chore: code cleanup by ruff fix

.ruff.toml

@@ -1 +1,4 @@
+select = ["E", "F", "I"]
 
+# Never enforce `E501` (line length violations).
+ignore = ["E501"]

cluster/__init__.py

@@ -1,6 +1,7 @@
 import torch
 from sklearn.cluster import KMeans
 
+
 def get_cluster_model(ckpt_path):
     checkpoint = torch.load(ckpt_path)
     kmeans_dict = {}

cluster/kmeans.py

@@ -1,7 +1,11 @@
-import torch,pynvml
-from torch.nn.functional import normalize
 from time import time
+
 import numpy as np
+import pynvml
+import torch
+from torch.nn.functional import normalize
+
+
 # device=torch.device("cuda:0")
 def _kpp(data: torch.Tensor, k: int, sample_size: int = -1):
     """ Picks k points in the data based on the kmeans++ method.

cluster/train_cluster.py

@@ -1,17 +1,17 @@
-import time
-import tqdm
+import argparse
+import logging
 import os
+import time
 from pathlib import Path
-import logging
-import argparse
-from kmeans import KMeansGPU
-import torch
+
 import numpy as np
-from sklearn.cluster import KMeans,MiniBatchKMeans
+import torch
+import tqdm
+from kmeans import KMeansGPU
+from sklearn.cluster import KMeans, MiniBatchKMeans
 
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
-import torch
 
 def train_cluster(in_dir, n_clusters, use_minibatch=True, verbose=False,use_gpu=False):#gpu_minibatch真拉，虽然库支持但是也不考虑
     logger.info(f"Loading features from {in_dir}")

data_utils.py

@@ -1,12 +1,13 @@
 import os
 import random
+
 import numpy as np
 import torch
 import torch.utils.data
 
 import utils
-from modules.mel_processing import spectrogram_torch, spectrogram_torch
-from utils import load_wav_to_torch, load_filepaths_and_text
+from modules.mel_processing import spectrogram_torch
+from utils import load_filepaths_and_text, load_wav_to_torch
 
 # import h5py
 

diffusion/data_loaders.py

@@ -1,12 +1,14 @@
 import os
 import random
-import numpy as np
+
 import librosa
+import numpy as np
 import torch
-import random
-from utils import repeat_expand_2d
-from tqdm import tqdm
 from torch.utils.data import Dataset
+from tqdm import tqdm
+
+from utils import repeat_expand_2d
+
 
 def traverse_dir(
         root_dir,

diffusion/diffusion.py

@@ -1,9 +1,10 @@
 from collections import deque
 from functools import partial
 from inspect import isfunction
-import torch.nn.functional as F
+
 import numpy as np
 import torch
+import torch.nn.functional as F
 from torch import nn
 from tqdm import tqdm
 
@@ -254,7 +255,11 @@ def forward(self,
 
             if method is not None and infer_speedup > 1:
                 if method == 'dpm-solver' or method == 'dpm-solver++':
-                    from .dpm_solver_pytorch import NoiseScheduleVP, model_wrapper, DPM_Solver
+                    from .dpm_solver_pytorch import (
+                        DPM_Solver,
+                        NoiseScheduleVP,
+                        model_wrapper,
+                    )
                     # 1. Define the noise schedule.
                     noise_schedule = NoiseScheduleVP(schedule='discrete', betas=self.betas[:t])
 
@@ -332,7 +337,7 @@ def wrapped(x, t, **kwargs):
                                 infer_speedup, cond=cond
                             )
                 elif method == 'unipc':
-                    from .uni_pc import NoiseScheduleVP, model_wrapper, UniPC
+                    from .uni_pc import NoiseScheduleVP, UniPC, model_wrapper
                     # 1. Define the noise schedule.
                     noise_schedule = NoiseScheduleVP(schedule='discrete', betas=self.betas[:t])
 

diffusion/diffusion_onnx.py

@@ -1,14 +1,14 @@
+import math
 from collections import deque
 from functools import partial
 from inspect import isfunction
-import torch.nn.functional as F
+
 import numpy as np
-from torch.nn import Conv1d
-from torch.nn import Mish
 import torch
+import torch.nn.functional as F
 from torch import nn
+from torch.nn import Conv1d, Mish
 from tqdm import tqdm
-import math
 
 
 def exists(x):
@@ -390,7 +390,11 @@ def org_forward(self,
 
             if method is not None and infer_speedup > 1:
                 if method == 'dpm-solver':
-                    from .dpm_solver_pytorch import NoiseScheduleVP, model_wrapper, DPM_Solver
+                    from .dpm_solver_pytorch import (
+                        DPM_Solver,
+                        NoiseScheduleVP,
+                        model_wrapper,
+                    )
                     # 1. Define the noise schedule.
                     noise_schedule = NoiseScheduleVP(schedule='discrete', betas=self.betas[:t])
 

diffusion/infer_gt_mel.py

@@ -1,5 +1,6 @@
 import torch
 import torch.nn.functional as F
+
 from diffusion.unit2mel import load_model_vocoder
 
 

diffusion/logger/saver.py

@@ -2,14 +2,16 @@
 author: wayn391@mastertones
 '''
 
+import datetime
 import os
 import time
-import yaml
-import datetime
-import torch
+
 import matplotlib.pyplot as plt
+import torch
+import yaml
 from torch.utils.tensorboard import SummaryWriter
 
+
 class Saver(object):
     def __init__(
             self, 

diffusion/logger/utils.py

@@ -1,7 +1,9 @@
-import os
-import yaml
 import json
+import os
+
 import torch
+import yaml
+
 
 def traverse_dir(
         root_dir,

diffusion/onnx_export.py

@@ -1,10 +1,12 @@
-from diffusion_onnx import GaussianDiffusion
 import os
-import yaml
+
+import numpy as np
 import torch
 import torch.nn as nn
-import numpy as np
 import torch.nn.functional as F
+import yaml
+from diffusion_onnx import GaussianDiffusion
+
 
 class DotDict(dict):
     def __getattr__(*args):         

diffusion/solver.py

@@ -1,12 +1,15 @@
 import time
+
+import librosa
 import numpy as np
 import torch
-import librosa
-from diffusion.logger.saver import Saver
-from diffusion.logger import utils
 from torch import autocast
 from torch.cuda.amp import GradScaler
 
+from diffusion.logger import utils
+from diffusion.logger.saver import Saver
+
+
 def test(args, model, vocoder, loader_test, saver):
     print(' [*] testing...')
     model.eval()

diffusion/uni_pc.py

@@ -1,6 +1,7 @@
-import torch
 import math
 
+import torch
+
 
 class NoiseScheduleVP:
     def __init__(

diffusion/unit2mel.py

@@ -1,11 +1,14 @@
 import os
-import yaml
+
+import numpy as np
 import torch
 import torch.nn as nn
-import numpy as np
+import yaml
+
 from .diffusion import GaussianDiffusion
-from .wavenet import WaveNet
 from .vocoder import Vocoder
+from .wavenet import WaveNet
+
 
 class DotDict(dict):
     def __getattr__(*args):         

diffusion/vocoder.py

@@ -1,9 +1,10 @@
 import torch
-from vdecoder.nsf_hifigan.nvSTFT import STFT
-from vdecoder.nsf_hifigan.models import load_model,load_config
 from torchaudio.transforms import Resample
 
-
+from vdecoder.nsf_hifigan.models import load_config, load_model
+from vdecoder.nsf_hifigan.nvSTFT import STFT
+
+
 class Vocoder:
     def __init__(self, vocoder_type, vocoder_ckpt, device = None):
         if device is None:

flask_api.py

@@ -7,7 +7,7 @@
 from flask import Flask, request, send_file
 from flask_cors import CORS
 
-from inference.infer_tool import Svc, RealTimeVC
+from inference.infer_tool import RealTimeVC, Svc
 
 app = Flask(__name__)
 

flask_api_full_song.py

@@ -1,10 +1,10 @@
 import io
+
 import numpy as np
 import soundfile
 from flask import Flask, request, send_file
 
-from inference import infer_tool
-from inference import slicer
+from inference import infer_tool, slicer
 
 app = Flask(__name__)
 

inference/infer_tool.py

@@ -1,26 +1,26 @@
+import gc
 import hashlib
 import io
 import json
 import logging
 import os
+import pickle
 import time
 from pathlib import Path
-from inference import slicer
-import gc
 
 import librosa
 import numpy as np
+
 # import onnxruntime
 import soundfile
 import torch
 import torchaudio
 
 import cluster
 import utils
-from models import SynthesizerTrn
-import pickle
-
 from diffusion.unit2mel import load_model_vocoder
+from inference import slicer
+from models import SynthesizerTrn
 
 logging.getLogger('matplotlib').setLevel(logging.WARNING)
 

inference/infer_tool_grad.py

@@ -1,16 +1,18 @@
+import io
 import logging
 import os
-import io
+
 import librosa
 import numpy as np
-from inference import slicer
 import parselmouth
 import soundfile
 import torch
 import torchaudio
 
 import utils
+from inference import slicer
 from models import SynthesizerTrn
+
 logging.getLogger('numba').setLevel(logging.WARNING)
 logging.getLogger('matplotlib').setLevel(logging.WARNING)
 

inference_main.py

@@ -1,8 +1,10 @@
 import logging
-from spkmix import spk_mix_map
+
 import soundfile
+
 from inference import infer_tool
 from inference.infer_tool import Svc
+from spkmix import spk_mix_map
 
 logging.getLogger('numba').setLevel(logging.WARNING)
 chunks_dict = infer_tool.read_temp("inference/chunks_temp.json")

models.py

@@ -1,18 +1,17 @@
 import torch
 from torch import nn
+from torch.nn import Conv1d, Conv2d
 from torch.nn import functional as F
+from torch.nn.utils import spectral_norm, weight_norm
 
 import modules.attentions as attentions
 import modules.commons as commons
 import modules.modules as modules
-
-from torch.nn import Conv1d, Conv2d
-from torch.nn.utils import weight_norm, spectral_norm
-
 import utils
 from modules.commons import get_padding
 from utils import f0_to_coarse
 
+
 class ResidualCouplingBlock(nn.Module):
     def __init__(self,
                  channels,

modules/F0Predictor/CrepeF0Predictor.py

@@ -1,7 +1,9 @@
-from modules.F0Predictor.F0Predictor import F0Predictor
-from modules.F0Predictor.crepe import CrepePitchExtractor
 import torch
 
+from modules.F0Predictor.crepe import CrepePitchExtractor
+from modules.F0Predictor.F0Predictor import F0Predictor
+
+
 class CrepeF0Predictor(F0Predictor):
     def __init__(self,hop_length=512,f0_min=50,f0_max=1100,device=None,sampling_rate=44100,threshold=0.05,model="full"):
         self.F0Creper = CrepePitchExtractor(hop_length=hop_length,f0_min=f0_min,f0_max=f0_max,device=device,threshold=threshold,model=model)