Skip to content

GitLab

"vscode:/vscode.git/clone" did not exist on "514fd0d1eaacdb36abe96f23bf53393b719a3b06"
Unverified Commit ee05c913 authored by zhoujun's avatar zhoujun Committed by GitHub
Browse files

Merge pull request #5 from PaddlePaddle/develop 

merge paddleocr
parents 7c09c97d 2bdaea56
Hide whitespace changes
Inline Side-by-side
Showing with 688 additions and 76 deletions
doc/doc_en/whl_en.md 0 → 100644
View file @ ee05c913
# paddleocr package
## Get started quickly
### install package
install by pypi
```bash
pip install paddleocr
```
build own whl package and install
```bash
python setup.py bdist_wheel
pip install dist/paddleocr-0.0.3-py3-none-any.whl
```
### 1. Use by code
* detection and recognition
```python
from paddleocr import PaddleOCR,draw_ocr
ocr = PaddleOCR() # need to run only once to download and load model into memory
img_path = 'PaddleOCR/doc/imgs_en/img_12.jpg'
result = ocr.ocr(img_path)
for line in result:
print(line)
# draw result
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='/path/to/PaddleOCR/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
Output will be a list, each item contains bounding box, text and recognition confidence
```bash
[[[442.0, 173.0], [1169.0, 173.0], [1169.0, 225.0], [442.0, 225.0]], ['ACKNOWLEDGEMENTS', 0.99283075]]
[[[393.0, 340.0], [1207.0, 342.0], [1207.0, 389.0], [393.0, 387.0]], ['We would like to thank all the designers and', 0.9357758]]
[[[399.0, 398.0], [1204.0, 398.0], [1204.0, 433.0], [399.0, 433.0]], ['contributors whohave been involved in the', 0.9592447]]
......
```
Visualization of results
<div align="center">
<img src="../imgs_results/whl/12_det_rec.jpg" width="800">
</div>
* only detection
```python
from paddleocr import PaddleOCR,draw_ocr
ocr = PaddleOCR() # need to run only once to download and load model into memory
img_path = 'PaddleOCR/doc/imgs_en/img_12.jpg'
result = ocr.ocr(img_path,rec=False)
for line in result:
print(line)
# draw result
from PIL import Image
image = Image.open(img_path).convert('RGB')
im_show = draw_ocr(image, result, txts=None, scores=None, font_path='/path/to/PaddleOCR/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
Output will be a list, each item only contains bounding box
```bash
[[756.0, 812.0], [805.0, 812.0], [805.0, 830.0], [756.0, 830.0]]
[[820.0, 803.0], [1085.0, 801.0], [1085.0, 836.0], [820.0, 838.0]]
[[393.0, 801.0], [715.0, 805.0], [715.0, 839.0], [393.0, 836.0]]
......
```
Visualization of results
<div align="center">
<img src="../imgs_results/whl/12_det.jpg" width="800">
</div>
* only recognition
```python
from paddleocr import PaddleOCR
ocr = PaddleOCR() # need to run only once to load model into memory
img_path = 'PaddleOCR/doc/imgs_words_en/word_10.png'
result = ocr.ocr(img_path,det=False)
for line in result:
print(line)
```
Output will be a list, each item contains text and recognition confidence
```bash
['PAIN', 0.990372]
```
### Use by command line
show help information
```bash
paddleocr -h
```
* detection and recognition
```bash
paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg
```
Output will be a list, each item contains bounding box, text and recognition confidence
```bash
[[[442.0, 173.0], [1169.0, 173.0], [1169.0, 225.0], [442.0, 225.0]], ['ACKNOWLEDGEMENTS', 0.99283075]]
[[[393.0, 340.0], [1207.0, 342.0], [1207.0, 389.0], [393.0, 387.0]], ['We would like to thank all the designers and', 0.9357758]]
[[[399.0, 398.0], [1204.0, 398.0], [1204.0, 433.0], [399.0, 433.0]], ['contributors whohave been involved in the', 0.9592447]]
......
```
* only detection
```bash
paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg --rec false
```
Output will be a list, each item only contains bounding box
```bash
[[756.0, 812.0], [805.0, 812.0], [805.0, 830.0], [756.0, 830.0]]
[[820.0, 803.0], [1085.0, 801.0], [1085.0, 836.0], [820.0, 838.0]]
[[393.0, 801.0], [715.0, 805.0], [715.0, 839.0], [393.0, 836.0]]
......
