mmocr 训练自己的数据集

电脑配置我这边是windows 10,显卡

NVIDIA GeForce GTX 1660 Ti 6G

1、拉取源代码

https://github.com/open-mmlab/mmocr.git

我这边拉的是master

2、环境配置

最好采用全新的环境

安装组件如果慢的话，可以指定国内源

国内常见包源

清华：https://pypi.tuna.tsinghua.edu.cn/simple

阿里云：http://mirrors.aliyun.com/pypi/simple/

中国科技大学 https://pypi.mirrors.ustc.edu.cn/simple/

华中理工大学：http://pypi.hustunique.com/

山东理工大学：http://pypi.sdutlinux.org/

豆瓣：http://pypi.douban.com/simple/

在指定命令后加 "-i 地址"

conda create --name openmmlab python=3.8 -y
切换到环境
conda activate openmmlab
安装pytorch
conda install pytorch torchvision -c pytorch
安装mmcv-full, 不要安装mmcv 这个里面不全

mim install mmcv-full -i https://pypi.tuna.tsinghua.edu.cn/simple

安装分割插件
pip install mmdet

具体可以参考官方文档：
https://mmocr.readthedocs.io/zh_CN/latest/install.html#id3

3、代码配置

先配置一个你要训练的模型生成配置代码，我这边用的是abinet (configs/textrecog/abinet/abinet_academic.py)这个模型：

在 tools/work_dirs/abinet_vision_only_academic/abinet_vision_only_academic.py 下生成了完整的配置文件（为了方便调整，原始文件是多个文件分块写的），把abinet_vision_only_academic.py 改一下名字放到

configs/textrecog/abinet/ 文件夹下，我这名字是 my_abinet_vision_only_academic.py

修改配置为

my_abinet_vision_only_academic.py

准备数据集，我这采用公用数据集：

 icdar_2015 下载地址：https://rrc.cvc.uab.es/?ch=4&com=downloads

也可以在百度网盘下载：

链接：https://pan.baidu.com/s/1et-BtES7FbvK-Z4wtkLuvw
提取码：kshx

修改配置文件：my_abinet_vision_only_academic.py

我这直接上代码吧注意点我标红了：

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

# 多少打印日志log_config = dict(interval=30, hooks=[dict(type='TextLoggerHook')])
dist_params = dict(backend='nccl')
log_level = 'INFO'
load_from = None
resume_from = None
workflow = [('train', 1)]
opencv_num_threads = 0
mp_start_method = 'fork'
optimizer = dict(type='Adam', lr=0.0001)
optimizer_config = dict(grad_clip=None)
lr_config = dict(
    policy='step',
    step=[16, 18],
    warmup='linear',
    warmup_iters=1,
    warmup_ratio=0.001,
    warmup_by_epoch=True)
runner = dict(type='EpochBasedRunner', max_epochs=20)
checkpoint_config = dict(interval=1)
img_norm_cfg = dict(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='ResizeOCR',
        height=32,
        min_width=128,
        max_width=128,
        keep_aspect_ratio=False,
        width_downsample_ratio=0.25),
    dict(
        type='RandomWrapper',
        p=0.5,
        transforms=[
            dict(
                type='OneOfWrapper',
                transforms=[
                    dict(type='RandomRotateTextDet', max_angle=15),
                    dict(
                        type='TorchVisionWrapper',
                        op='RandomAffine',
                        degrees=15,
                        translate=(0.3, 0.3),
                        scale=(0.5, 2.0),
                        shear=(-45, 45)),
                    dict(
                        type='TorchVisionWrapper',
                        op='RandomPerspective',
                        distortion_scale=0.5,
                        p=1)
                ])
        ]),
    dict(
        type='RandomWrapper',
        p=0.25,
        transforms=[
            dict(type='PyramidRescale'),
            dict(
                type='Albu',
                transforms=[
                    dict(type='GaussNoise', var_limit=(20, 20), p=0.5),
                    dict(type='MotionBlur', blur_limit=6, p=0.5)
                ])
        ]),
    dict(
        type='RandomWrapper',
        p=0.25,
        transforms=[
            dict(
                type='TorchVisionWrapper',
                op='ColorJitter',
                brightness=0.5,
                saturation=0.5,
                contrast=0.5,
                hue=0.1)
        ]),
    dict(type='ToTensorOCR'),
    dict(
        type='NormalizeOCR',
        mean=[0.485, 0.456, 0.406],
        std=[0.229, 0.224, 0.225]),
    dict(
        type='Collect',
        keys=['img'],
        meta_keys=[
            'filename', 'ori_shape', 'img_shape', 'text', 'valid_ratio',
            'resize_shape'
        ])
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiRotateAugOCR',
        rotate_degrees=[0, 90, 270],
        transforms=[
            dict(
                type='ResizeOCR',
                height=32,
                min_width=128,
                max_width=128,
                keep_aspect_ratio=False,
                width_downsample_ratio=0.25),
            dict(type='ToTensorOCR'),
            dict(
                type='NormalizeOCR',
                mean=[0.485, 0.456, 0.406],
                std=[0.229, 0.224, 0.225]),
            dict(
                type='Collect',
                keys=['img'],
                meta_keys=[
                    'filename', 'ori_shape', 'img_shape', 'valid_ratio',
                    'resize_shape', 'img_norm_cfg', 'ori_filename'
                ])
        ])
]# 数据集路径 由于我这边只有1个数据集 其他的删除，默认有很多
train_root = r'E:\PythonXX\pytorch\data\mmorc\mixture'
train_img_prefix1 = r'E:\PythonXX\pytorch\data\mmorc\mixture\icdar_2015'
train_ann_file1 = r'E:\PythonXX\pytorch\data\mmorc\mixture\icdar_2015\train_label.txt'
train1 = dict(
    type='OCRDataset',
    img_prefix=train_img_prefix1,
    ann_file=train_ann_file1,
    loader=dict(
        type='AnnFileLoader',
        repeat=1,
        parser=dict(type='LineStrParser', keys=['filename', 'text'])),
    pipeline=None,
    test_mode=False)
# 这个LineStrParser 默认是 LineJsonParser 因为我这边标签是文本文件# file_format='lmdb' 这个要去掉或者改成“txt” 因为我们不是这个格式 是文本格式标签train_list = [ dict( type='OCRDataset', img_prefix=train_img_prefix1, ann_file=train_ann_file1, loader=dict( type='AnnFileLoader', repeat=1, parser=dict(type='LineStrParser', keys=['filename', 'text'])), pipeline=None, test_mode=False) ]# 测试数据集
test_root = r'E:\PythonXX\pytorch\data\mmorc\mixture'
test_img_prefix1 = r'E:\PythonXX\pytorch\data\mmorc\mixture\icdar_2015'
test_ann_file1 = r'E:\PythonXX\pytorch\data\mmorc\mixture\icdar_2015\test_label.txt'
 
