feat: infer

.gitignore

@@ -1,2 +1,3 @@
 .DS_Store
 *.pyc
+__pycache__

README.md

@@ -1,5 +1,5 @@
 ---
-title: RVC Trainer
+title: ZeroRVC
 emoji: 🦀
 colorFrom: gray
 colorTo: gray
@@ -9,4 +9,6 @@ app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# ZeroRVC
+
+Run Retrieval-based Voice Conversion training and inference on HuggingFace ZeroGPU.

app.py

@@ -1,11 +1,12 @@
+from typing import Tuple
+from prelude import prelude
+
+prelude()
+
 import os
 import traceback
-
 import numpy as np
 from sklearn.cluster import MiniBatchKMeans
-
-os.environ["PYTORCH_JIT"] = "0v"
-
 from random import shuffle
 import gradio as gr
 import zipfile
@@ -18,23 +19,12 @@ from infer.modules.train.extract.extract_f0_rmvpe import FeatureInput
 from infer.modules.train.extract_feature_print import HubertFeatureExtractor
 from infer.modules.train.train import train
 from infer.lib.train.process_ckpt import extract_small_model
+from infer.modules.vc.modules import VC
+from configs.config import Config
+import demucs.separate
+import soundfile as sf
 from zero import zero
-
-# patch for jit script
-# if we find `def expand_2d_or_3d_tensor(x,` in /usr/local/lib/python3.10/site-packages/fairseq/models/model_utils.py
-# patch it with `def expand_2d_or_3d_tensor(x: Tensor,`
-FAIRSEQ_CODE = "/usr/local/lib/python3.10/site-packages/fairseq/models/model_utils.py"
-if os.path.exists(FAIRSEQ_CODE):
-    with open(FAIRSEQ_CODE, "r") as f:
-        lines = f.readlines()
-    with open(FAIRSEQ_CODE, "w") as f:
-        for line in lines:
-            if "def expand_2d_or_3d_tensor(x, trg_dim: int, padding_idx: int):" in line:
-                f.write(
-                    "def expand_2d_or_3d_tensor(x: Tensor, trg_dim: int, padding_idx: int) -> Tensor:\n"
-                )
-            else:
-                f.write(line)
+from model import device
 
 
 def extract_audio_files(zip_file: str, target_dir: str) -> list[str]:
@@ -189,13 +179,15 @@ def download_weight(exp_dir: str) -> str:
         raise gr.Error("No model found")
 
     latest_model = max(models, key=os.path.getctime)
+    print(f"Latest model: {latest_model}")
 
     name = os.path.basename(exp_dir)
+    out = os.path.join(exp_dir, f"{name}.pth")
     extract_small_model(
-        latest_model, name, "40k", True, "Model trained by ZeroGPU.", "v2"
+        latest_model, out, "40k", True, "Model trained by ZeroGPU.", "v2"
     )
 
-    return "assets/weights/%s.pth" % name
+    return out
 
 
 def train_index(exp_dir: str) -> str:
@@ -269,9 +261,70 @@ def restore_expdir(zip: str) -> str:
     return exp_dir
 
 
+@zero(duration=120)
+def infer(exp_dir: str, original_audio: str, f0add: int) -> Tuple[int, np.ndarray]:
+    name = os.path.basename(exp_dir)
+    model = os.path.join(exp_dir, f"{name}.pth")
+    if not os.path.exists(model):
+        raise gr.Error("Model not found")
+
+    index = glob(f"{exp_dir}/added_*.index")
+    if not index:
+        raise gr.Error("Index not found")
+
+    base = os.path.basename(original_audio)
+    base = os.path.splitext(base)[0]
+    demucs.separate.main(
+        ["--two-stems", "vocals", "-d", str(device), "-n", "htdemucs", original_audio]
+    )
+    out = os.path.join("separated", "htdemucs", base, "vocals.wav")
+
+    cfg = Config()
+    vc = VC(cfg)
+    vc.get_vc(model)
+    _, wav_opt = vc.vc_single(
+        0,
+        out,
+        f0add,
+        None,
+        "rmvpe",
+        index,
+        None,
+        0.5,
+        3,
+        0,
+        1,
+        0.33,
+    )
+
+    sr = wav_opt[0]
+    data = wav_opt[1]
+
+    return sr, data
+
+
+def merge(exp_dir: str, original_audio: str, vocal: Tuple[int, np.ndarray]) -> str:
+    base = os.path.basename(original_audio)
+    base = os.path.splitext(base)[0]
+    music = os.path.join("separated", "htdemucs", base, "no-vocals.wav")
+
+    tmp = os.path.join(exp_dir, "tmp.wav")
+    sf.write(tmp, vocal[1], vocal[0])
+
+    os.system(
+        f"ffmpeg -i {music} -i {tmp} -filter_complex '[1]volume=2[a];[0][a]amix=inputs=2:duration=first:dropout_transition=2' {tmp}.merged.mp3"
+    )
+
+    return f"{tmp}.merged.mp3"
+
+
 with gr.Blocks() as app:
     # allow user to manually select the experiment directory
-    exp_dir = gr.Textbox(label="Experiment directory (don't touch it unless you know what you are doing)", visible=True, interactive=True)
+    exp_dir = gr.Textbox(
+        label="Experiment directory (don't touch it unless you know what you are doing)",
+        visible=True,
+        interactive=True,
+    )
 
     with gr.Tabs():
         with gr.Tab(label="New / Restore"):
@@ -284,10 +337,10 @@ with gr.Blocks() as app:
                     preprocess_output = gr.Textbox(
                         label="Preprocessing output", lines=5
                     )
-                with gr.Column():
-                    preprocess_btn = gr.Button(
-                        value="Start New Experiment", variant="primary"
-                    )
+
+                preprocess_btn = gr.Button(
+                    value="Start New Experiment", variant="primary"
+                )
 
             with gr.Row():
                 restore_zip_file = gr.File(
@@ -327,6 +380,26 @@ with gr.Blocks() as app:
                 )
                 download_expdir_output = gr.File(label="Download experiment directory")
 
+        with gr.Tab(label="Inference"):
+            with gr.Row():
+                original_audio = gr.Audio(
+                    label="Upload original audio",
+                    type="filepath",
+                    show_download_button=True,
+                )
+                f0add = gr.Slider(
+                    label="F0 add",
+                    minimum=-16,
+                    maximum=16,
+                    step=1,
+                    value=0,
+                )
+                infer_btn = gr.Button(value="Infer", variant="primary")
+            with gr.Row():
+                infer_output = gr.Audio(label="Inferred audio")
+            with gr.Row():
+                merge_output = gr.Audio(label="Merged audio")
+
     preprocess_btn.click(
         fn=preprocess,
         inputs=[zip_file],
@@ -343,6 +416,10 @@ with gr.Blocks() as app:
         fn=train_model,
         inputs=[exp_dir],
         outputs=[latest_model],
+    ).success(
+        fn=train_model,
+        inputs=[exp_dir],
+        outputs=[latest_model],
     )
 
     train_index_btn.click(
@@ -369,4 +446,14 @@ with gr.Blocks() as app:
         outputs=[exp_dir],
     )
 
