Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,19 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+background.jpg filter=lfs diff=lfs merge=lfs -text
+flagged/image/5ad4b2887a6e21ed21ad368cb12b06241a8c06b4/tmpir4uhzvh.jpg filter=lfs diff=lfs merge=lfs -text
+flagged/image/c60b52ea5b844eb28fc764fc77bee9bd47ce39a9/tmp_dno9712.jpg filter=lfs diff=lfs merge=lfs -text
+flagged/image/fc61756a7b0971534c5b49be71caf17d819f35ad/tmpv0t201p6.jpg filter=lfs diff=lfs merge=lfs -text
+flagged/output/68a659c360681de00197f50432d89fbd98efdefc/tmp_cajstfi.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:34:30[[:space:]]@2021-02-03/144/test/mark_new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:34:30[[:space:]]@2021-02-03/144/test/new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:52:58[[:space:]]@2021-02-03/144/test/mark_new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:52:58[[:space:]]@2021-02-03/144/test/new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:55:35[[:space:]]@2021-02-03/144/test/mark_new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:55:35[[:space:]]@2021-02-03/144/test/new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:56:55[[:space:]]@2021-02-03/144/test/mark_new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:56:55[[:space:]]@2021-02-03/144/test/new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:59:53[[:space:]]@2021-02-03/144/test/mark_new_image_1.png filter=lfs diff=lfs merge=lfs -text
+flat/2023-07-13/2023-07-13[[:space:]]18:59:53[[:space:]]@2021-02-03/144/test/new_image_1.png filter=lfs diff=lfs merge=lfs -text
+testdata/new_image_1.jpg filter=lfs diff=lfs merge=lfs -text

ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2afacdaa13ab08049326e835064ecf6ae6873a10ecc3e2b8d4b4aa486e9b3c20
+size 159331175

ICDAR2021/2021-02-03 16:15:55/143/model_parameter

@@ -0,0 +1,5 @@
+Download the model parameters from the link below:
+
+Google：  https://drive.google.com/file/d/1rKS03wHMoAXp8uR1h3z7NAVoBY84NZlp/view?usp=sharing
+
+Baidu:   https://pan.baidu.com/s/1eSoiE3LoSQmTIL-SpbzPRw?pwd=rngi       password: rngi 

README.md

@@ -1,12 +1,56 @@
 ---
 title: Source
-emoji: 🏢
-colorFrom: red
-colorTo: purple
+app_file: demo.py
 sdk: gradio
 sdk_version: 3.40.1
-app_file: app.py
-pinned: false
 ---
+# Description
+Due to work, more information will be released.
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Performance
+**Execution Time when getting a 1024x960 corrected image.**
+|   Method  | Time  |
+|  ----  | ----  |
+| TPS  | - |
+| Interpolation  | - |
+<!-- 
+<small>\* TPS:flatByfiducial_TPS,
+  
+  Interpolation:flatByfiducial_interpolation</small> -->
+
+# Testing
+1、Download model parameter and source codes 
+
+2、Resize the input image into 992x992
+
+3、Running 
+
+- In GPU 0:
+`python test.py --data_path_test=./your/test/data/path/ --parallel 0 --schema test --batch_size 1`
+
+- In CPU:
+`python test.py --data_path_test=./your/test/data/path/ --parallel None --schema test --batch_size 1`
+
+# Training
+a) Download training data in [here](https://github.com/gwxie/Document-Dewarping-with-Control-Points/tree/main/Source/dataset/fiducial1024).
+
+b) Run `python train.py --data_path_train=./your/train/data/path/ --data_path_validate=./your/validate/data/path/ --data_path_test=./your/test/data/path/ --batch_size 32 --schema train --parallel 01`
+
+# Use your Dataset
+The training dataset can be synthesised using the [scripts](https://github.com/gwxie/Synthesize-Distorted-Image-and-Its-Control-Points).
+
+# Q&A
+1. Q:How to adjust the output image resolution？
+A: Adjusting the ‘flat_shap’ or 'output_shape' in utilsV4.py.
+https://github.com/gwxie/Document-Dewarping-with-Control-Points/blob/4c74853b0eb93f7c6006a774c2eb42c64f363531/Source/utilsV4.py#L99
+or
+https://github.com/gwxie/Document-Dewarping-with-Control-Points/blob/4c74853b0eb93f7c6006a774c2eb42c64f363531/Source/utilsV4.py#L162
+
+2. Q:How to open '*.gw' files?
+https://github.com/gwxie/Document-Dewarping-with-Control-Points/blob/0f4e9ac577fb001a719fb63b05cfa915fe3c9866/Source/dataloader.py#L146
+'im' is input images; 'lbl' is control points; 'segment' is the intervals of points between the horizontal and vertical directions;
+
+3. Q:How to train the model again with 61 points?
+https://github.com/gwxie/Document-Dewarping-with-Control-Points/blob/1ad92be2995ee7f4ecdd04157350f2b44ecbd7e9/Source/dataloader.py#L185-L186
+https://github.com/gwxie/Document-Dewarping-with-Control-Points/blob/1ad92be2995ee7f4ecdd04157350f2b44ecbd7e9/Source/dataloader.py#L74-L75
+set  self.col_gap = 0; self.row_gap = 0