test1 = dict(
    type='OCRDataset',
    img_prefix=test_img_prefix1,
    ann_file=test_ann_file1,
    loader=dict(
        type='AnnFileLoader',
        repeat=1,
        file_format='txt',
        parser=dict(
            type='LineStrParser',
            keys=['filename', 'text'],
            keys_idx=[0, 1],
            separator=' ')),
    pipeline=None,
    test_mode=True)
test_list = [
    dict(
        type='OCRDataset',
        img_prefix=test_img_prefix1,
        ann_file=test_ann_file1,
        loader=dict(
            type='AnnFileLoader',
            repeat=1,
            file_format='txt',
            parser=dict(
                type='LineStrParser',
                keys=['filename', 'text'],
                keys_idx=[0, 1],
                separator=' ')),
        pipeline=None,
        test_mode=True)
]
num_chars = 38 #字符分类数量
max_seq_len = 10 #识别字符最大长度
label_convertor = dict(
    type='ABIConvertor',
    dict_type='DICT36',#识别字符类型 10个数字+26字母
    with_unknown=True, #如果报这个错就改
    with_padding=False,
    lower=True)
model = dict(
    type='ABINet',
    backbone=dict(type='ResNetABI'),
    encoder=dict(
        type='ABIVisionModel',
        encoder=dict(
            type='TransformerEncoder',
            n_layers=3,
            n_head=8,
            d_model=512,
            d_inner=2048,
            dropout=0.1,
            max_len=256),
        decoder=dict(
            type='ABIVisionDecoder',
            in_channels=512,
            num_channels=64,
            attn_height=8,
            attn_width=32,
            attn_mode='nearest',
            use_result='feature',
            num_chars=num_chars,
            max_seq_len=max_seq_len,
            init_cfg=dict(type='Xavier', layer='Conv2d'))),
    loss=dict(
        type='ABILoss',
        enc_weight=1.0,
        dec_weight=1.0,
        fusion_weight=1.0,
        num_classes=num_chars),
    label_convertor=dict(
        type='ABIConvertor',
        dict_type='DICT36',
        with_unknown=True,
        with_padding=False,
        lower=True),
    max_seq_len=max_seq_len,
    iter_size=1)
data = dict(
    samples_per_gpu=40, # 批次每一批次训练多少张图片
    workers_per_gpu=1,  # 几个GUP训练
    val_dataloader=dict(samples_per_gpu=1),
    test_dataloader=dict(samples_per_gpu=1),
    train=dict(
        type='UniformConcatDataset',
        datasets=[
            dict(
                type='OCRDataset',
                img_prefix=train_img_prefix1,
                ann_file=train_ann_file1,
                loader=dict(
                    type='AnnFileLoader',
                    repeat=1,
 