+    infer_btn.click(
+        fn=infer,
+        inputs=[exp_dir, original_audio, f0add],
+        outputs=[infer_output],
+    ).success(
+        fn=merge,
+        inputs=[exp_dir, original_audio, infer_output],
+        outputs=[merge_output],
+    )
+
     app.launch()

assets/pretrained_v2/D40k.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:471378e894e7191f89a94eda8288c5947b16bbe0b10c3f1f17efdb7a1d998242
-size 142875703

assets/pretrained_v2/G40k.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a3843da7fde33db1dab176146c70d6c2df06eafe9457f4e3aa10024e9c6a4b69
-size 72959671

config.json

@@ -67,7 +67,7 @@
         "c_mel": 45,
         "epochs": 20000,
         "eps": 1e-09,
-        "fp16_run": false,
+        "fp16_run": true,
         "init_lr_ratio": 1,
         "learning_rate": 0.0001,
         "log_interval": 200,

configs/config.py

@@ -0,0 +1,245 @@
+import argparse
+import os
+import sys
+import json
+import shutil
+from multiprocessing import cpu_count
+
+import torch
+import logging
+from model import device, fp16
+
+logger = logging.getLogger(__name__)
+
+
+version_config_list = [
+    "v1/32k.json",
+    "v1/40k.json",
+    "v1/48k.json",
+    "v2/48k.json",
+    "v2/32k.json",
+]
+
+
+def singleton_variable(func):
+    def wrapper(*args, **kwargs):
+        if not wrapper.instance:
+            wrapper.instance = func(*args, **kwargs)
+        return wrapper.instance
+
+    wrapper.instance = None
+    return wrapper
+
+
+@singleton_variable
+class Config:
+    def __init__(self):
+        self.device = str(device)
+        self.is_half = fp16
+        self.use_jit = False
+        self.n_cpu = 0
+        self.gpu_name = None
+        self.json_config = self.load_config_json()
+        self.gpu_mem = None
+        (
+            self.python_cmd,
+            self.listen_port,
+            self.iscolab,
+            self.noparallel,
+            self.noautoopen,
+            self.dml,
+        ) = self.arg_parse()
+        self.instead = ""
+        self.preprocess_per = 3.7
+        self.x_pad, self.x_query, self.x_center, self.x_max = self.device_config()
+
+    @staticmethod
+    def load_config_json() -> dict:
+        d = {}
+        # for config_file in version_config_list:
+        #     p = f"configs/inuse/{config_file}"
+        #     if not os.path.exists(p):
+        #         shutil.copy(f"configs/{config_file}", p)
+        #     with open(f"configs/inuse/{config_file}", "r") as f:
+        #         d[config_file] = json.load(f)
+        return d
+
+    @staticmethod
+    def arg_parse() -> tuple:
+        exe = sys.executable or "python"
+        parser = argparse.ArgumentParser()
+        parser.add_argument("--port", type=int, default=7865, help="Listen port")
+        parser.add_argument("--pycmd", type=str, default=exe, help="Python command")
+        parser.add_argument("--colab", action="store_true", help="Launch in colab")
+        parser.add_argument(
+            "--noparallel", action="store_true", help="Disable parallel processing"
+        )
+        parser.add_argument(
+            "--noautoopen",
+            action="store_true",
+            help="Do not open in browser automatically",
+        )
+        parser.add_argument(
+            "--dml",
+            action="store_true",
+            help="torch_dml",
+        )
+        cmd_opts = parser.parse_args()
+
+        cmd_opts.port = cmd_opts.port if 0 <= cmd_opts.port <= 65535 else 7865
+
+        return (
+            cmd_opts.pycmd,
+            cmd_opts.port,
+            cmd_opts.colab,
+            cmd_opts.noparallel,
+            cmd_opts.noautoopen,
+            cmd_opts.dml,
+        )
+
+    # has_mps is only available in nightly pytorch (for now) and MasOS 12.3+.
+    # check `getattr` and try it for compatibility
+    @staticmethod
+    def has_mps() -> bool:
+        if not torch.backends.mps.is_available():
+            return False
+        try:
+            torch.zeros(1).to(torch.device("mps"))
+            return True
+        except Exception:
+            return False
+
+    @staticmethod
+    def has_xpu() -> bool:
+        if hasattr(torch, "xpu") and torch.xpu.is_available():
+            return True
+        else:
+            return False
+
+    def use_fp32_config(self):
+        for config_file in version_config_list:
+            self.json_config[config_file]["train"]["fp16_run"] = False
+            with open(f"configs/inuse/{config_file}", "r") as f:
+                strr = f.read().replace("true", "false")
+            with open(f"configs/inuse/{config_file}", "w") as f:
+                f.write(strr)
+            logger.info("overwrite " + config_file)
+        self.preprocess_per = 3.0
+        logger.info("overwrite preprocess_per to %d" % (self.preprocess_per))
+
+    def device_config(self) -> tuple:
+        if torch.cuda.is_available():
+            if self.has_xpu():
+                self.device = self.instead = "xpu:0"
+                self.is_half = True
+            i_device = int(self.device.split(":")[-1])
+            self.gpu_name = torch.cuda.get_device_name(i_device)
+            if (
+                ("16" in self.gpu_name and "V100" not in self.gpu_name.upper())
+                or "P40" in self.gpu_name.upper()
+                or "P10" in self.gpu_name.upper()
+                or "1060" in self.gpu_name
+                or "1070" in self.gpu_name
+                or "1080" in self.gpu_name
+            ):
+                logger.info("Found GPU %s, force to fp32", self.gpu_name)
+                self.is_half = False
+                self.use_fp32_config()
+            else:
+                logger.info("Found GPU %s", self.gpu_name)
+            self.gpu_mem = int(
+                torch.cuda.get_device_properties(i_device).total_memory
+                / 1024
+                / 1024
+                / 1024
+                + 0.4
+            )
+            if self.gpu_mem <= 4:
+                self.preprocess_per = 3.0
+        elif self.has_mps():
+            logger.info("No supported Nvidia GPU found")
+            self.device = self.instead = "mps"
+            self.is_half = False
+            self.use_fp32_config()
+        else:
+            logger.info("No supported Nvidia GPU found")
+            self.device = self.instead = "cpu"
+            self.is_half = False
+            self.use_fp32_config()
+
+        if self.n_cpu == 0:
+            self.n_cpu = cpu_count()
+
+        if self.is_half:
+            # 6G显存配置
+            x_pad = 3
+            x_query = 10
+            x_center = 60
+            x_max = 65
+        else:
+            # 5G显存配置
+            x_pad = 1
+            x_query = 6
+            x_center = 38
+            x_max = 41
+
+        if self.gpu_mem is not None and self.gpu_mem <= 4:
+            x_pad = 1
+            x_query = 5
+            x_center = 30
+            x_max = 32
+        if self.dml:
+            logger.info("Use DirectML instead")
+            if (
+                os.path.exists(
+                    "runtime\Lib\site-packages\onnxruntime\capi\DirectML.dll"
+                )
+                == False
+            ):
+                try:
+                    os.rename(
+                        "runtime\Lib\site-packages\onnxruntime",
+                        "runtime\Lib\site-packages\onnxruntime-cuda",
+                    )
+                except:
+                    pass
+                try:
+                    os.rename(
+                        "runtime\Lib\site-packages\onnxruntime-dml",
+                        "runtime\Lib\site-packages\onnxruntime",
+                    )
+                except:
+                    pass
+            # if self.device != "cpu":
+            import torch_directml
+
+            self.device = torch_directml.device(torch_directml.default_device())
+            self.is_half = False
+        else:
+            if self.instead:
+                logger.info(f"Use {self.instead} instead")
+            if (
+                os.path.exists(
+                    "runtime\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"
+                )
+                == False
+            ):
+                try:
+                    os.rename(
+                        "runtime\Lib\site-packages\onnxruntime",
+                        "runtime\Lib\site-packages\onnxruntime-dml",
+                    )
+                except:
+                    pass
+                try:
+                    os.rename(
+                        "runtime\Lib\site-packages\onnxruntime-cuda",
+                        "runtime\Lib\site-packages\onnxruntime",
+                    )
+                except:
+                    pass
+        logger.info(
+            "Half-precision floating-point: %s, device: %s"
+            % (self.is_half, self.device)
+        )
+        return x_pad, x_query, x_center, x_max