```
* only recognition
```bash
paddleocr --image_dir PaddleOCR/doc/imgs_words_en/word_10.png --det false
```
Output will be a list, each item contains text and recognition confidence
```bash
['PAIN', 0.990372]
```
## Use custom model
When the built-in model cannot meet the needs, you need to use your own trained model.
First, refer to the first section of [inference_en.md](./inference_en.md) to convert your det and rec model to inference model, and then use it as follows
### 1. Use by code
```python
from paddleocr import PaddleOCR,draw_ocr
# The path of detection and recognition model must contain model and params files
ocr = PaddleOCR(det_model_dir='{your_det_model_dir}',rec_model_dir='{your_rec_model_dir}å')
img_path = 'PaddleOCR/doc/imgs_en/img_12.jpg'
result = ocr.ocr(img_path)
for line in result:
print(line)
# draw result
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='/path/to/PaddleOCR/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
```
### Use by command line
```bash
paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --det_model_dir {your_det_model_dir} --rec_model_dir {your_rec_model_dir}
```
## Parameter Description
| Parameter | Description | Default value |
|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------|
| use_gpu | use GPU or not | TRUE |
| gpu_mem | GPU memory size used for initialization | 8000M |
| image_dir | The images path or folder path for predicting when used by the command line | |
| det_algorithm | Type of detection algorithm selected | DB |
| det_model_dir | the text detection inference model folder. There are two ways to transfer parameters, 1. None: Automatically download the built-in model to `~/.paddleocr/det`; 2. The path of the inference model converted by yourself, the model and params files must be included in the model path | None |
| det_max_side_len | The maximum size of the long side of the image. When the long side exceeds this value, the long side will be resized to this size, and the short side will be scaled proportionally | 960 |
| det_db_thresh | Binarization threshold value of DB output map | 0.3 |
| det_db_box_thresh | The threshold value of the DB output box. Boxes score lower than this value will be discarded | 0.5 |
| det_db_unclip_ratio | The expanded ratio of DB output box | 2 |
| det_east_score_thresh | Binarization threshold value of EAST output map | 0.8 |
| det_east_cover_thresh | The threshold value of the EAST output box. Boxes score lower than this value will be discarded | 0.1 |
| det_east_nms_thresh | The NMS threshold value of EAST model output box | 0.2 |
| rec_algorithm | Type of recognition algorithm selected | CRNN |
| rec_model_dir | the text recognition inference model folder. There are two ways to transfer parameters, 1. None: Automatically download the built-in model to `~/.paddleocr/rec`; 2. The path of the inference model converted by yourself, the model and params files must be included in the model path | None |
| rec_image_shape | image shape of recognition algorithm | "3,32,320" |
| rec_char_type | Character type of recognition algorithm, Chinese (ch) or English (en) | ch |
| rec_batch_num | When performing recognition, the batchsize of forward images | 30 |
| max_text_length | The maximum text length that the recognition algorithm can recognize | 25 |
| rec_char_dict_path | the alphabet path which needs to be modified to your own path when `rec_model_Name` use mode 2 | ./ppocr/utils/ppocr_keys_v1.txt |
| use_space_char | Whether to recognize spaces | TRUE |
| enable_mkldnn | Whether to enable mkldnn | FALSE |
| det | Enable detction when `ppocr.ocr` func exec | TRUE |
| rec | Enable detction when `ppocr.ocr` func exec | TRUE |
paddleocr.py 0 → 100644
View file @ ee05c913
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import sys
__dir__ = os.path.dirname(__file__)
sys.path.append(os.path.join(__dir__, ''))
import cv2
import numpy as np
from pathlib import Path
import tarfile
import requests
from tqdm import tqdm
from tools.infer import predict_system
from ppocr.utils.utility import initial_logger
logger = initial_logger()
from ppocr.utils.utility import check_and_read_gif, get_image_file_list
__all__ = ['PaddleOCR']
model_params = {
'det': 'https://paddleocr.bj.bcebos.com/ch_models/ch_det_mv3_db_infer.tar',
'rec':
'https://paddleocr.bj.bcebos.com/ch_models/ch_rec_mv3_crnn_enhance_infer.tar',
}
SUPPORT_DET_MODEL = ['DB']
SUPPORT_REC_MODEL = ['CRNN']
BASE_DIR = os.path.expanduser("~/.paddleocr/")
def download_with_progressbar(url, save_path):
response = requests.get(url, stream=True)