css.css

@@ -0,0 +1,3 @@
+body {
+    background-color: #bd1414 !important;
+}

dataloader.py

@@ -0,0 +1,189 @@
+import os
+import pickle
+from os.path import join as pjoin
+import collections
+import json
+import torch
+import numpy as np
+
+import re
+import cv2
+
+from torch.utils import data
+
+
+def get_data_path(name):
+	"""Extract path to data from config file.
+
+	Args:
+		name (str): The name of the dataset.
+
+	Returns:
+		(str): The path to the root directory containing the dataset.
+	"""
+	with open('../xgw/segmentation/config.json') as f:
+		js = f.read()
+	# js = open('config.json').read()
+	data = json.loads(js)
+	return os.path.expanduser(data[name]['data_path'])
+
+def getDatasets(dir):
+	return os.listdir(dir)
+'''
+Resize the input image into 1024x960 (zooming in or out along the longest side and keeping the aspect ration, then filling zero for padding. )
+'''
+def resize_image(origin_img, long_edge=1024, short_edge=960):
+	# long_edge, short_edge = 2048, 1920
+	# long_edge, short_edge = 1024, 960
+	# long_edge, short_edge = 512, 480
+
+	im_lr = origin_img.shape[0]
+	im_ud = origin_img.shape[1]
+	new_img = np.zeros([long_edge, short_edge, 3], dtype=np.uint8)
+	new_shape = new_img.shape[:2]
+	if im_lr > im_ud:
+		img_shrink, base_img_shrink = long_edge, long_edge
+		im_ud = int(im_ud / im_lr * base_img_shrink)
+		im_ud += 32-im_ud%32
+		im_ud = min(im_ud, short_edge)
+		im_lr = img_shrink
+		origin_img = cv2.resize(origin_img, (im_ud, im_lr), interpolation=cv2.INTER_CUBIC)
+		new_img[:, (new_shape[1]-im_ud)//2:new_shape[1]-(new_shape[1]-im_ud)//2] = origin_img
+		# mask = np.full(new_shape, 255, dtype='uint8')
+		# mask[:, (new_shape[1] - im_ud) // 2:new_shape[1] - (new_shape[1] - im_ud) // 2] = 0
+	else:
+		img_shrink, base_img_shrink = short_edge, short_edge
+		im_lr = int(im_lr / im_ud * base_img_shrink)
+		im_lr += 32-im_lr%32
+		im_lr = min(im_lr, long_edge)
+		im_ud = img_shrink
+		origin_img = cv2.resize(origin_img, (im_ud, im_lr), interpolation=cv2.INTER_CUBIC)
+		new_img[(new_shape[0] - im_lr) // 2:new_shape[0] - (new_shape[0] - im_lr) // 2, :] = origin_img
+	return new_img
+
+class PerturbedDatastsForFiducialPoints_pickle_color_v2_v2(data.Dataset):
+	def __init__(self, root, split='1-1', img_shrink=None, is_return_img_name=False, preproccess=False):
+		self.root = os.path.expanduser(root)
+		self.split = split
+		self.img_shrink = img_shrink
+		self.is_return_img_name = is_return_img_name
+		self.preproccess = preproccess
+		# self.mean = np.array([104.00699, 116.66877, 122.67892])
+		self.images = collections.defaultdict(list)
+		self.labels = collections.defaultdict(list)
+		self.row_gap = 1  # value:0, 1, 2;  POINTS NUM: 61, 31, 21
+		self.col_gap = 1
+		datasets = ['validate', 'test', 'train']
+
+		if self.split == 'test' or self.split == 'eval':
+			img_file_list = getDatasets(os.path.join(self.root))
+			self.images[self.split] = img_file_list
+			# self.images[self.split] = sorted(img_file_list, key=lambda num: (
+			# int(re.match(r'(\d+)_(\d+)( copy.png)', num, re.IGNORECASE).group(1)), int(re.match(r'(\d+)_(\d+)( copy.png)', num, re.IGNORECASE).group(2))))
+		elif self.split in datasets:
+			img_file_list = []
+			img_file_list_ = getDatasets(os.path.join(self.root, 'color'))
+			for id_ in img_file_list_:
+				img_file_list.append(id_.rstrip())
+
+			self.images[self.split] = sorted(img_file_list, key=lambda num: (
+			re.match(r'(\w+\d*)_(\d+)_(\d+)_(\w+)', num, re.IGNORECASE).group(1), int(re.match(r'(\w+\d*)_(\d+)_(\d+)_(\w+)', num, re.IGNORECASE).group(2))
+			, int(re.match(r'(\w+\d*)_(\d+)_(\d+)_(\w+)', num, re.IGNORECASE).group(3)), re.match(r'(\w+\d*)_(\d+)_(\d+)_(\w+)', num, re.IGNORECASE).group(4)))
+		else:
+			raise Exception('load data error')
+		# self.checkImg()
+
+	def checkImg(self):
+		if self.split == 'validate':
+			for im_name in self.images[self.split]:
+				# if 'SinglePage' in im_name:
+				im_path = pjoin(self.root, self.split, 'color', im_name)
+				try:
+					with open(im_path, 'rb') as f:
+						perturbed_data = pickle.load(f)
+
+					im_shape = perturbed_data.shape
+				except:
+					print(im_name)
+					# os.remove(im_path)
+
+	def __len__(self):
+		return len(self.images[self.split])
+
+	def __getitem__(self, item):
+		if self.split == 'test':
+			im_name = self.images[self.split][item]
+			im_path = pjoin(self.root, im_name)
+
+			im = cv2.imread(im_path, flags=cv2.IMREAD_COLOR)
+
+			im = self.resize_im(im)
+			im = self.transform_im(im)
+
+			if self.is_return_img_name:
+				return im, im_name
+			return im
+		elif self.split == 'eval':
+			im_name = self.images[self.split][item]
+			im_path = pjoin(self.root, im_name)
+
+			img = cv2.imread(im_path, flags=cv2.IMREAD_COLOR)
+
+			im = self.resize_im(img)
+			im = self.transform_im(im)
+
+			if self.is_return_img_name:
+				return im, im_name
+			return im, img
+			# return im, img, im_name
+
+		else:
+			im_name = self.images[self.split][item]
+
+			im_path = pjoin(self.root, 'color', im_name)
+
+			with open(im_path, 'rb') as f:
+				perturbed_data = pickle.load(f)
+
+			im = perturbed_data.get('image')
+			lbl = perturbed_data.get('fiducial_points')
+			segment = perturbed_data.get('segment')
+
+			im = self.resize_im(im)
+			im = im.transpose(2, 0, 1)
+
+			lbl = self.resize_lbl(lbl)
+			lbl, segment = self.fiducal_points_lbl(lbl, segment)
+			lbl = lbl.transpose(2, 0, 1)
+
+			im = torch.from_numpy(im)
+			lbl = torch.from_numpy(lbl).float()
+			segment = torch.from_numpy(segment).float()
+
+			if self.is_return_img_name:
+				return im, lbl, segment, im_name
+
+			return im, lbl, segment
+
+	def transform_im(self, im):
+		im = im.transpose(2, 0, 1)
+		im = torch.from_numpy(im).float()
+
+		return im
+
+	def resize_im(self, im):
+		im = cv2.resize(im, (992, 992), interpolation=cv2.INTER_LINEAR)
+		# im = cv2.resize(im, (496, 496), interpolation=cv2.INTER_LINEAR)
+		return im
+
+	def resize_lbl(self, lbl):
+		lbl = lbl/[960, 1024]*[992, 992]
+		# lbl = lbl/[960, 1024]*[496, 496]
+		return lbl
+
+	def fiducal_points_lbl(self, fiducial_points, segment):
+
+		fiducial_point_gaps = [1, 2, 3, 4, 5, 6, 10, 12, 15, 20, 30, 60]  # POINTS NUM: 61, 31, 21, 16, 13, 11, 7, 6, 5, 4, 3, 2
+		fiducial_points = fiducial_points[::fiducial_point_gaps[self.row_gap], ::fiducial_point_gaps[self.col_gap], :]
+		segment = segment * [fiducial_point_gaps[self.col_gap], fiducial_point_gaps[self.row_gap]]
+		return fiducial_points, segment

dataset/fiducial1024/README.md

@@ -0,0 +1,11 @@
+# Download training Data
+
+Link: https://pan.baidu.com/s/1bZ9B7oZuxsMNRDZGz89Mbg?pwd=9fck  
+
+Password: 9fck
+
+
+# unzip
+```bash
+cat fiducial1024.tar.gz_0* > fiducial1024.tar.gz | tar -zxv
+```