infer/lib/audio.py

@@ -1,8 +1,8 @@
 import platform, os
+import traceback
 import ffmpeg
 import numpy as np
 import av
-from io import BytesIO
 
 
 def wav2(i, o, format):

infer/lib/rmvpe.py

@@ -1,24 +1,14 @@
 from io import BytesIO
 import os
-from typing import List, Optional, Tuple
+from typing import List
 import numpy as np
 import torch
 
 from infer.lib import jit
 
-try:
-    # Fix "Torch not compiled with CUDA enabled"
-    import intel_extension_for_pytorch as ipex  # pylint: disable=import-error, unused-import
-
-    if torch.xpu.is_available():
-        from infer.modules.ipex import ipex_init
-
-        ipex_init()
-except Exception:  # pylint: disable=broad-exception-caught
-    pass
 import torch.nn as nn
 import torch.nn.functional as F
-from librosa.util import normalize, pad_center, tiny
+from librosa.util import pad_center
 from scipy.signal import get_window
 
 import logging

infer/lib/train/process_ckpt.py

@@ -61,7 +61,7 @@ def show_info(path):
         return traceback.format_exc()
 
 
-def extract_small_model(path, name, sr, if_f0, info, version):
+def extract_small_model(path, out, sr, if_f0, info, version):
     try:
         ckpt = torch.load(path, map_location="cpu")
         if "model" in ckpt:
@@ -185,7 +185,7 @@ def extract_small_model(path, name, sr, if_f0, info, version):
         opt["version"] = version
         opt["sr"] = sr
         opt["f0"] = int(if_f0)
-        torch.save(opt, "assets/weights/%s.pth" % name)
+        torch.save(opt, out)
         return "Success."
     except:
         return traceback.format_exc()

infer/lib/uvr5_pack/lib_v5/dataset.py

@@ -1,183 +0,0 @@
-import os
-import random
-
-import numpy as np
-import torch
-import torch.utils.data
-from tqdm import tqdm
-
-from . import spec_utils
-
-
-class VocalRemoverValidationSet(torch.utils.data.Dataset):
-    def __init__(self, patch_list):
-        self.patch_list = patch_list
-
-    def __len__(self):
-        return len(self.patch_list)
-
-    def __getitem__(self, idx):
-        path = self.patch_list[idx]
-        data = np.load(path)
-
-        X, y = data["X"], data["y"]
-
-        X_mag = np.abs(X)
-        y_mag = np.abs(y)
-
-        return X_mag, y_mag
-
-
-def make_pair(mix_dir, inst_dir):
-    input_exts = [".wav", ".m4a", ".mp3", ".mp4", ".flac"]
-
-    X_list = sorted(
-        [
-            os.path.join(mix_dir, fname)
-            for fname in os.listdir(mix_dir)
-            if os.path.splitext(fname)[1] in input_exts
-        ]
-    )
-    y_list = sorted(
-        [
-            os.path.join(inst_dir, fname)
-            for fname in os.listdir(inst_dir)
-            if os.path.splitext(fname)[1] in input_exts
-        ]
-    )
-
-    filelist = list(zip(X_list, y_list))
-
-    return filelist
-
-
-def train_val_split(dataset_dir, split_mode, val_rate, val_filelist):
-    if split_mode == "random":
-        filelist = make_pair(
-            os.path.join(dataset_dir, "mixtures"),
-            os.path.join(dataset_dir, "instruments"),
-        )
-
-        random.shuffle(filelist)
-
-        if len(val_filelist) == 0:
-            val_size = int(len(filelist) * val_rate)
-            train_filelist = filelist[:-val_size]
-            val_filelist = filelist[-val_size:]
-        else:
-            train_filelist = [
-                pair for pair in filelist if list(pair) not in val_filelist
-            ]
-    elif split_mode == "subdirs":
-        if len(val_filelist) != 0:
-            raise ValueError(
-                "The `val_filelist` option is not available in `subdirs` mode"
-            )
-
-        train_filelist = make_pair(
-            os.path.join(dataset_dir, "training/mixtures"),
-            os.path.join(dataset_dir, "training/instruments"),
-        )
-
-        val_filelist = make_pair(
-            os.path.join(dataset_dir, "validation/mixtures"),
-            os.path.join(dataset_dir, "validation/instruments"),
-        )
-
-    return train_filelist, val_filelist
-
-
-def augment(X, y, reduction_rate, reduction_mask, mixup_rate, mixup_alpha):
-    perm = np.random.permutation(len(X))
-    for i, idx in enumerate(tqdm(perm)):
-        if np.random.uniform() < reduction_rate:
-            y[idx] = spec_utils.reduce_vocal_aggressively(
-                X[idx], y[idx], reduction_mask
-            )
-
-        if np.random.uniform() < 0.5:
-            # swap channel
-            X[idx] = X[idx, ::-1]
-            y[idx] = y[idx, ::-1]
-        if np.random.uniform() < 0.02:
-            # mono
-            X[idx] = X[idx].mean(axis=0, keepdims=True)
-            y[idx] = y[idx].mean(axis=0, keepdims=True)
-        if np.random.uniform() < 0.02:
-            # inst
-            X[idx] = y[idx]
-
-        if np.random.uniform() < mixup_rate and i < len(perm) - 1:
-            lam = np.random.beta(mixup_alpha, mixup_alpha)
-            X[idx] = lam * X[idx] + (1 - lam) * X[perm[i + 1]]
-            y[idx] = lam * y[idx] + (1 - lam) * y[perm[i + 1]]
-
-    return X, y
-
-
-def make_padding(width, cropsize, offset):
-    left = offset
-    roi_size = cropsize - left * 2
-    if roi_size == 0:
-        roi_size = cropsize
-    right = roi_size - (width % roi_size) + left
-
-    return left, right, roi_size
-
-
-def make_training_set(filelist, cropsize, patches, sr, hop_length, n_fft, offset):
-    len_dataset = patches * len(filelist)
-
-    X_dataset = np.zeros((len_dataset, 2, n_fft // 2 + 1, cropsize), dtype=np.complex64)
-    y_dataset = np.zeros((len_dataset, 2, n_fft // 2 + 1, cropsize), dtype=np.complex64)
-
-    for i, (X_path, y_path) in enumerate(tqdm(filelist)):
-        X, y = spec_utils.cache_or_load(X_path, y_path, sr, hop_length, n_fft)
-        coef = np.max([np.abs(X).max(), np.abs(y).max()])
-        X, y = X / coef, y / coef
-
-        l, r, roi_size = make_padding(X.shape[2], cropsize, offset)
-        X_pad = np.pad(X, ((0, 0), (0, 0), (l, r)), mode="constant")
-        y_pad = np.pad(y, ((0, 0), (0, 0), (l, r)), mode="constant")
-
-        starts = np.random.randint(0, X_pad.shape[2] - cropsize, patches)
-        ends = starts + cropsize
-        for j in range(patches):
-            idx = i * patches + j
-            X_dataset[idx] = X_pad[:, :, starts[j] : ends[j]]
-            y_dataset[idx] = y_pad[:, :, starts[j] : ends[j]]
-
-    return X_dataset, y_dataset
-
-
-def make_validation_set(filelist, cropsize, sr, hop_length, n_fft, offset):
-    patch_list = []
-    patch_dir = "cs{}_sr{}_hl{}_nf{}_of{}".format(
-        cropsize, sr, hop_length, n_fft, offset
-    )
-    os.makedirs(patch_dir, exist_ok=True)
-
-    for i, (X_path, y_path) in enumerate(tqdm(filelist)):
-        basename = os.path.splitext(os.path.basename(X_path))[0]
-
-        X, y = spec_utils.cache_or_load(X_path, y_path, sr, hop_length, n_fft)
-        coef = np.max([np.abs(X).max(), np.abs(y).max()])
-        X, y = X / coef, y / coef
-
-        l, r, roi_size = make_padding(X.shape[2], cropsize, offset)
-        X_pad = np.pad(X, ((0, 0), (0, 0), (l, r)), mode="constant")
-        y_pad = np.pad(y, ((0, 0), (0, 0), (l, r)), mode="constant")
-
-        len_dataset = int(np.ceil(X.shape[2] / roi_size))
-        for j in range(len_dataset):
-            outpath = os.path.join(patch_dir, "{}_p{}.npz".format(basename, j))
-            start = j * roi_size
-            if not os.path.exists(outpath):
-                np.savez(
-                    outpath,
-                    X=X_pad[:, :, start : start + cropsize],
-                    y=y_pad[:, :, start : start + cropsize],
-                )
-            patch_list.append(outpath)
-
-    return VocalRemoverValidationSet(patch_list)