total_size_in_bytes = int(response.headers.get('content-length', 0))
block_size = 1024 # 1 Kibibyte
progress_bar = tqdm(total=total_size_in_bytes, unit='iB', unit_scale=True)
with open(save_path, 'wb') as file:
for data in response.iter_content(block_size):
progress_bar.update(len(data))
file.write(data)
progress_bar.close()
if total_size_in_bytes != 0 and progress_bar.n != total_size_in_bytes:
logger.error("ERROR, something went wrong")
sys.exit(0)
def maybe_download(model_storage_directory, url):
# using custom model
if not os.path.exists(os.path.join(
model_storage_directory, 'model')) or not os.path.exists(
os.path.join(model_storage_directory, 'params')):
tmp_path = os.path.join(model_storage_directory, url.split('/')[-1])
print('download {} to {}'.format(url, tmp_path))
os.makedirs(model_storage_directory, exist_ok=True)
download_with_progressbar(url, tmp_path)
with tarfile.open(tmp_path, 'r') as tarObj:
for member in tarObj.getmembers():
if "model" in member.name:
filename = 'model'
elif "params" in member.name:
filename = 'params'
else:
continue
file = tarObj.extractfile(member)
with open(
os.path.join(model_storage_directory, filename),
'wb') as f:
f.write(file.read())
os.remove(tmp_path)
def parse_args():
import argparse
def str2bool(v):
return v.lower() in ("true", "t", "1")
parser = argparse.ArgumentParser()
# params for prediction engine
parser.add_argument("--use_gpu", type=str2bool, default=True)
parser.add_argument("--ir_optim", type=str2bool, default=True)
parser.add_argument("--use_tensorrt", type=str2bool, default=False)
parser.add_argument("--gpu_mem", type=int, default=8000)
# params for text detector
parser.add_argument("--image_dir", type=str)
parser.add_argument("--det_algorithm", type=str, default='DB')
parser.add_argument("--det_model_dir", type=str, default=None)
parser.add_argument("--det_max_side_len", type=float, default=960)
# DB parmas
parser.add_argument("--det_db_thresh", type=float, default=0.3)
parser.add_argument("--det_db_box_thresh", type=float, default=0.5)
parser.add_argument("--det_db_unclip_ratio", type=float, default=2.0)
# EAST parmas
parser.add_argument("--det_east_score_thresh", type=float, default=0.8)
parser.add_argument("--det_east_cover_thresh", type=float, default=0.1)
parser.add_argument("--det_east_nms_thresh", type=float, default=0.2)
# params for text recognizer
parser.add_argument("--rec_algorithm", type=str, default='CRNN')
parser.add_argument("--rec_model_dir", type=str, default=None)
parser.add_argument("--rec_image_shape", type=str, default="3, 32, 320")
parser.add_argument("--rec_char_type", type=str, default='ch')
parser.add_argument("--rec_batch_num", type=int, default=30)
parser.add_argument("--max_text_length", type=int, default=25)
parser.add_argument(
"--rec_char_dict_path",
type=str,
default="./ppocr/utils/ppocr_keys_v1.txt")
parser.add_argument("--use_space_char", type=bool, default=True)
parser.add_argument("--enable_mkldnn", type=bool, default=False)
parser.add_argument("--det", type=str2bool, default=True)
parser.add_argument("--rec", type=str2bool, default=True)
return parser.parse_args()
class PaddleOCR(predict_system.TextSystem):
def __init__(self, **kwargs):
"""
paddleocr package
args:
**kwargs: other params show in paddleocr --help
"""
postprocess_params = parse_args()
postprocess_params.__dict__.update(**kwargs)
# init model dir
if postprocess_params.det_model_dir is None:
postprocess_params.det_model_dir = os.path.join(BASE_DIR, 'det')
if postprocess_params.rec_model_dir is None:
postprocess_params.rec_model_dir = os.path.join(BASE_DIR, 'rec')
print(postprocess_params)
# download model
maybe_download(postprocess_params.det_model_dir, model_params['det'])
maybe_download(postprocess_params.rec_model_dir, model_params['rec'])
if postprocess_params.det_algorithm not in SUPPORT_DET_MODEL:
logger.error('det_algorithm must in {}'.format(SUPPORT_DET_MODEL))
sys.exit(0)
if postprocess_params.rec_algorithm not in SUPPORT_REC_MODEL:
logger.error('rec_algorithm must in {}'.format(SUPPORT_REC_MODEL))
sys.exit(0)
postprocess_params.rec_char_dict_path = Path(
__file__).parent / postprocess_params.rec_char_dict_path
# init det_model and rec_model
super().__init__(postprocess_params)
def ocr(self, img, det=True, rec=True):
"""
ocr with paddleocr
args：
img: img for ocr, support ndarray, img_path and list or ndarray
det: use text detection or not, if false, only rec will be exec. default is True
rec: use text recognition or not, if false, only det will be exec. default is True
"""
assert isinstance(img, (np.ndarray, list, str))
if isinstance(img, str):
