Use pre-commit to reformat code

Use pre-commit to reformat code

classification/configs/accelerate/deepspeed/ds_config_zero3_offload.json

@@ -17,4 +17,4 @@
     "steps_per_print": "inf",
     "train_batch_size": "auto",
     "train_micro_batch_size_per_gpu": "auto"
-}
\ No newline at end of file
+}

classification/configs/accelerate/deepspeed/ds_config_zero3_offload_wo_loss_scale.json

@@ -18,4 +18,4 @@
     "steps_per_print": "inf",
     "train_batch_size": "auto",
     "train_micro_batch_size_per_gpu": "auto"
-}
\ No newline at end of file
+}

classification/configs/accelerate/dist_8gpus_ddp_fp16.yaml

@@ -10,4 +10,4 @@ main_training_function: main
 mixed_precision: 'fp16'
 num_machines: 1
 num_processes: 8
-use_cpu: false
\ No newline at end of file
+use_cpu: false

classification/configs/accelerate/dist_8gpus_zero1.yaml

@@ -9,4 +9,4 @@ main_training_function: main
 num_machines: 1
 num_processes: 8
 rdzv_backend: static
-use_cpu: false
\ No newline at end of file
+use_cpu: false

classification/configs/accelerate/dist_8gpus_zero1_wo_loss_scale.yaml

@@ -9,4 +9,4 @@ main_training_function: main
 num_machines: 1
 num_processes: 8
 rdzv_backend: static
-use_cpu: false
\ No newline at end of file
+use_cpu: false

classification/configs/accelerate/dist_8gpus_zero3_offload.yaml

@@ -9,4 +9,4 @@ main_training_function: main
 num_machines: 1
 num_processes: 8
 rdzv_backend: static
-use_cpu: false
\ No newline at end of file
+use_cpu: false

classification/configs/accelerate/dist_8gpus_zero3_offload_wo_loss_scale.yaml

@@ -9,4 +9,4 @@ main_training_function: main
 num_machines: 1
 num_processes: 8
 rdzv_backend: static
-use_cpu: false
\ No newline at end of file
+use_cpu: false

classification/configs/internimage_b_1k_224.yaml

@@ -16,4 +16,4 @@ TRAIN:
   EMA:
     ENABLE: True
     DECAY: 0.9999
-  BASE_LR: 5e-4
\ No newline at end of file
+  BASE_LR: 5e-4

classification/configs/internimage_g_22kto1k_512.yaml

@@ -39,4 +39,4 @@ TRAIN:
   OPTIMIZER:
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/internimage_h_22kto1k_384.yaml

@@ -40,4 +40,4 @@ TRAIN:
   OPTIMIZER:
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/internimage_h_22kto1k_640.yaml

@@ -41,4 +41,4 @@ TRAIN:
     USE_ZERO: True
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/internimage_xl_22kto1k_384.yaml

@@ -34,4 +34,4 @@ TRAIN:
   OPTIMIZER:
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/without_lr_decay/internimage_b_1k_224.yaml

@@ -16,4 +16,4 @@ TRAIN:
   EMA:
     ENABLE: True
     DECAY: 0.9999
-  BASE_LR: 5e-4
\ No newline at end of file
+  BASE_LR: 5e-4

classification/configs/without_lr_decay/internimage_g_22kto1k_512.yaml

@@ -37,4 +37,4 @@ TRAIN:
   OPTIMIZER:
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/without_lr_decay/internimage_h_22kto1k_640.yaml

@@ -39,4 +39,4 @@ TRAIN:
     USE_ZERO: True
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/configs/without_lr_decay/internimage_xl_22kto1k_384.yaml

@@ -32,4 +32,4 @@ TRAIN:
   OPTIMIZER:
     DCN_LR_MUL: 0.1
 AMP_OPT_LEVEL: O0
-EVAL_FREQ: 1
\ No newline at end of file
+EVAL_FREQ: 1

classification/dataset/__init__.py

@@ -4,4 +4,4 @@
 # Licensed under The MIT License [see LICENSE for details]
 # --------------------------------------------------------
 