infer/lib/uvr5_pack/lib_v5/layers.py

@@ -1,118 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 5, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_123812KB .py

@@ -1,118 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 5, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_123821KB.py

@@ -1,118 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 5, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_33966KB.py

@@ -1,126 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16, 32, 64), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv6 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv7 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 7, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        feat6 = self.conv6(x)
-        feat7 = self.conv7(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5, feat6, feat7), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_537227KB.py

@@ -1,126 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16, 32, 64), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv6 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv7 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 7, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        feat6 = self.conv6(x)
-        feat7 = self.conv7(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5, feat6, feat7), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_537238KB.py

@@ -1,126 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class SeperableConv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(SeperableConv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nin,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                groups=nin,
-                bias=False,
-            ),
-            nn.Conv2d(nin, nout, kernel_size=1, bias=False),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, stride, pad, activ=activ)
-
-    def __call__(self, x):
-        skip = self.conv1(x)
-        h = self.conv2(skip)
-
-        return h, skip
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-        h = self.conv(x)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 16, 32, 64), activ=nn.ReLU):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nin, 1, 1, 0, activ=activ)
-        self.conv3 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv6 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.conv7 = SeperableConv2DBNActiv(
-            nin, nin, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = nn.Sequential(
-            Conv2DBNActiv(nin * 7, nout, 1, 1, 0, activ=activ), nn.Dropout2d(0.1)
-        )
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        feat6 = self.conv6(x)
-        feat7 = self.conv7(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5, feat6, feat7), dim=1)
-        bottle = self.bottleneck(out)
-        return bottle

infer/lib/uvr5_pack/lib_v5/layers_new.py

@@ -1,125 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class Conv2DBNActiv(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, dilation=1, activ=nn.ReLU):
-        super(Conv2DBNActiv, self).__init__()
-        self.conv = nn.Sequential(
-            nn.Conv2d(
-                nin,
-                nout,
-                kernel_size=ksize,
-                stride=stride,
-                padding=pad,
-                dilation=dilation,
-                bias=False,
-            ),
-            nn.BatchNorm2d(nout),
-            activ(),
-        )
-
-    def __call__(self, x):
-        return self.conv(x)
-
-
-class Encoder(nn.Module):
-    def __init__(self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.LeakyReLU):
-        super(Encoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, stride, pad, activ=activ)
-        self.conv2 = Conv2DBNActiv(nout, nout, ksize, 1, pad, activ=activ)
-
-    def __call__(self, x):
-        h = self.conv1(x)
-        h = self.conv2(h)
-
-        return h
-
-
-class Decoder(nn.Module):
-    def __init__(
-        self, nin, nout, ksize=3, stride=1, pad=1, activ=nn.ReLU, dropout=False
-    ):
-        super(Decoder, self).__init__()
-        self.conv1 = Conv2DBNActiv(nin, nout, ksize, 1, pad, activ=activ)
-        # self.conv2 = Conv2DBNActiv(nout, nout, ksize, 1, pad, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def __call__(self, x, skip=None):
-        x = F.interpolate(x, scale_factor=2, mode="bilinear", align_corners=True)
-
-        if skip is not None:
-            skip = spec_utils.crop_center(skip, x)
-            x = torch.cat([x, skip], dim=1)
-
-        h = self.conv1(x)
-        # h = self.conv2(h)
-
-        if self.dropout is not None:
-            h = self.dropout(h)
-
-        return h
-
-
-class ASPPModule(nn.Module):
-    def __init__(self, nin, nout, dilations=(4, 8, 12), activ=nn.ReLU, dropout=False):
-        super(ASPPModule, self).__init__()
-        self.conv1 = nn.Sequential(
-            nn.AdaptiveAvgPool2d((1, None)),
-            Conv2DBNActiv(nin, nout, 1, 1, 0, activ=activ),
-        )
-        self.conv2 = Conv2DBNActiv(nin, nout, 1, 1, 0, activ=activ)
-        self.conv3 = Conv2DBNActiv(
-            nin, nout, 3, 1, dilations[0], dilations[0], activ=activ
-        )
-        self.conv4 = Conv2DBNActiv(
-            nin, nout, 3, 1, dilations[1], dilations[1], activ=activ
-        )
-        self.conv5 = Conv2DBNActiv(
-            nin, nout, 3, 1, dilations[2], dilations[2], activ=activ
-        )
-        self.bottleneck = Conv2DBNActiv(nout * 5, nout, 1, 1, 0, activ=activ)
-        self.dropout = nn.Dropout2d(0.1) if dropout else None
-
-    def forward(self, x):
-        _, _, h, w = x.size()
-        feat1 = F.interpolate(
-            self.conv1(x), size=(h, w), mode="bilinear", align_corners=True
-        )
-        feat2 = self.conv2(x)
-        feat3 = self.conv3(x)
-        feat4 = self.conv4(x)
-        feat5 = self.conv5(x)
-        out = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1)
-        out = self.bottleneck(out)
-
-        if self.dropout is not None:
-            out = self.dropout(out)
-
-        return out
-
-
-class LSTMModule(nn.Module):
-    def __init__(self, nin_conv, nin_lstm, nout_lstm):
-        super(LSTMModule, self).__init__()
-        self.conv = Conv2DBNActiv(nin_conv, 1, 1, 1, 0)
-        self.lstm = nn.LSTM(
-            input_size=nin_lstm, hidden_size=nout_lstm // 2, bidirectional=True
-        )
-        self.dense = nn.Sequential(
-            nn.Linear(nout_lstm, nin_lstm), nn.BatchNorm1d(nin_lstm), nn.ReLU()
-        )
-
-    def forward(self, x):
-        N, _, nbins, nframes = x.size()
-        h = self.conv(x)[:, 0]  # N, nbins, nframes
-        h = h.permute(2, 0, 1)  # nframes, N, nbins
-        h, _ = self.lstm(h)
-        h = self.dense(h.reshape(-1, h.size()[-1]))  # nframes * N, nbins
-        h = h.reshape(nframes, N, 1, nbins)
-        h = h.permute(1, 2, 3, 0)
-
-        return h

