+#     )
+# else:
+from infer.lib.infer_pack.models import (
+    SynthesizerTrnMs768NSFsid as RVC_Model_f0,
+    SynthesizerTrnMs768NSFsid_nono as RVC_Model_nof0,
+    MultiPeriodDiscriminatorV2 as MultiPeriodDiscriminator,
+)
 
 from infer.lib.train.losses import (
     discriminator_loss,
@@ -76,8 +76,6 @@ from infer.lib.train.losses import (
 from infer.lib.train.mel_processing import mel_spectrogram_torch, spec_to_mel_torch
 from infer.lib.train.process_ckpt import savee
 
-global_step = 0
-
 
 class EpochRecorder:
     def __init__(self):
@@ -92,33 +90,31 @@ class EpochRecorder:
         return f"[{current_time}] | ({elapsed_time_str})"
 
 
-def main():
-    n_gpus = torch.cuda.device_count()
+def train(exp_dir: str):
+    state = {"global_step": 0}
+
+    hps = utils.get_hparams_from_dir(exp_dir)
+    hps.experiment_dir = exp_dir
+    hps.save_every_epoch = False
+    hps.name = os.path.basename(exp_dir)
+    hps.total_epoch = 100
+    hps.pretrainG = ""
+    hps.pretrainD = ""
+    hps.version = "v2"
+    hps.gpus = "0"
+    hps.train.batch_size = 8
+    hps.sample_rate = "40k"
+    hps.if_f0 = 1
+    hps.if_latest = 1
+    hps.save_every_weights = "0"
+    hps.if_cache_data_in_gpu = True
+    hps.data.training_files = "%s/filelist.txt" % exp_dir
 
-    if torch.cuda.is_available() == False and torch.backends.mps.is_available() == True:
-        n_gpus = 1
-    if n_gpus < 1:
-        # patch to unblock people without gpus. there is probably a better way.
-        print("NO GPU DETECTED: falling back to CPU - this may take a while")
-        n_gpus = 1
-    os.environ["MASTER_ADDR"] = "localhost"
-    os.environ["MASTER_PORT"] = str(randint(20000, 55555))
-    children = []
     logger = utils.get_logger(hps.model_dir)
-    for i in range(n_gpus):
-        subproc = mp.Process(
-            target=run,
-            args=(i, n_gpus, hps, logger),
-        )
-        children.append(subproc)
-        subproc.start()
-
-    for i in range(n_gpus):
-        children[i].join()
+    run(0, 1, hps, logger, state)
 
 
-def run(rank, n_gpus, hps, logger: logging.Logger):
-    global global_step
+def run(rank, n_gpus, hps, logger: logging.Logger, state):
     if rank == 0:
         # logger = utils.get_logger(hps.model_dir)
         logger.info(hps)
@@ -215,13 +211,14 @@ def run(rank, n_gpus, hps, logger: logging.Logger):
         _, _, _, epoch_str = utils.load_checkpoint(
             utils.latest_checkpoint_path(hps.model_dir, "G_*.pth"), net_g, optim_g
         )
-        global_step = (epoch_str - 1) * len(train_loader)
+        state["global_step"] = (epoch_str - 1) * len(train_loader)
+        print("loaded", epoch_str)
         # epoch_str = 1
         # global_step = 0
     except:  # 如果首次不能加载，加载pretrain
         # traceback.print_exc()
         epoch_str = 1
-        global_step = 0
+        state["global_step"] = 0
         if hps.pretrainG != "":
             if rank == 0:
                 logger.info("loaded pretrained %s" % (hps.pretrainG))
@@ -252,6 +249,7 @@ def run(rank, n_gpus, hps, logger: logging.Logger):
                         torch.load(hps.pretrainD, map_location="cpu")["model"]
                     )
                 )
+        print("new")
 
     scheduler_g = torch.optim.lr_scheduler.ExponentialLR(
         optim_g, gamma=hps.train.lr_decay, last_epoch=epoch_str - 2
@@ -264,6 +262,7 @@ def run(rank, n_gpus, hps, logger: logging.Logger):
 
     cache = []
     for epoch in range(epoch_str, hps.train.epochs + 1):
+        print("epoch", epoch)
         if rank == 0:
             train_and_evaluate(
                 rank,
@@ -277,6 +276,7 @@ def run(rank, n_gpus, hps, logger: logging.Logger):
                 logger,
                 [writer, writer_eval],
                 cache,
+                state,
             )
         else:
             train_and_evaluate(
@@ -291,13 +291,25 @@ def run(rank, n_gpus, hps, logger: logging.Logger):
                 None,
                 None,
                 cache,
+                state,
             )
         scheduler_g.step()
         scheduler_d.step()
 
