allow packaging and running from a wheel

.python-version

@@ -0,0 +1 @@
+3.11

manga_translator/__init__.py

@@ -4,4 +4,4 @@
 colorama.init(autoreset=True)
 load_dotenv()
 
-from .manga_translator import *
+from manga_translator.manga_translator import *

manga_translator/__main__.py

@@ -3,13 +3,12 @@
 import logging
 from argparse import Namespace
 
-from manga_translator import Config
+from manga_translator.config import Config
 from manga_translator.args import parser, reparse
-from .manga_translator import (
+from manga_translator.manga_translator import (
     set_main_logger, load_dictionary, apply_dictionary,
 )
-from .args import parser
-from .utils import (
+from manga_translator.utils import (
     BASE_PATH,
     init_logging,
     get_logger,
@@ -21,7 +20,8 @@
 
 async def dispatch(args: Namespace):
     args_dict = vars(args)
-
+    global logger
+    logger = get_logger(args.mode)
     logger.info(f'Running in {args.mode} mode')
 
     if args.mode == 'local':
@@ -77,9 +77,7 @@ async def dispatch(args: Namespace):
         config = Config.schema()
         print(json.dumps(config, indent=2))
 
-
-
-if __name__ == '__main__':
+def main():
     args = None
     init_logging()
     try:
@@ -98,3 +96,6 @@ async def dispatch(args: Namespace):
     except Exception as e:
         logger.error(f'{e.__class__.__name__}: {e}',
                      exc_info=e if args and args.verbose else None)
+
+if __name__ == '__main__':
+    main()

manga_translator/args.py

@@ -2,12 +2,6 @@
 import os
 from urllib.parse import unquote
 
-from .detection import DETECTORS
-from .ocr import OCRS
-from .inpainting import INPAINTERS
-from .translators import VALID_LANGUAGES, TRANSLATORS, TranslatorChain
-from .upscaling import UPSCALERS
-from .colorization import COLORIZERS
 from .save import OUTPUT_FORMATS
 
 def url_decode(s):

manga_translator/colorization/manga_colorization_v2_utils/denoising/denoiser.py

@@ -11,17 +11,14 @@
 this program. If not, see <http://www.gnu.org/licenses/>.
 """
 import os
-import argparse
-import time
 
 
 import numpy as np
 import cv2
 import torch
 import torch.nn as nn
-from torch.autograd import Variable
 from .models import FFDNet
-from .utils import normalize, variable_to_cv2_image, remove_dataparallel_wrapper, is_rgb
+from .utils import normalize, variable_to_cv2_image, remove_dataparallel_wrapper
 
 class FFDNetDenoiser:
     def __init__(self, _device, _sigma = 25, _weights_dir = 'denoising/models/', _in_ch = 3):

manga_translator/colorization/manga_colorization_v2_utils/networks/extractor.py

@@ -1,4 +1,3 @@
-import torch
 import torch.nn as nn
 import math
 

manga_translator/colorization/manga_colorization_v2_utils/networks/models.py

@@ -1,9 +1,6 @@
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import torchvision.models as M
-import math
-from torch import Tensor
 from torch.nn import Parameter
 
 from .extractor import SEResNeXt_Origin, BottleneckX_Origin

manga_translator/config.py

@@ -27,9 +27,9 @@ def __init__(self, string: str):
             else:
                 translator = Translator[trans]
             if translator not in TRANSLATORS:
-                raise ValueError(f'Invalid choice: %s (choose from %s)' % (trans, ', '.join(map(repr, TRANSLATORS))))
+                raise ValueError('Invalid choice: %s (choose from %s)' % (trans, ', '.join(map(repr, TRANSLATORS))))
             if lang not in VALID_LANGUAGES:
-                raise ValueError(f'Invalid choice: %s (choose from %s)' % (lang, ', '.join(map(repr, VALID_LANGUAGES))))
+                raise ValueError('Invalid choice: %s (choose from %s)' % (lang, ', '.join(map(repr, VALID_LANGUAGES))))
             self.chain.append((translator, lang))
         self.translators, self.langs = list(zip(*self.chain))
 

manga_translator/detection/craft.py

@@ -10,17 +10,12 @@
 
 import os
 import shutil
-import numpy as np
-import torch
 import cv2
-import einops
-from typing import List, Tuple
 
-from .default_utils.DBNet_resnet34 import TextDetection as TextDetectionDefault
-from .default_utils import imgproc, dbnet_utils, craft_utils
+from .default_utils import imgproc, craft_utils
 from .common import OfflineDetector
-from ..utils import TextBlock, Quadrilateral, det_rearrange_forward
-from shapely.geometry import Polygon, MultiPoint
+from ..utils import Quadrilateral
+from shapely.geometry import MultiPoint
 from shapely import affinity
 
 from .craft_utils.vgg16_bn import vgg16_bn, init_weights

manga_translator/detection/craft_utils/refiner.py

@@ -6,8 +6,6 @@
 # -*- coding: utf-8 -*-
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
-from torch.autograd import Variable
 from .vgg16_bn import init_weights
 