infer/lib/uvr5_pack/lib_v5/model_param_init.py

@@ -1,69 +0,0 @@
-import json
-import os
-import pathlib
-
-default_param = {}
-default_param["bins"] = 768
-default_param["unstable_bins"] = 9  # training only
-default_param["reduction_bins"] = 762  # training only
-default_param["sr"] = 44100
-default_param["pre_filter_start"] = 757
-default_param["pre_filter_stop"] = 768
-default_param["band"] = {}
-
-
-default_param["band"][1] = {
-    "sr": 11025,
-    "hl": 128,
-    "n_fft": 960,
-    "crop_start": 0,
-    "crop_stop": 245,
-    "lpf_start": 61,  # inference only
-    "res_type": "polyphase",
-}
-
-default_param["band"][2] = {
-    "sr": 44100,
-    "hl": 512,
-    "n_fft": 1536,
-    "crop_start": 24,
-    "crop_stop": 547,
-    "hpf_start": 81,  # inference only
-    "res_type": "sinc_best",
-}
-
-
-def int_keys(d):
-    r = {}
-    for k, v in d:
-        if k.isdigit():
-            k = int(k)
-        r[k] = v
-    return r
-
-
-class ModelParameters(object):
-    def __init__(self, config_path=""):
-        if ".pth" == pathlib.Path(config_path).suffix:
-            import zipfile
-
-            with zipfile.ZipFile(config_path, "r") as zip:
-                self.param = json.loads(
-                    zip.read("param.json"), object_pairs_hook=int_keys
-                )
-        elif ".json" == pathlib.Path(config_path).suffix:
-            with open(config_path, "r") as f:
-                self.param = json.loads(f.read(), object_pairs_hook=int_keys)
-        else:
-            self.param = default_param
-
-        for k in [
-            "mid_side",
-            "mid_side_b",
-            "mid_side_b2",
-            "stereo_w",
-            "stereo_n",
-            "reverse",
-        ]:
-            if not k in self.param:
-                self.param[k] = False

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr16000_hl512.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 16000,
-			"hl": 512,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 1024,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 16000,
-	"pre_filter_start": 1023,
-	"pre_filter_stop": 1024
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr32000_hl512.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 32000,
-			"hl": 512,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 1024,
-			"hpf_start": -1,
-			"res_type": "kaiser_fast"
-		}
-	},
-	"sr": 32000,
-	"pre_filter_start": 1000,
-	"pre_filter_stop": 1021
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr33075_hl384.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 33075,
-			"hl": 384,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 1024,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 33075,
-	"pre_filter_start": 1000,
-	"pre_filter_stop": 1021
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr44100_hl1024.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 44100,
-			"hl": 1024,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 1024,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 1023,
-	"pre_filter_stop": 1024
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr44100_hl256.json

@@ -1,19 +0,0 @@
-{
-	"bins": 256,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 44100,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 0,
-			"crop_stop": 256,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 256,
-	"pre_filter_stop": 256
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr44100_hl512.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 1024,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 1023,
-	"pre_filter_stop": 1024
-}

infer/lib/uvr5_pack/lib_v5/modelparams/1band_sr44100_hl512_cut.json

@@ -1,19 +0,0 @@
-{
-	"bins": 1024,
-	"unstable_bins": 0,
-	"reduction_bins": 0,
-	"band": {
-		"1": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 700,
-			"hpf_start": -1,
-			"res_type": "sinc_best"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 1023,
-	"pre_filter_stop": 700
-}

infer/lib/uvr5_pack/lib_v5/modelparams/2band_32000.json

@@ -1,30 +0,0 @@
-{
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 705,
-	"band": {
-		"1": {
-			"sr": 6000,
-			"hl": 66,
-			"n_fft": 512,
-			"crop_start": 0,
-			"crop_stop": 240,
-			"lpf_start": 60,
-			"lpf_stop": 118,
-			"res_type": "sinc_fastest"
-		},
-		"2": {
-			"sr": 32000,
-			"hl": 352,
-			"n_fft": 1024,
-			"crop_start": 22,
-			"crop_stop": 505,
-			"hpf_start": 44,
-			"hpf_stop": 23,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 32000,
-	"pre_filter_start": 710,
-	"pre_filter_stop": 731
-}

infer/lib/uvr5_pack/lib_v5/modelparams/2band_44100_lofi.json

@@ -1,30 +0,0 @@
-{
-	"bins": 512,
-	"unstable_bins": 7,
-	"reduction_bins": 510,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 160,
-			"n_fft": 768,
-			"crop_start": 0,
-			"crop_stop": 192,
-			"lpf_start": 41,
-			"lpf_stop": 139,
-			"res_type": "sinc_fastest"
-		},
-		"2": {
-			"sr": 44100,
-			"hl": 640,
-			"n_fft": 1024,
-			"crop_start": 10,
-			"crop_stop": 320,
-			"hpf_start": 47,
-			"hpf_stop": 15,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 510,
-	"pre_filter_stop": 512
-}