 
 def train_and_evaluate(
-    rank, epoch, hps, nets, optims, schedulers, scaler, loaders, logger, writers, cache
+    rank,
+    epoch,
+    hps,
+    nets,
+    optims,
+    schedulers,
+    scaler,
+    loaders,
+    logger,
+    writers,
+    cache,
+    state,
 ):
     net_g, net_d = nets
     optim_g, optim_d = optims
@@ -306,7 +318,6 @@ def train_and_evaluate(
         writer, writer_eval = writers
 
     train_loader.batch_sampler.set_epoch(epoch)
-    global global_step
 
     net_g.train()
     net_d.train()
@@ -500,7 +511,7 @@ def train_and_evaluate(
         scaler.update()
 
         if rank == 0:
-            if global_step % hps.train.log_interval == 0:
+            if state["global_step"] % hps.train.log_interval == 0:
                 lr = optim_g.param_groups[0]["lr"]
                 logger.info(
                     "Train Epoch: {} [{:.0f}%]".format(
@@ -513,7 +524,7 @@ def train_and_evaluate(
                 if loss_kl > 9:
                     loss_kl = 9
 
-                logger.info([global_step, lr])
+                logger.info([state["global_step"], lr])
                 logger.info(
                     f"loss_disc={loss_disc:.3f}, loss_gen={loss_gen:.3f}, loss_fm={loss_fm:.3f},loss_mel={loss_mel:.3f}, loss_kl={loss_kl:.3f}"
                 )
@@ -554,11 +565,11 @@ def train_and_evaluate(
                 }
                 utils.summarize(
                     writer=writer,
-                    global_step=global_step,
+                    global_step=state["global_step"],
                     images=image_dict,
                     scalars=scalar_dict,
                 )
-        global_step += 1
+        state["global_step"] += 1
     # /Run steps
 
     if epoch % hps.save_every_epoch == 0 and rank == 0:
@@ -568,14 +579,14 @@ def train_and_evaluate(
                 optim_g,
                 hps.train.learning_rate,
                 epoch,
-                os.path.join(hps.model_dir, "G_{}.pth".format(global_step)),
+                os.path.join(hps.model_dir, "G_{}.pth".format(state["global_step"])),
             )
             utils.save_checkpoint(
                 net_d,
                 optim_d,
                 hps.train.learning_rate,
                 epoch,
-                os.path.join(hps.model_dir, "D_{}.pth".format(global_step)),
+                os.path.join(hps.model_dir, "D_{}.pth".format(state["global_step"])),
             )
         else:
             utils.save_checkpoint(
@@ -606,7 +617,7 @@ def train_and_evaluate(
                         ckpt,
                         hps.sample_rate,
                         hps.if_f0,
-                        hps.name + "_e%s_s%s" % (epoch, global_step),
+                        hps.name + "_e%s_s%s" % (epoch, state["global_step"]),
                         epoch,
                         hps.version,
                         hps,
@@ -633,8 +644,3 @@ def train_and_evaluate(
         )
         sleep(1)
         os._exit(2333333)
-
-
-if __name__ == "__main__":
-    torch.multiprocessing.set_start_method("spawn")
-    main()

logs/mute/0_gt_wavs/mute32k.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9edcf85ec77e88bd01edf3d887bdc418d3596d573f7ad2694da546f41dae6baf
+size 192078

logs/mute/0_gt_wavs/mute40k.spec.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bbca900dcff32be4d664383a705d53ebc6829027ac8a07d78308d472c9087a1
+size 1230339

logs/mute/0_gt_wavs/mute40k.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67a816e77b50cb9f016e49e5c01f07e080c4e3b82b7a8ac3e64bcb143f90f31b
+size 240078

logs/mute/0_gt_wavs/mute48k.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f2bb4daaa106e351aebb001e5a25de985c0b472f22e8d60676bc924a79056ee
+size 288078

logs/mute/1_16k_wavs/mute.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e233e86ba1be365e1133f157d56b61110086b89650ecfbdfc013c759e466250
+size 96078

logs/mute/2a_f0/mute.wav.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b9acf9ab7facdb032e1d687fe35182670b0b94566c4b209ae48c239d19956a6
+size 1332

logs/mute/2b-f0nsf/mute.wav.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30792849c8e72d67e6691754077f2888b101cb741e9c7f193c91dd9692870c87
+size 2536

logs/mute/3_feature256/mute.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64d5abbac078e19a3f649c0d78a02cb33a71407ded3ddf2db78e6b803d0c0126
+size 152704

logs/mute/3_feature768/mute.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16ef62b957887ac9f0913aa5158f18983afff1ef5a3e4c5fd067ac20fc380d54
+size 457856