image_file = img
img, flag = check_and_read_gif(image_file)
if not flag:
img = cv2.imread(image_file)
if img is None:
logger.error("error in loading image:{}".format(image_file))
return None
if det and rec:
dt_boxes, rec_res = self.__call__(img)
return [[box.tolist(), res] for box, res in zip(dt_boxes, rec_res)]
elif det and not rec:
dt_boxes, elapse = self.text_detector(img)
if dt_boxes is None:
return None
return [box.tolist() for box in dt_boxes]
else:
if not isinstance(img, list):
img = [img]
rec_res, elapse = self.text_recognizer(img)
return rec_res
def main():
# for com
args = parse_args()
image_file_list = get_image_file_list(args.image_dir)
if len(image_file_list) == 0:
logger.error('no images find in {}'.format(args.image_dir))
return
ocr_engine = PaddleOCR()
for img_path in image_file_list:
print(img_path)
result = ocr_engine.ocr(img_path, det=args.det, rec=args.rec)
for line in result:
print(line)
ppocr/data/rec/dataset_traversal.py
View file @ ee05c913
......@@ -214,6 +214,8 @@ class SimpleReader(object):
self.mode = params['mode']
self.infer_img = params['infer_img']
self.use_tps = False
if "num_heads" in params:
self.num_heads = params['num_heads']
if "tps" in params:
self.use_tps = True
self.use_distort = False
......@@ -237,7 +239,7 @@ class SimpleReader(object):
def get_device_num():
if self.use_gpu:
gpus = os.environ.get("CUDA_VISIBLE_DEVICES", 1)
gpus = os.environ.get("CUDA_VISIBLE_DEVICES", '1')
gpu_num = len(gpus.split(','))
return gpu_num
else:
......@@ -251,12 +253,19 @@ class SimpleReader(object):
img = cv2.imread(single_img)
if img.shape[-1] == 1 or len(list(img.shape)) == 2:
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
norm_img = process_image(
img=img,
image_shape=self.image_shape,
char_ops=self.char_ops,
tps=self.use_tps,
infer_mode=True)
if self.loss_type == 'srn':
norm_img = process_image_srn(
img=img,
image_shape=self.image_shape,
num_heads=self.num_heads,
max_text_length=self.max_text_length)
else:
norm_img = process_image(
img=img,
image_shape=self.image_shape,
char_ops=self.char_ops,
tps=self.use_tps,
infer_mode=True)
yield norm_img
else:
with open(self.label_file_path, "rb") as fin:
......@@ -286,14 +295,25 @@ class SimpleReader(object):
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
label = substr[1]
outs = process_image(
img=img,
image_shape=self.image_shape,
label=label,
char_ops=self.char_ops,
loss_type=self.loss_type,
max_text_length=self.max_text_length,
distort=self.use_distort)
if self.loss_type == "srn":
outs = process_image_srn(
img=img,
image_shape=self.image_shape,
num_heads=self.num_heads,
max_text_length=self.max_text_length,
label=label,
char_ops=self.char_ops,
loss_type=self.loss_type)
else:
outs = process_image(
img=img,
image_shape=self.image_shape,
label=label,
char_ops=self.char_ops,
loss_type=self.loss_type,
max_text_length=self.max_text_length,
distort=self.use_distort)
if outs is None:
continue
yield outs
......
ppocr/data/rec/img_tools.py
View file @ ee05c913
......@@ -410,7 +410,8 @@ def resize_norm_img_srn(img, image_shape):
def srn_other_inputs(image_shape,
num_heads,
max_text_length):
max_text_length,
char_num):
imgC, imgH, imgW = image_shape
feature_dim = int((imgH / 8) * (imgW / 8))
......@@ -418,7 +419,7 @@ def srn_other_inputs(image_shape,
encoder_word_pos = np.array(range(0, feature_dim)).reshape((feature_dim, 1)).astype('int64')
gsrm_word_pos = np.array(range(0, max_text_length)).reshape((max_text_length, 1)).astype('int64')
lbl_weight = np.array([37] * max_text_length).reshape((-1,1)).astype('int64')
lbl_weight = np.array([int(char_num-1)] * max_text_length).reshape((-1,1)).astype('int64')
gsrm_attn_bias_data = np.ones((1, max_text_length, max_text_length))
gsrm_slf_attn_bias1 = np.triu(gsrm_attn_bias_data, 1).reshape([-1, 1, max_text_length, max_text_length])
......@@ -441,17 +442,18 @@ def process_image_srn(img,
loss_type=None):
norm_img = resize_norm_img_srn(img, image_shape)
norm_img = norm_img[np.newaxis, :]
char_num = char_ops.get_char_num()
[lbl_weight, encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, gsrm_slf_attn_bias2] = \
srn_other_inputs(image_shape, num_heads, max_text_length)
srn_other_inputs(image_shape, num_heads, max_text_length,char_num)
if label is not None:
char_num = char_ops.get_char_num()
text = char_ops.encode(label)
if len(text) == 0 or len(text) > max_text_length:
return None
else:
if loss_type == "srn":
text_padded = [37] * max_text_length
text_padded = [int(char_num-1)] * max_text_length
for i in range(len(text)):
text_padded[i] = text[i]
lbl_weight[i] = [1.0]
......