infer/lib/uvr5_pack/lib_v5/modelparams/2band_48000.json

@@ -1,30 +0,0 @@
-{
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 705,
-	"band": {
-		"1": {
-			"sr": 6000,
-			"hl": 66,
-			"n_fft": 512,
-			"crop_start": 0,
-			"crop_stop": 240,
-			"lpf_start": 60,
-			"lpf_stop": 240,
-			"res_type": "sinc_fastest"
-		},
-		"2": {
-			"sr": 48000,
-			"hl": 528,
-			"n_fft": 1536,
-			"crop_start": 22,
-			"crop_stop": 505,
-			"hpf_start": 82,
-			"hpf_stop": 22,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 48000,
-	"pre_filter_start": 710,
-	"pre_filter_stop": 731
-}

infer/lib/uvr5_pack/lib_v5/modelparams/3band_44100.json

@@ -1,42 +0,0 @@
-{
-	"bins": 768,
-	"unstable_bins": 5,
-	"reduction_bins": 733,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 768,
-			"crop_start": 0,
-			"crop_stop": 278,
-			"lpf_start": 28,
-			"lpf_stop": 140,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 768,
-			"crop_start": 14,
-			"crop_stop": 322,
-			"hpf_start": 70,
-			"hpf_stop": 14,
-			"lpf_start": 283,
-			"lpf_stop": 314,
-			"res_type": "polyphase"
-		},	
-		"3": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 131,
-			"crop_stop": 313,
-			"hpf_start": 154,
-			"hpf_stop": 141,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 757,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/3band_44100_mid.json

@@ -1,43 +0,0 @@
-{
-	"mid_side": true,
-	"bins": 768,
-	"unstable_bins": 5,
-	"reduction_bins": 733,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 768,
-			"crop_start": 0,
-			"crop_stop": 278,
-			"lpf_start": 28,
-			"lpf_stop": 140,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 768,
-			"crop_start": 14,
-			"crop_stop": 322,
-			"hpf_start": 70,
-			"hpf_stop": 14,
-			"lpf_start": 283,
-			"lpf_stop": 314,
-			"res_type": "polyphase"
-		},	
-		"3": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 131,
-			"crop_stop": 313,
-			"hpf_start": 154,
-			"hpf_stop": 141,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 757,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/3band_44100_msb2.json

@@ -1,43 +0,0 @@
-{
-	"mid_side_b2": true,
-	"bins": 640,
-	"unstable_bins": 7,
-	"reduction_bins": 565,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 108,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 187,
-			"lpf_start": 92,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 22050,
-			"hl": 216,
-			"n_fft": 768,
-			"crop_start": 0,
-			"crop_stop": 212,
-			"hpf_start": 68,
-			"hpf_stop": 34,
-			"lpf_start": 174,
-			"lpf_stop": 209,
-			"res_type": "polyphase"
-		},	
-		"3": {
-			"sr": 44100,
-			"hl": 432,
-			"n_fft": 640,
-			"crop_start": 66,
-			"crop_stop": 307,
-			"hpf_start": 86,
-			"hpf_stop": 72,
-			"res_type": "kaiser_fast"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 639,
-	"pre_filter_stop": 640
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100.json

@@ -1,54 +0,0 @@
-{
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100_mid.json

@@ -1,55 +0,0 @@
-{
-	"bins": 768,
-	"unstable_bins": 7,
-	"mid_side": true,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100_msb.json

@@ -1,55 +0,0 @@
-{
-	"mid_side_b": true,
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100_msb2.json

@@ -1,55 +0,0 @@
-{
-	"mid_side_b": true,
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100_reverse.json

@@ -1,55 +0,0 @@
-{
-	"reverse": true,
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_44100_sw.json

@@ -1,55 +0,0 @@
-{
-	"stereo_w": true,
-	"bins": 768,
-	"unstable_bins": 7,
-	"reduction_bins": 668,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 1024,
-			"crop_start": 0,
-			"crop_stop": 186,
-			"lpf_start": 37,
-			"lpf_stop": 73,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 11025,
-			"hl": 128,
-			"n_fft": 512,
-			"crop_start": 4,
-			"crop_stop": 185,			
-			"hpf_start": 36,
-			"hpf_stop": 18,
-			"lpf_start": 93,
-			"lpf_stop": 185,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 22050,
-			"hl": 256,
-			"n_fft": 512,
-			"crop_start": 46,
-			"crop_stop": 186,
-			"hpf_start": 93,
-			"hpf_stop": 46,
-			"lpf_start": 164,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 512,
-			"n_fft": 768,
-			"crop_start": 121,
-			"crop_stop": 382,
-			"hpf_start": 138,
-			"hpf_stop": 123,
-			"res_type": "sinc_medium"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 740,
-	"pre_filter_stop": 768
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_v2.json

@@ -1,54 +0,0 @@
-{
-	"bins": 672,
-	"unstable_bins": 8,
-	"reduction_bins": 637,
-	"band": {
-		"1": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 640,
-			"crop_start": 0,
-			"crop_stop": 85,
-			"lpf_start": 25,
-			"lpf_stop": 53,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 320,
-			"crop_start": 4,
-			"crop_stop": 87,
-			"hpf_start": 25,
-			"hpf_stop": 12,
-			"lpf_start": 31,
-			"lpf_stop": 62,
-			"res_type": "polyphase"
-		},		
-		"3": {
-			"sr": 14700,
-			"hl": 160,
-			"n_fft": 512,
-			"crop_start": 17,
-			"crop_stop": 216,
-			"hpf_start": 48,
-			"hpf_stop": 24,
-			"lpf_start": 139,
-			"lpf_stop": 210,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 480,
-			"n_fft": 960,
-			"crop_start": 78,
-			"crop_stop": 383,
-			"hpf_start": 130,
-			"hpf_stop": 86,
-			"res_type": "kaiser_fast"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 668,
-	"pre_filter_stop": 672
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_v2_sn.json

@@ -1,55 +0,0 @@
-{
-	"bins": 672,
-	"unstable_bins": 8,
-	"reduction_bins": 637,
-	"band": {
-		"1": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 640,
-			"crop_start": 0,
-			"crop_stop": 85,
-			"lpf_start": 25,
-			"lpf_stop": 53,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 320,
-			"crop_start": 4,
-			"crop_stop": 87,
-			"hpf_start": 25,
-			"hpf_stop": 12,
-			"lpf_start": 31,
-			"lpf_stop": 62,
-			"res_type": "polyphase"
-		},		
-		"3": {
-			"sr": 14700,
-			"hl": 160,
-			"n_fft": 512,
-			"crop_start": 17,
-			"crop_stop": 216,
-			"hpf_start": 48,
-			"hpf_stop": 24,
-			"lpf_start": 139,
-			"lpf_stop": 210,
-			"res_type": "polyphase"
-		},	
-		"4": {
-			"sr": 44100,
-			"hl": 480,
-			"n_fft": 960,
-			"crop_start": 78,
-			"crop_stop": 383,
-			"hpf_start": 130,
-			"hpf_stop": 86,
-			"convert_channels": "stereo_n",
-			"res_type": "kaiser_fast"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 668,
-	"pre_filter_stop": 672
-}