 

manga_translator/detection/ctd.py

@@ -10,7 +10,7 @@
 from .ctd_utils.utils.yolov5_utils import non_max_suppression
 from .ctd_utils.utils.db_utils import SegDetectorRepresenter
 from .ctd_utils.utils.imgproc_utils import letterbox
-from .ctd_utils.textmask import REFINEMASK_INPAINT, refine_mask
+from .ctd_utils.textmask import refine_mask
 from .common import OfflineDetector
 from ..utils import Quadrilateral, det_rearrange_forward
 

manga_translator/detection/ctd_utils/utils/io_utils.py

@@ -1,4 +1,3 @@
-import os
 import os.path as osp
 import glob
 from pathlib import Path

manga_translator/detection/ctd_utils/yolov5/common.py

@@ -3,23 +3,12 @@
 Common modules
 """
 
-import json
 import math
-import platform
 import warnings
-from collections import OrderedDict, namedtuple
-from copy import copy
-from pathlib import Path
 
-import cv2
-import numpy as np
-import requests
 import torch
 import torch.nn as nn
-from PIL import Image
-from torch.cuda import amp
 
-from ..utils.yolov5_utils import make_divisible, initialize_weights, check_anchor_order, check_version, fuse_conv_and_bn
 
 def autopad(k, p=None):  # kernel, padding
     # Pad to 'same'

manga_translator/detection/ctd_utils/yolov5/yolo.py

@@ -1,5 +1,3 @@
-from operator import mod
-from cv2 import imshow
 # from utils.yolov5_utils import scale_img
 from copy import deepcopy
 from .common import *

manga_translator/detection/dbnet_convnext.py

@@ -6,13 +6,9 @@
 import numpy as np
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
-import torch.nn.init as init
 
-from torchvision.models import resnet34
 
 import einops
-import math
 
 from timm.layers import trunc_normal_, AvgPool2dSame, DropPath, Mlp, GlobalResponseNormMlp, \
 	LayerNorm2d, LayerNorm, create_conv2d, get_act_layer, make_divisible, to_ntuple
@@ -493,7 +489,7 @@ def forward(self, x) :
 import os
 from .default_utils import imgproc, dbnet_utils, craft_utils
 from .common import OfflineDetector
-from ..utils import TextBlock, Quadrilateral, det_rearrange_forward
+from ..utils import Quadrilateral, det_rearrange_forward
 
 MODEL = None
 def det_batch_forward_default(batch: np.ndarray, device: str):

manga_translator/detection/default.py

@@ -4,12 +4,11 @@
 import torch
 import cv2
 import einops
-from typing import List, Tuple
 
 from .default_utils.DBNet_resnet34 import TextDetection as TextDetectionDefault
 from .default_utils import imgproc, dbnet_utils, craft_utils
 from .common import OfflineDetector
-from ..utils import TextBlock, Quadrilateral, det_rearrange_forward
+from ..utils import Quadrilateral, det_rearrange_forward
 
 MODEL = None
 def det_batch_forward_default(batch: np.ndarray, device: str):

manga_translator/detection/default_utils/CRAFT_resnet34.py

@@ -2,12 +2,10 @@
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import torch.nn.init as init
 
 from torchvision.models import resnet34
 
 import einops
-import math
 
 class ImageMultiheadSelfAttention(nn.Module):
     def __init__(self, planes):

manga_translator/detection/default_utils/DBNet_resnet101.py

@@ -1,7 +1,6 @@
 
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
 
 from torchvision.models import resnet101
 

manga_translator/inpainting/booru_tagger.py

@@ -1,16 +1,12 @@
 import os
-import gc
 import pandas as pd
 import numpy as np
 from onnxruntime import InferenceSession
-from typing import Tuple, List, Dict
-from io import BytesIO
+from typing import Dict
 from PIL import Image
 
 import cv2
-from pathlib import Path
 
-from tqdm import tqdm
 
 def make_square(img, target_size):
     old_size = img.shape[:2]

manga_translator/inpainting/guided_ldm_inpainting.py

@@ -1,27 +1,13 @@
 
 
-import einops
 import torch
-import torch as th
-import torch.nn as nn
 import numpy as np
 from tqdm import tqdm
-import cv2
-
-from .ldm.modules.diffusionmodules.util import (
-    conv_nd,
-    linear,
-    zero_module,
-    timestep_embedding,
-)
+
 from .ldm.modules.diffusionmodules.util import noise_like
-from einops import rearrange, repeat
-from torchvision.utils import make_grid
-from .ldm.modules.attention import SpatialTransformer
-from .ldm.modules.diffusionmodules.openaimodel import UNetModel, TimestepEmbedSequential, ResBlock, Downsample, AttentionBlock
 from .ldm.models.diffusion.ddpm import LatentDiffusion
 from .ldm.models.diffusion.ddim import DDIMSampler
-from .ldm.util import log_txt_as_img, exists, instantiate_from_config
+from .ldm.util import instantiate_from_config
 