-from .build import build_loader, build_loader2
\ No newline at end of file
+from .build import build_loader, build_loader2

classification/dataset/build.py

@@ -5,14 +5,15 @@
 # --------------------------------------------------------
 
 import os
-import torch
+
 import numpy as np
+import torch
 import torch.distributed as dist
+from timm.data import Mixup, create_transform
 from torchvision import transforms
-from timm.data import Mixup
-from timm.data import create_transform
+
 from .cached_image_folder import ImageCephDataset
-from .samplers import SubsetRandomSampler, NodeDistributedSampler
+from .samplers import NodeDistributedSampler, SubsetRandomSampler
 
 try:
     from torchvision.transforms import InterpolationMode
@@ -50,7 +51,7 @@ class TTA(torch.nn.Module):
         return out
 
     def __repr__(self) -> str:
-        return f"{self.__class__.__name__}(size={self.size}, scale={self.scales})"
+        return f'{self.__class__.__name__}(size={self.size}, scale={self.scales})'
 
 
 def build_loader(config):
@@ -58,16 +59,16 @@ def build_loader(config):
     dataset_train, config.MODEL.NUM_CLASSES = build_dataset('train',
                                                             config=config)
     config.freeze()
-    print(f"local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}"
-          "successfully build train dataset")
+    print(f'local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}'
+          'successfully build train dataset')
 
     dataset_val, _ = build_dataset('val', config=config)
-    print(f"local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}"
-          "successfully build val dataset")
+    print(f'local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}'
+          'successfully build val dataset')
 
     dataset_test, _ = build_dataset('test', config=config)
-    print(f"local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}"
-          "successfully build test dataset")
+    print(f'local rank {config.LOCAL_RANK} / global rank {dist.get_rank()}'
+          'successfully build test dataset')
 
     num_tasks = dist.get_world_size()
     global_rank = dist.get_rank()

classification/dataset/cached_image_folder.py

@@ -5,22 +5,24 @@
 # --------------------------------------------------------
 
 import io
+import json
+import logging
+import math
 import os
+import os.path as osp
 import re
 import time
-import json
-import math
+from abc import abstractmethod
+
 import mmcv
 import torch
-import logging
-import os.path as osp
-from PIL import Image
-from tqdm import tqdm, trange
-from abc import abstractmethod
-import torch.utils.data as data
 import torch.distributed as dist
+import torch.utils.data as data
 from mmcv.fileio import FileClient
-from .zipreader import is_zip_path, ZipReader
+from PIL import Image
+from tqdm import tqdm, trange
+
+from .zipreader import ZipReader, is_zip_path
 
 _logger = logging.getLogger(__name__)
 
@@ -66,7 +68,7 @@ def make_dataset(dir, class_to_idx, extensions):
 
 def make_dataset_with_ann(ann_file, img_prefix, extensions):
     images = []
-    with open(ann_file, "r") as f:
+    with open(ann_file, 'r') as f:
         contents = f.readlines()
         for line_str in contents:
             path_contents = [c for c in line_str.split('\t')]
@@ -108,7 +110,7 @@ class DatasetFolder(data.Dataset):
                  img_prefix='',
                  transform=None,
                  target_transform=None,
-                 cache_mode="no"):
+                 cache_mode='no'):
         # image folder mode
         if ann_file == '':
             _, class_to_idx = find_classes(root)
@@ -120,9 +122,9 @@ class DatasetFolder(data.Dataset):
                                             extensions)
 
         if len(samples) == 0:
-            raise (RuntimeError("Found 0 files in subfolders of: " + root +
-                                "\n" + "Supported extensions are: " +
-                                ",".join(extensions)))
+            raise (RuntimeError('Found 0 files in subfolders of: ' + root +
+                                '\n' + 'Supported extensions are: ' +
+                                ','.join(extensions)))
 