                    parser=dict(
                        type='LineStrParser', keys=['filename', 'text'])),
                pipeline=None,
                test_mode=False)
        ],
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='ResizeOCR',
                height=32,
                min_width=128,
                max_width=128,
                keep_aspect_ratio=False,
                width_downsample_ratio=0.25),
            dict(
                type='RandomWrapper',
                p=0.5,
                transforms=[
                    dict(
                        type='OneOfWrapper',
                        transforms=[
                            dict(type='RandomRotateTextDet', max_angle=15),
                            dict(
                                type='TorchVisionWrapper',
                                op='RandomAffine',
                                degrees=15,
                                translate=(0.3, 0.3),
                                scale=(0.5, 2.0),
                                shear=(-45, 45)),
                            dict(
                                type='TorchVisionWrapper',
                                op='RandomPerspective',
                                distortion_scale=0.5,
                                p=1)
                        ])
                ]),
            dict(
                type='RandomWrapper',
                p=0.25,
                transforms=[
                    dict(type='PyramidRescale'),
                    dict(
                        type='Albu',
                        transforms=[
                            dict(type='GaussNoise', var_limit=(20, 20), p=0.5),
                            dict(type='MotionBlur', blur_limit=6, p=0.5)
                        ])
                ]),
            dict(
                type='RandomWrapper',
                p=0.25,
                transforms=[
                    dict(
                        type='TorchVisionWrapper',
                        op='ColorJitter',
                        brightness=0.5,
                        saturation=0.5,
                        contrast=0.5,
                        hue=0.1)
                ]),
            dict(type='ToTensorOCR'),
            dict(
                type='NormalizeOCR',
                mean=[0.485, 0.456, 0.406],
                std=[0.229, 0.224, 0.225]),
            dict(
                type='Collect',
                keys=['img'],
                meta_keys=[
                    'filename', 'ori_shape', 'img_shape', 'text',
                    'valid_ratio', 'resize_shape'
                ])
        ]),
    val=dict(
        type='UniformConcatDataset',
        datasets=[
            dict(
                type='OCRDataset',
                img_prefix=test_img_prefix1,
                ann_file=test_ann_file1,
                loader=dict(
                    type='AnnFileLoader',
                    repeat=1,
                    file_format='txt',
                    parser=dict(
                        type='LineStrParser',
                        keys=['filename', 'text'],
                        keys_idx=[0, 1],
                        separator=' ')),
                pipeline=None,
                test_mode=True)
        ],
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiRotateAugOCR',
                rotate_degrees=[0, 90, 270],
                transforms=[
                    dict(
                        type='ResizeOCR',
                        height=32,
                        min_width=128,
                        max_width=128,
                        keep_aspect_ratio=False,
                        width_downsample_ratio=0.25),
                    dict(type='ToTensorOCR'),
                    dict(
                        type='NormalizeOCR',
                        mean=[0.485, 0.456, 0.406],
                        std=[0.229, 0.224, 0.225]),
                    dict(
                        type='Collect',
                        keys=['img'],
                        meta_keys=[
                            'filename', 'ori_shape', 'img_shape',
                            'valid_ratio', 'resize_shape', 'img_norm_cfg',
                            'ori_filename'
                        ])
                ])
        ]),
    test=dict(
        type='UniformConcatDataset',
        datasets=[
            dict(
                type='OCRDataset',
                img_prefix=test_img_prefix1,
                ann_file=test_ann_file1,
                loader=dict(
                    type='AnnFileLoader',
                    repeat=1,
                    file_format='txt',
                    parser=dict(
                        type='LineStrParser',
                        keys=['filename', 'text'],
                        keys_idx=[0, 1],
                        separator=' ')),
                pipeline=None,
                test_mode=True)
        ],
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiRotateAugOCR',
                rotate_degrees=[0, 90, 270],
                transforms=[
                    dict(
                        type='ResizeOCR',
                        height=32,
                        min_width=128,
                        max_width=128,
                        keep_aspect_ratio=False,
                        width_downsample_ratio=0.25),
                    dict(type='ToTensorOCR'),
                    dict(
                        type='NormalizeOCR',
                        mean=[0.485, 0.456, 0.406],
                        std=[0.229, 0.224, 0.225]),
                    dict(
                        type='Collect',
                        keys=['img'],
                        meta_keys=[
                            'filename', 'ori_shape', 'img_shape',
                            'valid_ratio', 'resize_shape', 'img_norm_cfg',
                            'ori_filename'
                        ])
                ])
        ]))
evaluation = dict(interval=1, metric='acc')
work_dir = r'./work_dirs/abinet_vision_only_academic'
gpu_ids = [0]

4、常见报错处理

报错：

加载配置文件报错：

During handling of the above exception,another exception occurred:
Traceback (most recent call last):
File "c:\ProqramData\Winiconda\envs\pytorch\lib\site-packages(mcvutils(reqistry.py",line 72，in build.from_cfgraise type(e)(f'iobj_cls. __name__}: ie}')
1no.Envow : Jnifonwtonatiatoaset:0A就nAateset : AnmFileloder: E: ythrnXxX pytorch)datalmorcelmixturelicdarn.2815)train_label.txt:
python-BaseException