ppocr/modeling/backbones/rec_resnet50_fpn.py → ppocr/modeling/backbones/rec_resnet_fpn.py
View file @ ee05c913
......@@ -22,12 +22,12 @@ import paddle
import paddle.fluid as fluid
from paddle.fluid.param_attr import ParamAttr
__all__ = ["ResNet", "ResNet18", "ResNet34", "ResNet50", "ResNet101", "ResNet152"]
__all__ = [
"ResNet", "ResNet18", "ResNet34", "ResNet50", "ResNet101", "ResNet152"
]
Trainable = True
w_nolr = fluid.ParamAttr(
trainable = Trainable)
w_nolr = fluid.ParamAttr(trainable=Trainable)
train_parameters = {
"input_size": [3, 224, 224],
"input_mean": [0.485, 0.456, 0.406],
......@@ -40,12 +40,12 @@ train_parameters = {
}
}
class ResNet():
def __init__(self, params):
self.layers = params['layers']
self.params = train_parameters
def __call__(self, input):
layers = self.layers
supported_layers = [18, 34, 50, 101, 152]
......@@ -60,12 +60,17 @@ class ResNet():
depth = [3, 4, 23, 3]
elif layers == 152:
depth = [3, 8, 36, 3]
stride_list = [(2,2),(2,2),(1,1),(1,1)]
stride_list = [(2, 2), (2, 2), (1, 1), (1, 1)]
num_filters = [64, 128, 256, 512]
conv = self.conv_bn_layer(
input=input, num_filters=64, filter_size=7, stride=2, act='relu', name="conv1")
F = []
input=input,
num_filters=64,
filter_size=7,
stride=2,
act='relu',
name="conv1")
F = []
if layers >= 50:
for block in range(len(depth)):
for i in range(depth[block]):
......@@ -79,26 +84,67 @@ class ResNet():
conv = self.bottleneck_block(
input=conv,
num_filters=num_filters[block],
stride=stride_list[block] if i == 0 else 1, name=conv_name)
stride=stride_list[block] if i == 0 else 1,
name=conv_name)
F.append(conv)
else:
for block in range(len(depth)):
for i in range(depth[block]):
conv_name = "res" + str(block + 2) + chr(97 + i)
if i == 0 and block != 0:
stride = (2, 1)
else:
stride = (1, 1)
conv = self.basic_block(
input=conv,
num_filters=num_filters[block],
stride=stride,
if_first=block == i == 0,
name=conv_name)
F.append(conv)
base = F[-1]
for i in [-2, -3]:
for i in [-2, -3]:
b, c, w, h = F[i].shape
if (w,h) == base.shape[2:]:
if (w, h) == base.shape[2:]:
base = base
else:
base = fluid.layers.conv2d_transpose( input=base, num_filters=c,filter_size=4, stride=2,
padding=1,act=None,
base = fluid.layers.conv2d_transpose(
input=base,
num_filters=c,
filter_size=4,
stride=2,
padding=1,
act=None,
param_attr=w_nolr,
bias_attr=w_nolr)
base = fluid.layers.batch_norm(base, act = "relu", param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.batch_norm(
base, act="relu", param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.concat([base, F[i]], axis=1)
base = fluid.layers.conv2d(base, num_filters=c, filter_size=1, param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.conv2d(base, num_filters=c, filter_size=3,padding = 1, param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.batch_norm(base, act = "relu", param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.conv2d(base, num_filters=512, filter_size=1,bias_attr=w_nolr,param_attr=w_nolr)
base = fluid.layers.conv2d(
base,
num_filters=c,
filter_size=1,
param_attr=w_nolr,
bias_attr=w_nolr)
base = fluid.layers.conv2d(
base,
num_filters=c,
filter_size=3,
padding=1,
param_attr=w_nolr,
bias_attr=w_nolr)
base = fluid.layers.batch_norm(
base, act="relu", param_attr=w_nolr, bias_attr=w_nolr)
base = fluid.layers.conv2d(
base,
num_filters=512,
filter_size=1,
bias_attr=w_nolr,
param_attr=w_nolr)
return base
......@@ -113,13 +159,14 @@ class ResNet():
conv = fluid.layers.conv2d(
input=input,
num_filters=num_filters,
filter_size= 2 if stride==(1,1) else filter_size,
dilation = 2 if stride==(1,1) else 1,
filter_size=2 if stride == (1, 1) else filter_size,
dilation=2 if stride == (1, 1) else 1,
stride=stride,
padding=(filter_size - 1) // 2,
groups=groups,
act=None,
param_attr=ParamAttr(name=name + "_weights",trainable = Trainable),
param_attr=ParamAttr(