         self.root = root
         self.loader = loader
@@ -136,11 +138,11 @@ class DatasetFolder(data.Dataset):
         self.target_transform = target_transform
 
         self.cache_mode = cache_mode
-        if self.cache_mode != "no":
+        if self.cache_mode != 'no':
             self.init_cache()
 
     def init_cache(self):
-        assert self.cache_mode in ["part", "full"]
+        assert self.cache_mode in ['part', 'full']
         n_sample = len(self.samples)
         global_rank = dist.get_rank()
         world_size = dist.get_world_size()
@@ -155,9 +157,9 @@ class DatasetFolder(data.Dataset):
                 )
                 start_time = time.time()
             path, target = self.samples[index]
-            if self.cache_mode == "full":
+            if self.cache_mode == 'full':
                 samples_bytes[index] = (ZipReader.read(path), target)
-            elif self.cache_mode == "part" and index % world_size == global_rank:
+            elif self.cache_mode == 'part' and index % world_size == global_rank:
                 samples_bytes[index] = (ZipReader.read(path), target)
             else:
                 samples_bytes[index] = (path, target)
@@ -260,7 +262,7 @@ class CachedImageFolder(DatasetFolder):
                  transform=None,
                  target_transform=None,
                  loader=default_img_loader,
-                 cache_mode="no"):
+                 cache_mode='no'):
         super(CachedImageFolder,
               self).__init__(root,
                              loader,
@@ -394,7 +396,7 @@ class ParserCephImage(Parser):
                 local_size = int(os.environ.get('LOCAL_SIZE', 1))
             self.local_rank = local_rank
             self.local_size = local_size
-            self.rank = int(os.environ["RANK"])
+            self.rank = int(os.environ['RANK'])
             self.world_size = int(os.environ['WORLD_SIZE'])
             self.num_replicas = int(os.environ['WORLD_SIZE'])
             self.num_parts = local_size
@@ -405,7 +407,7 @@ class ParserCephImage(Parser):
             self.load_onto_memory_v2()
 
     def load_onto_memory(self):
-        print("Loading images onto memory...", self.local_rank,
+        print('Loading images onto memory...', self.local_rank,
               self.local_size)
         if self.file_client is None:
             self.file_client = FileClient(self.io_backend, **self.kwargs)
@@ -417,7 +419,7 @@ class ParserCephImage(Parser):
             img_bytes = self.file_client.get(path)
             self.holder[path] = img_bytes
 
-        print("Loading complete!")
+        print('Loading complete!')
 
     def load_onto_memory_v2(self):
         # print("Loading images onto memory...", self.local_rank, self.local_size)
@@ -446,7 +448,7 @@ class ParserCephImage(Parser):
 
             self.holder[path] = img_bytes
 
-        print("Loading complete!")
+        print('Loading complete!')
 
     def __getitem__(self, index):
         if self.file_client is None:

classification/dataset/samplers.py

@@ -4,12 +4,13 @@
 # Licensed under The MIT License [see LICENSE for details]
 # --------------------------------------------------------
 
-import torch
-import os
 import math
-from torch.utils.data.sampler import Sampler
-import torch.distributed as dist
+import os
+
 import numpy as np
+import torch
+import torch.distributed as dist
+from torch.utils.data.sampler import Sampler
 
 
 class SubsetRandomSampler(torch.utils.data.Sampler):
@@ -57,12 +58,12 @@ class NodeDistributedSampler(Sampler):
         if num_replicas is None:
             if not dist.is_available():
                 raise RuntimeError(
-                    "Requires distributed package to be available")
+                    'Requires distributed package to be available')
             num_replicas = dist.get_world_size()
         if rank is None:
             if not dist.is_available():
                 raise RuntimeError(
-                    "Requires distributed package to be available")
+                    'Requires distributed package to be available')
             rank = dist.get_rank()
         if local_rank is None:
             local_rank = int(os.environ.get('LOCAL_RANK', 0))