 class GuidedDDIMSample(DDIMSampler) :
     def __init__(self, *args, **kwargs):

manga_translator/inpainting/inpainting_lama.py

@@ -149,7 +149,7 @@ def forward(self, x: torch.Tensor) -> Tuple[torch.Tensor, List[torch.Tensor]]:
         Predict scores and get intermediate activations. Useful for feature matching loss
         :return tuple (scores, list of intermediate activations)
         """
-        raise NotImplemented()
+        raise NotImplementedError()
 
 
 def get_conv_block_ctor(kind='default'):
@@ -687,7 +687,6 @@ def get_discriminator():
     }
     return FFCNLayerDiscriminator(3, norm_layer = nn.Identity, init_conv_kwargs = init_conv_kwargs, conv_kwargs = conv_kwargs)
 
-from torchsummary import summary
 
 def test_model():
     dis = get_generator()

manga_translator/inpainting/ldm/models/diffusion/ddpm.py

@@ -112,7 +112,7 @@ def __init__(self,
             self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys, only_model=load_only_unet)
             if reset_ema:
                 assert self.use_ema
-                print(f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
+                print("Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
                 self.model_ema = LitEma(self.model)
         if reset_num_ema_updates:
             print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ")
@@ -227,7 +227,7 @@ def init_from_ckpt(self, path, ignore_keys=list(), only_model=False):
                     desc="Fitting old weights to new weights",
                     total=n_params
             ):
-                if not name in sd:
+                if name not in sd:
                     continue
                 old_shape = sd[name].shape
                 new_shape = param.shape
@@ -574,7 +574,7 @@ def __init__(self,
             if reset_ema:
                 assert self.use_ema
                 print(
-                    f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
+                    "Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
                 self.model_ema = LitEma(self.model)
         if reset_num_ema_updates:
             print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ")

manga_translator/inpainting/ldm/models/diffusion/dpm_solver/dpm_solver.py

@@ -1,5 +1,4 @@
 import torch
-import torch.nn.functional as F
 import math
 from tqdm import tqdm
 

manga_translator/inpainting/ldm/models/diffusion/plms.py

@@ -3,7 +3,6 @@
 import torch
 import numpy as np
 from tqdm import tqdm
-from functools import partial
 
 from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like
 from ldm.models.diffusion.sampling_util import norm_thresholding

manga_translator/inpainting/ldm/models/diffusion/sampling_util.py

@@ -1,5 +1,3 @@
-import torch
-import numpy as np
 
 
 def append_dims(x, target_dims):

manga_translator/inpainting/ldm/modules/diffusionmodules/util.py

@@ -8,7 +8,6 @@
 # thanks!
 
 
-import os
 import math
 import torch
 import torch.nn as nn

manga_translator/inpainting/ldm/modules/ema.py

@@ -45,7 +45,7 @@ def forward(self, model):
                     shadow_params[sname] = shadow_params[sname].type_as(m_param[key])
                     shadow_params[sname].sub_(one_minus_decay * (shadow_params[sname] - m_param[key]))
                 else:
-                    assert not key in self.m_name2s_name
+                    assert key not in self.m_name2s_name
 
     def copy_to(self, model):
         m_param = dict(model.named_parameters())
@@ -54,7 +54,7 @@ def copy_to(self, model):
             if m_param[key].requires_grad:
                 m_param[key].data.copy_(shadow_params[self.m_name2s_name[key]].data)
             else:
-                assert not key in self.m_name2s_name
+                assert key not in self.m_name2s_name
 
     def store(self, parameters):
         """

manga_translator/inpainting/ldm/modules/encoders/modules.py

@@ -5,7 +5,7 @@
 from transformers import T5Tokenizer, T5EncoderModel, CLIPTokenizer, CLIPTextModel
 
 import open_clip
-from ldm.util import default, count_params
+from ldm.util import count_params
 
 
 class AbstractEncoder(nn.Module):

manga_translator/inpainting/ldm/modules/midas/midas/blocks.py

@@ -5,7 +5,6 @@
     _make_pretrained_vitb_rn50_384,
     _make_pretrained_vitl16_384,
     _make_pretrained_vitb16_384,
-    forward_vit,
 )
 
 def _make_encoder(backbone, features, use_pretrained, groups=1, expand=False, exportable=True, hooks=None, use_vit_only=False, use_readout="ignore",):