name=name + "_weights", trainable=Trainable),
bias_attr=False,
name=name + '.conv2d.output.1')
......@@ -127,28 +174,35 @@ class ResNet():
bn_name = "bn_" + name
else:
bn_name = "bn" + name[3:]
return fluid.layers.batch_norm(input=conv,
act=act,
name=bn_name + '.output.1',
param_attr=ParamAttr(name=bn_name + '_scale',trainable = Trainable),
bias_attr=ParamAttr(bn_name + '_offset',trainable = Trainable),
moving_mean_name=bn_name + '_mean',
moving_variance_name=bn_name + '_variance', )
return fluid.layers.batch_norm(
input=conv,
act=act,
name=bn_name + '.output.1',
param_attr=ParamAttr(
name=bn_name + '_scale', trainable=Trainable),
bias_attr=ParamAttr(
bn_name + '_offset', trainable=Trainable),
moving_mean_name=bn_name + '_mean',
moving_variance_name=bn_name + '_variance', )
def shortcut(self, input, ch_out, stride, is_first, name):
ch_in = input.shape[1]
if ch_in != ch_out or stride != 1 or is_first == True:
if stride == (1,1):
if stride == (1, 1):
return self.conv_bn_layer(input, ch_out, 1, 1, name=name)
else: #stride == (2,2)
else: #stride == (2,2)
return self.conv_bn_layer(input, ch_out, 1, stride, name=name)
else:
return input
def bottleneck_block(self, input, num_filters, stride, name):
conv0 = self.conv_bn_layer(
input=input, num_filters=num_filters, filter_size=1, act='relu', name=name + "_branch2a")
input=input,
num_filters=num_filters,
filter_size=1,
act='relu',
name=name + "_branch2a")
conv1 = self.conv_bn_layer(
input=conv0,
num_filters=num_filters,
......@@ -157,16 +211,36 @@ class ResNet():
act='relu',
name=name + "_branch2b")
conv2 = self.conv_bn_layer(
input=conv1, num_filters=num_filters * 4, filter_size=1, act=None, name=name + "_branch2c")
input=conv1,
num_filters=num_filters * 4,
filter_size=1,
act=None,
name=name + "_branch2c")
short = self.shortcut(input, num_filters * 4, stride, is_first=False, name=name + "_branch1")
short = self.shortcut(
input,
num_filters * 4,
stride,
is_first=False,
name=name + "_branch1")
return fluid.layers.elementwise_add(x=short, y=conv2, act='relu', name=name + ".add.output.5")
return fluid.layers.elementwise_add(
x=short, y=conv2, act='relu', name=name + ".add.output.5")
def basic_block(self, input, num_filters, stride, is_first, name):
conv0 = self.conv_bn_layer(input=input, num_filters=num_filters, filter_size=3, act='relu', stride=stride,
name=name + "_branch2a")
conv1 = self.conv_bn_layer(input=conv0, num_filters=num_filters, filter_size=3, act=None,
name=name + "_branch2b")
short = self.shortcut(input, num_filters, stride, is_first, name=name + "_branch1")
conv0 = self.conv_bn_layer(
input=input,
num_filters=num_filters,
filter_size=3,
act='relu',
stride=stride,
name=name + "_branch2a")
conv1 = self.conv_bn_layer(
input=conv0,
num_filters=num_filters,
filter_size=3,
act=None,
name=name + "_branch2b")
short = self.shortcut(
input, num_filters, stride, is_first, name=name + "_branch1")
return fluid.layers.elementwise_add(x=short, y=conv1, act='relu')
ppocr/modeling/heads/det_sast_head.py
View file @ ee05c913
......@@ -49,7 +49,7 @@ class SASTHead(object):
for i in range(4):
if i == 0:
g[i] = deconv_bn_layer(input=h[i], num_filters=num_outputs[i + 1], act=None, name='fpn_up_g0')
print("g[{}] shape: {}".format(i, g[i].shape))
#print("g[{}] shape: {}".format(i, g[i].shape))
else:
g[i] = fluid.layers.elementwise_add(x=g[i - 1], y=h[i])
g[i] = fluid.layers.relu(g[i])
......@@ -58,7 +58,7 @@ class SASTHead(object):
g[i] = conv_bn_layer(input=g[i], num_filters=num_outputs[i],
filter_size=3, stride=1, act='relu', name='fpn_up_g%d_1'%i)
g[i] = deconv_bn_layer(input=g[i], num_filters=num_outputs[i + 1], act=None, name='fpn_up_g%d_2'%i)
print("g[{}] shape: {}".format(i, g[i].shape))
#print("g[{}] shape: {}".format(i, g[i].shape))
g[4] = fluid.layers.elementwise_add(x=g[3], y=h[4])
g[4] = fluid.layers.relu(g[4])
......