infer/lib/uvr5_pack/lib_v5/modelparams/4band_v3.json

@@ -1,54 +0,0 @@
-{
-	"bins": 672,
-	"unstable_bins": 8,
-	"reduction_bins": 530,
-	"band": {
-		"1": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 640,
-			"crop_start": 0,
-			"crop_stop": 85,
-			"lpf_start": 25,
-			"lpf_stop": 53,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 7350,
-			"hl": 80,
-			"n_fft": 320,
-			"crop_start": 4,
-			"crop_stop": 87,
-			"hpf_start": 25,
-			"hpf_stop": 12,
-			"lpf_start": 31,
-			"lpf_stop": 62,
-			"res_type": "polyphase"
-		},
-		"3": {
-			"sr": 14700,
-			"hl": 160,
-			"n_fft": 512,
-			"crop_start": 17,
-			"crop_stop": 216,
-			"hpf_start": 48,
-			"hpf_stop": 24,
-			"lpf_start": 139,
-			"lpf_stop": 210,
-			"res_type": "polyphase"
-		},
-		"4": {
-			"sr": 44100,
-			"hl": 480,
-			"n_fft": 960,
-			"crop_start": 78,
-			"crop_stop": 383,
-			"hpf_start": 130,
-			"hpf_stop": 86,
-			"res_type": "kaiser_fast"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 668,
-	"pre_filter_stop": 672
-}

infer/lib/uvr5_pack/lib_v5/modelparams/ensemble.json

@@ -1,43 +0,0 @@
-{
-	"mid_side_b2": true,
-	"bins": 1280,
-	"unstable_bins": 7,
-	"reduction_bins": 565,
-	"band": {
-		"1": {
-			"sr": 11025,
-			"hl": 108,
-			"n_fft": 2048,
-			"crop_start": 0,
-			"crop_stop": 374,
-			"lpf_start": 92,
-			"lpf_stop": 186,
-			"res_type": "polyphase"
-		},
-		"2": {
-			"sr": 22050,
-			"hl": 216,
-			"n_fft": 1536,
-			"crop_start": 0,
-			"crop_stop": 424,
-			"hpf_start": 68,
-			"hpf_stop": 34,
-			"lpf_start": 348,
-			"lpf_stop": 418,
-			"res_type": "polyphase"
-		},	
-		"3": {
-			"sr": 44100,
-			"hl": 432,
-			"n_fft": 1280,
-			"crop_start": 132,
-			"crop_stop": 614,
-			"hpf_start": 172,
-			"hpf_stop": 144,
-			"res_type": "polyphase"
-		}
-	},
-	"sr": 44100,
-	"pre_filter_start": 1280,
-	"pre_filter_stop": 1280
-}

infer/lib/uvr5_pack/lib_v5/nets.py

@@ -1,123 +0,0 @@
-import layers
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import spec_utils
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 16)
-        self.stg1_high_band_net = BaseASPPNet(2, 16)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(18, 8, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(8, 16)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(34, 16, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(16, 32)
-
-        self.out = nn.Conv2d(32, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(16, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(16, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_123812KB.py

@@ -1,122 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_123821KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 32)
-        self.stg1_high_band_net = BaseASPPNet(2, 32)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(34, 16, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(16, 32)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(66, 32, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(32, 64)
-
-        self.out = nn.Conv2d(64, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(32, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(32, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_123821KB.py

@@ -1,122 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_123821KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 32)
-        self.stg1_high_band_net = BaseASPPNet(2, 32)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(34, 16, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(16, 32)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(66, 32, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(32, 64)
-
-        self.out = nn.Conv2d(64, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(32, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(32, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_33966KB.py

@@ -1,122 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_33966KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16, 32)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 16)
-        self.stg1_high_band_net = BaseASPPNet(2, 16)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(18, 8, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(8, 16)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(34, 16, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(16, 32)
-
-        self.out = nn.Conv2d(32, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(16, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(16, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_537227KB.py

@@ -1,123 +0,0 @@
-import numpy as np
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_537238KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 64)
-        self.stg1_high_band_net = BaseASPPNet(2, 64)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(66, 32, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(32, 64)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(130, 64, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(64, 128)
-
-        self.out = nn.Conv2d(128, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(64, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(64, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_537238KB.py

@@ -1,123 +0,0 @@
-import numpy as np
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_537238KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 64)
-        self.stg1_high_band_net = BaseASPPNet(2, 64)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(66, 32, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(32, 64)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(130, 64, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(64, 128)
-
-        self.out = nn.Conv2d(128, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(64, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(64, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_61968KB.py

@@ -1,122 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_123821KB as layers
-
-
-class BaseASPPNet(nn.Module):
-    def __init__(self, nin, ch, dilations=(4, 8, 16)):
-        super(BaseASPPNet, self).__init__()
-        self.enc1 = layers.Encoder(nin, ch, 3, 2, 1)
-        self.enc2 = layers.Encoder(ch, ch * 2, 3, 2, 1)
-        self.enc3 = layers.Encoder(ch * 2, ch * 4, 3, 2, 1)
-        self.enc4 = layers.Encoder(ch * 4, ch * 8, 3, 2, 1)
-
-        self.aspp = layers.ASPPModule(ch * 8, ch * 16, dilations)
-
-        self.dec4 = layers.Decoder(ch * (8 + 16), ch * 8, 3, 1, 1)
-        self.dec3 = layers.Decoder(ch * (4 + 8), ch * 4, 3, 1, 1)
-        self.dec2 = layers.Decoder(ch * (2 + 4), ch * 2, 3, 1, 1)
-        self.dec1 = layers.Decoder(ch * (1 + 2), ch, 3, 1, 1)
-
-    def __call__(self, x):
-        h, e1 = self.enc1(x)
-        h, e2 = self.enc2(h)
-        h, e3 = self.enc3(h)
-        h, e4 = self.enc4(h)
-
-        h = self.aspp(h)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedASPPNet(nn.Module):
-    def __init__(self, n_fft):
-        super(CascadedASPPNet, self).__init__()
-        self.stg1_low_band_net = BaseASPPNet(2, 32)
-        self.stg1_high_band_net = BaseASPPNet(2, 32)
-
-        self.stg2_bridge = layers.Conv2DBNActiv(34, 16, 1, 1, 0)
-        self.stg2_full_band_net = BaseASPPNet(16, 32)
-
-        self.stg3_bridge = layers.Conv2DBNActiv(66, 32, 1, 1, 0)
-        self.stg3_full_band_net = BaseASPPNet(32, 64)
-
-        self.out = nn.Conv2d(64, 2, 1, bias=False)
-        self.aux1_out = nn.Conv2d(32, 2, 1, bias=False)
-        self.aux2_out = nn.Conv2d(32, 2, 1, bias=False)
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-
-        self.offset = 128
-
-    def forward(self, x, aggressiveness=None):
-        mix = x.detach()
-        x = x.clone()
-
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        aux1 = torch.cat(
-            [
-                self.stg1_low_band_net(x[:, :, :bandw]),
-                self.stg1_high_band_net(x[:, :, bandw:]),
-            ],
-            dim=2,
-        )
-
-        h = torch.cat([x, aux1], dim=1)
-        aux2 = self.stg2_full_band_net(self.stg2_bridge(h))
-
-        h = torch.cat([x, aux1, aux2], dim=1)
-        h = self.stg3_full_band_net(self.stg3_bridge(h))
-
-        mask = torch.sigmoid(self.out(h))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux1 = torch.sigmoid(self.aux1_out(aux1))
-            aux1 = F.pad(
-                input=aux1,
-                pad=(0, 0, 0, self.output_bin - aux1.size()[2]),
-                mode="replicate",
-            )
-            aux2 = torch.sigmoid(self.aux2_out(aux2))
-            aux2 = F.pad(
-                input=aux2,
-                pad=(0, 0, 0, self.output_bin - aux2.size()[2]),
-                mode="replicate",
-            )
-            return mask * mix, aux1 * mix, aux2 * mix
-        else:
-            if aggressiveness:
-                mask[:, :, : aggressiveness["split_bin"]] = torch.pow(
-                    mask[:, :, : aggressiveness["split_bin"]],
-                    1 + aggressiveness["value"] / 3,
-                )
-                mask[:, :, aggressiveness["split_bin"] :] = torch.pow(
-                    mask[:, :, aggressiveness["split_bin"] :],
-                    1 + aggressiveness["value"],
-                )
-
-            return mask * mix
-
-    def predict(self, x_mag, aggressiveness=None):
-        h = self.forward(x_mag, aggressiveness)
-
-        if self.offset > 0:
-            h = h[:, :, :, self.offset : -self.offset]
-            assert h.size()[3] > 0
-
-        return h