demo.py

@@ -0,0 +1,65 @@
+import gradio as gr
+from PIL import Image
+import os
+import subprocess
+import glob
+
+
+def get_latest_file(directory):
+    # 获取所有的文件和文件夹
+    all_items = glob.glob(os.path.join(directory, '*'))
+
+    # 分离文件和文件夹
+    folders = [item for item in all_items if os.path.isdir(item)]
+    files = [item for item in all_items if os.path.isfile(item)]
+
+        # 先检查文件夹，如果有文件夹，则返回最新的文件夹
+    if folders:
+        latest_folder = max(folders, key=os.path.getmtime)
+        return latest_folder
+
+        # 如果没有文件夹，那么检查文件，并返回最新的文件
+    elif files:
+        latest_file = max(files, key=os.path.getmtime)
+        return latest_file
+
+        # 如果没有文件和文件夹，那么返回None
+    else:
+        return None
+
+def save_image(image):
+    # 指定目录
+    directory = '/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/testdata'
+
+    # 检查目录是否存在，不存在则创建
+    if not os.path.exists(directory):
+        os.makedirs(directory)
+
+    # 保存图片到指定目录
+    image.save(os.path.join(directory, 'new_image_1.jpg'))
+
+def process_image(image):
+    # 使用PIL.Image将numpy数组转换回图像
+    pil_image = Image.fromarray(image)
+
+    # 将图像保存到指定目录
+    save_image(pil_image)
+
+    program_path = '/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/test.py'
+    subprocess.run(['python3', program_path, '--data_path_test=./testdata'])
+
+    directory_image = '/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'  # 你的目录路径
+    latest_file = get_latest_file(get_latest_file(directory_image))
+    image_path = f'{latest_file}/144/test/new_image_1.jpg'
+    new_image = Image.open(image_path)
+    return new_image
+
+iface = gr.Interface(fn=process_image,
+                     inputs="image",
+                     outputs="image",
+                     title="Document dewarping platform",
+                     description="This page recognises wrinkled documents and processes them into flat ones, you can upload your local images and download the processed files.",
+                     allow_flagging="never"
+                     )
+iface.launch(share = True)
+

flagged/log.csv

@@ -0,0 +1,6 @@
+image,output,flag,username,timestamp
+,,,,2023-07-12 18:59:12.763581
+/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/image/c60b52ea5b844eb28fc764fc77bee9bd47ce39a9/tmp_dno9712.jpg,/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/output/68a659c360681de00197f50432d89fbd98efdefc/tmp_cajstfi.png,,,2023-07-13 18:34:59.129831
+/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/image/fc61756a7b0971534c5b49be71caf17d819f35ad/tmpv0t201p6.jpg,/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/output/020e114a87a9ea1d41251128026eaf06751a1803/tmpa50oe7qi.png,,,2023-08-23 22:08:00.276130
+/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/image/5ad4b2887a6e21ed21ad368cb12b06241a8c06b4/tmpir4uhzvh.jpg,/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flagged/output/870c2f8157465905417c545291ec216d6c9e9d1e/tmpgkg2h7fc.png,,,2023-08-23 22:08:23.861292
+,,,,2023-08-23 22:09:54.461844

flat/2023-07-11/2023-07-11 15:21:45 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./test1.jpg', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)

flat/2023-07-11/2023-07-11 15:27:11 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 9.533

flat/2023-07-11/2023-07-11 17:10:10 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 8.614

flat/2023-07-12/2023-07-12 14:53:08 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 8.535

flat/2023-07-12/2023-07-12 14:59:01 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 8.410

flat/2023-07-12/2023-07-12 19:01:14 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 9.356

flat/2023-07-13/2023-07-13 17:09:54 @2021-02-03_Document-Dewarping-with-Control-Points.log

@@ -0,0 +1,2 @@
+Namespace(arch='Document-Dewarping-with-Control-Points', img_shrink=None, n_epoch=300, optimizer='adam', l_rate=0.0002, print_freq=60, data_path_train=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/color'), data_path_validate=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/dataset/fiducial1024/fiducial1024_v1/validate'), data_path_test='./testdata', output_path=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/flat'), resume=PosixPath('/Users/lixiangfang/Desktop/Surf code/Document-Dewarping-with-Control-Points/Source/ICDAR2021/2021-02-03 16:15:55/143/2021-02-03 16_15_55flat_img_by_fiducial_points-fiducial1024_v1.pkl'), batch_size=1, schema='test', parallel=None)
+test time : 9.125