ppocr/optimizer.py
View file @ ee05c913
......@@ -14,14 +14,50 @@
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import math
import paddle.fluid as fluid
from paddle.fluid.regularizer import L2Decay
from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
import paddle.fluid.layers.ops as ops
from ppocr.utils.utility import initial_logger
logger = initial_logger()
def cosine_decay_with_warmup(learning_rate,
step_each_epoch,
epochs=500,
warmup_minibatch=1000):
"""Applies cosine decay to the learning rate.
lr = 0.05 * (math.cos(epoch * (math.pi / 120)) + 1)
decrease lr for every mini-batch and start with warmup.
"""
global_step = _decay_step_counter()
lr = fluid.layers.tensor.create_global_var(
shape=[1],
value=0.0,
dtype='float32',
persistable=True,
name="learning_rate")
warmup_minibatch = fluid.layers.fill_constant(
shape=[1],
dtype='float32',
value=float(warmup_minibatch),
force_cpu=True)
with fluid.layers.control_flow.Switch() as switch:
with switch.case(global_step < warmup_minibatch):
decayed_lr = learning_rate * (1.0 * global_step / warmup_minibatch)
fluid.layers.tensor.assign(input=decayed_lr, output=lr)
with switch.default():
decayed_lr = learning_rate * \
(ops.cos((global_step - warmup_minibatch) * (math.pi / (epochs * step_each_epoch))) + 1)/2
fluid.layers.tensor.assign(input=decayed_lr, output=lr)
return lr
def AdamDecay(params, parameter_list=None):
"""
define optimizer function
......@@ -36,7 +72,9 @@ def AdamDecay(params, parameter_list=None):
l2_decay = params.get("l2_decay", 0.0)
if 'decay' in params:
supported_decay_mode = ["cosine_decay", "piecewise_decay"]
supported_decay_mode = [
"cosine_decay", "cosine_decay_warmup", "piecewise_decay"
]
params = params['decay']
decay_mode = params['function']
assert decay_mode in supported_decay_mode, "Supported decay mode is {}, but got {}".format(
......@@ -49,6 +87,15 @@ def AdamDecay(params, parameter_list=None):
learning_rate=base_lr,
step_each_epoch=step_each_epoch,
epochs=total_epoch)
elif decay_mode == "cosine_decay_warmup":
step_each_epoch = params['step_each_epoch']
total_epoch = params['total_epoch']
warmup_minibatch = params.get("warmup_minibatch", 1000)
base_lr = cosine_decay_with_warmup(
learning_rate=base_lr,
step_each_epoch=step_each_epoch,
epochs=total_epoch,
warmup_minibatch=warmup_minibatch)
elif decay_mode == "piecewise_decay":
boundaries = params["boundaries"]
decay_rate = params["decay_rate"]
......@@ -104,5 +151,5 @@ def RMSProp(params, parameter_list=None):
optimizer = fluid.optimizer.RMSProp(
learning_rate=base_lr,
regularization=fluid.regularizer.L2Decay(regularization_coeff=l2_decay))
return optimizer
\ No newline at end of file
return optimizer
ppocr/postprocess/east_postprocess.py
View file @ ee05c913
......@@ -22,9 +22,9 @@ import cv2
import os
import sys
__dir__ = os.path.dirname(__file__)
__dir__ = os.path.dirname(os.path.abspath(__file__))
sys.path.append(__dir__)
sys.path.append(os.path.join(__dir__, '..'))