infer/lib/uvr5_pack/lib_v5/nets_new.py

@@ -1,133 +0,0 @@
-import torch
-import torch.nn.functional as F
-from torch import nn
-
-from . import layers_new
-
-
-class BaseNet(nn.Module):
-    def __init__(
-        self, nin, nout, nin_lstm, nout_lstm, dilations=((4, 2), (8, 4), (12, 6))
-    ):
-        super(BaseNet, self).__init__()
-        self.enc1 = layers_new.Conv2DBNActiv(nin, nout, 3, 1, 1)
-        self.enc2 = layers_new.Encoder(nout, nout * 2, 3, 2, 1)
-        self.enc3 = layers_new.Encoder(nout * 2, nout * 4, 3, 2, 1)
-        self.enc4 = layers_new.Encoder(nout * 4, nout * 6, 3, 2, 1)
-        self.enc5 = layers_new.Encoder(nout * 6, nout * 8, 3, 2, 1)
-
-        self.aspp = layers_new.ASPPModule(nout * 8, nout * 8, dilations, dropout=True)
-
-        self.dec4 = layers_new.Decoder(nout * (6 + 8), nout * 6, 3, 1, 1)
-        self.dec3 = layers_new.Decoder(nout * (4 + 6), nout * 4, 3, 1, 1)
-        self.dec2 = layers_new.Decoder(nout * (2 + 4), nout * 2, 3, 1, 1)
-        self.lstm_dec2 = layers_new.LSTMModule(nout * 2, nin_lstm, nout_lstm)
-        self.dec1 = layers_new.Decoder(nout * (1 + 2) + 1, nout * 1, 3, 1, 1)
-
-    def __call__(self, x):
-        e1 = self.enc1(x)
-        e2 = self.enc2(e1)
-        e3 = self.enc3(e2)
-        e4 = self.enc4(e3)
-        e5 = self.enc5(e4)
-
-        h = self.aspp(e5)
-
-        h = self.dec4(h, e4)
-        h = self.dec3(h, e3)
-        h = self.dec2(h, e2)
-        h = torch.cat([h, self.lstm_dec2(h)], dim=1)
-        h = self.dec1(h, e1)
-
-        return h
-
-
-class CascadedNet(nn.Module):
-    def __init__(self, n_fft, nout=32, nout_lstm=128):
-        super(CascadedNet, self).__init__()
-
-        self.max_bin = n_fft // 2
-        self.output_bin = n_fft // 2 + 1
-        self.nin_lstm = self.max_bin // 2
-        self.offset = 64
-
-        self.stg1_low_band_net = nn.Sequential(
-            BaseNet(2, nout // 2, self.nin_lstm // 2, nout_lstm),
-            layers_new.Conv2DBNActiv(nout // 2, nout // 4, 1, 1, 0),
-        )
-
-        self.stg1_high_band_net = BaseNet(
-            2, nout // 4, self.nin_lstm // 2, nout_lstm // 2
-        )
-
-        self.stg2_low_band_net = nn.Sequential(
-            BaseNet(nout // 4 + 2, nout, self.nin_lstm // 2, nout_lstm),
-            layers_new.Conv2DBNActiv(nout, nout // 2, 1, 1, 0),
-        )
-        self.stg2_high_band_net = BaseNet(
-            nout // 4 + 2, nout // 2, self.nin_lstm // 2, nout_lstm // 2
-        )
-
-        self.stg3_full_band_net = BaseNet(
-            3 * nout // 4 + 2, nout, self.nin_lstm, nout_lstm
-        )
-
-        self.out = nn.Conv2d(nout, 2, 1, bias=False)
-        self.aux_out = nn.Conv2d(3 * nout // 4, 2, 1, bias=False)
-
-    def forward(self, x):
-        x = x[:, :, : self.max_bin]
-
-        bandw = x.size()[2] // 2
-        l1_in = x[:, :, :bandw]
-        h1_in = x[:, :, bandw:]
-        l1 = self.stg1_low_band_net(l1_in)
-        h1 = self.stg1_high_band_net(h1_in)
-        aux1 = torch.cat([l1, h1], dim=2)
-
-        l2_in = torch.cat([l1_in, l1], dim=1)
-        h2_in = torch.cat([h1_in, h1], dim=1)
-        l2 = self.stg2_low_band_net(l2_in)
-        h2 = self.stg2_high_band_net(h2_in)
-        aux2 = torch.cat([l2, h2], dim=2)
-
-        f3_in = torch.cat([x, aux1, aux2], dim=1)
-        f3 = self.stg3_full_band_net(f3_in)
-
-        mask = torch.sigmoid(self.out(f3))
-        mask = F.pad(
-            input=mask,
-            pad=(0, 0, 0, self.output_bin - mask.size()[2]),
-            mode="replicate",
-        )
-
-        if self.training:
-            aux = torch.cat([aux1, aux2], dim=1)
-            aux = torch.sigmoid(self.aux_out(aux))
-            aux = F.pad(
-                input=aux,
-                pad=(0, 0, 0, self.output_bin - aux.size()[2]),
-                mode="replicate",
-            )
-            return mask, aux
-        else:
-            return mask
-
-    def predict_mask(self, x):
-        mask = self.forward(x)
-
-        if self.offset > 0:
-            mask = mask[:, :, :, self.offset : -self.offset]
-            assert mask.size()[3] > 0
-
-        return mask
-
-    def predict(self, x, aggressiveness=None):
-        mask = self.forward(x)
-        pred_mag = x * mask
-
-        if self.offset > 0:
-            pred_mag = pred_mag[:, :, :, self.offset : -self.offset]
-            assert pred_mag.size()[3] > 0
-
-        return pred_mag