sys.path.append(os.path.abspath(os.path.join(__dir__, '..')))
class EASTPostPocess(object):
......
ppocr/postprocess/lanms/.ycm_extra_conf.py
View file @ ee05c913
......@@ -25,7 +25,7 @@ import ycm_core
# These are the compilation flags that will be used in case there's no
# compilation database set (by default, one is not set).
# CHANGE THIS LIST OF FLAGS. YES, THIS IS THE DROID YOU HAVE BEEN LOOKING FOR.
sys.path.append(os.path.dirname(__file__))
sys.path.append(os.path.dirname(os.path.abspath(__file__)))
BASE_DIR = os.path.dirname(os.path.realpath(__file__))
......
ppocr/utils/character.py
View file @ ee05c913
......@@ -26,8 +26,6 @@ class CharacterOps(object):
self.character_type = config['character_type']
self.loss_type = config['loss_type']
self.max_text_len = config['max_text_length']
if self.loss_type == "srn" and self.character_type != "en":
raise Exception("SRN can only support in character_type == en")
if self.character_type == "en":
self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz"
dict_character = list(self.character_str)
......@@ -160,13 +158,15 @@ def cal_predicts_accuracy_srn(char_ops,
acc_num = 0
img_num = 0
char_num = char_ops.get_char_num()
total_len = preds.shape[0]
img_num = int(total_len / max_text_len)
for i in range(img_num):
cur_label = []
cur_pred = []
for j in range(max_text_len):
if labels[j + i * max_text_len] != 37: #0
if labels[j + i * max_text_len] != int(char_num-1): #0
cur_label.append(labels[j + i * max_text_len][0])
else:
break
......@@ -178,7 +178,7 @@ def cal_predicts_accuracy_srn(char_ops,
elif j == len(cur_label) and j == max_text_len:
acc_num += 1
break
elif j == len(cur_label) and preds[j + i * max_text_len][0] == 37:
elif j == len(cur_label) and preds[j + i * max_text_len][0] == int(char_num-1):
acc_num += 1
break
acc = acc_num * 1.0 / img_num
......
requirments.txt
View file @ ee05c913
shapely
imgaug
pyclipper
lmdb
\ No newline at end of file
lmdb
tqdm
numpy
\ No newline at end of file
setup.py 0 → 100644
View file @ ee05c913
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from setuptools import setup
from io import open
with open('requirments.txt', encoding="utf-8-sig") as f:
requirements = f.readlines()
requirements.append('tqdm')
def readme():
with open('doc/doc_en/whl_en.md', encoding="utf-8-sig") as f:
README = f.read()
return README
setup(
name='paddleocr',
packages=['paddleocr'],
package_dir={'paddleocr': ''},
include_package_data=True,
entry_points={"console_scripts": ["paddleocr= paddleocr.paddleocr:main"]},
version='0.0.3',
install_requires=requirements,
license='Apache License 2.0',
description='Awesome OCR toolkits based on PaddlePaddle （8.6M ultra-lightweight pre-trained model, support training and deployment among server, mobile, embeded and IoT devices',
long_description=readme(),
long_description_content_type='text/markdown',
url='https://github.com/PaddlePaddle/PaddleOCR',
download_url='https://github.com/PaddlePaddle/PaddleOCR.git',
keywords=[
'ocr textdetection textrecognition paddleocr crnn east star-net rosetta ocrlite db chineseocr chinesetextdetection chinesetextrecognition'
],
classifiers=[
'Intended Audience :: Developers', 'Operating System :: OS Independent',
'Natural Language :: Chinese (Simplified)',
'Programming Language :: Python :: 3',
'Programming Language :: Python :: 3.2',
'Programming Language :: Python :: 3.3',
'Programming Language :: Python :: 3.4',
'Programming Language :: Python :: 3.5',
'Programming Language :: Python :: 3.6',
'Programming Language :: Python :: 3.7', 'Topic :: Utilities'
], )
tools/eval.py
View file @ ee05c913
......@@ -18,9 +18,9 @@ from __future__ import print_function
import os
import sys
__dir__ = os.path.dirname(__file__)
__dir__ = os.path.dirname(os.path.abspath(__file__))
sys.path.append(__dir__)
sys.path.append(os.path.join(__dir__, '..'))
sys.path.append(os.path.abspath(os.path.join(__dir__, '..')))
def set_paddle_flags(**kwargs):
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment