I've been testing different images using the OCR.ipynb notebook and anytime there are numbers they are not recognized. Does the character model not recognize numbers? Or am I doing something wrong. Thanks! Awesome work on this

@Breta01 Currently the words are comes through the indexes in random format, How can we bring those words in correct order like " top-to-bottom-left-to-right", I have followed this link https://www.pyimagesearch.com/2015/04/20/sorting-contours-using-python-and-opencv/ But still I can't achieve completely, Any guidance can you please provide?

Hi breta unable to train the word_classifier_ctc file following is the error iam getting AttributeError Traceback (most recent call last) in 23 # Plotting loss 24 tmpLoss = loss.eval(fd) ---> 25 trainPlot.updateCost(tmpLoss, i_batch // LOSS_ITER) 26 27 if i_batch % TEST_ITER == 0:

AttributeError: 'TrainingPlot' object has no attribute 'updateCost'

I'm using an WSL ubuntu environment, I have install the environment.yaml exactly like you have mentioned but I have no idea why I'm getting this error.

If there are any requirement issues please let me know. Please help me fix this error.

##Please help me with this error. I am just trying to run the model and i encountered this error. I tried few things but none of them worked.

NameError Traceback (most recent call last) in 1 # Crop image and get bounding boxes ----> 2 crop = page.detection(image) 3 implt(crop) 4 boxes = words.detection(crop) 5 lines = words.sort_words(boxes)

E:\handwriting-ocr-master\handwriting-ocr-master\src\ocr\page.py in detection(image) 11 """Finding Page.""" 12 # Edge detection ---> 13 image_edges = _edges_detection(image, 200, 250) 14 15 # Close gaps between edges (double page clouse => rectangle kernel)

E:\handwriting-ocr-master\handwriting-ocr-master\src\ocr\page.py in _edges_detection(img, minVal, maxVal) 28 def _edges_detection(img, minVal, maxVal): 29 """Preprocessing (gray, thresh, filter, border) + Canny edge detection.""" ---> 30 img = cv2.cvtColor(cv2.resize(img), cv2.COLOR_BGR2GRAY) 31 32 img = cv2.bilateralFilter(img, 9, 75, 75)

NameError: name 'resize' is not defined

Hi,

Kindly help me with how to reproduce gap class models for my dataset. Its an emergency. I am stuck at this point. I understood that pred values are taken from gaps. maybe since my char class -class model is not compatible with gap-class that am not able to separate characters properly. ???

Hi Breta, while trying Recogition using CTC model, I get the next error: 'ValueError: Cannot feed value of shape (1, 64, 58, 1) for Tensor 'inputs:0', which has shape '(?, ?, 2400)'

I am using one of your pictures.

Thanks!

I see a very weird error while I try to train the CTC model when the gpu version of tensorflow is used. the cpu version does not have problem. the error generated from the line train_step.run(fd) try: for i_batch in range(TRAIN_STEPS): fd = train_iterator.next_feed(BATCH_SIZE) train_step.run(fd) <---------

the error is: NotFoundError: Resource __per_step_4/_tensor_arraysmap/TensorArray_1_85/N10tensorflow11TensorArrayE does not exist.

I could not find helpful materials to figure out what is the problem.

the total log is attached.

NotFoundError Traceback (most recent call last) /mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args) 1321 try: -> 1322 return fn(*args) 1323 except errors.OpError as e:

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run_fn(feed_dict, fetch_list, target_list, options, run_metadata) 1306 return self._call_tf_sessionrun( -> 1307 options, feed_dict, fetch_list, target_list, run_metadata) 1308

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _call_tf_sessionrun(self, options, feed_dict, fetch_list, target_list, run_metadata) 1408 self._session, options, feed_dict, fetch_list, target_list, -> 1409 run_metadata) 1410 else:

NotFoundError: Resource __per_step_4/_tensor_arraysmap/TensorArray_1_85/N10tensorflow11TensorArrayE does not exist. [[Node: gradients/map/TensorArrayStack/TensorArrayGatherV3_grad/TensorArrayGrad/TensorArrayGradV3 = TensorArrayGradV3[_class=["loc:@map/TensorArray_1"], source="gradients", _device="/job:localhost/replica:0/task:0/device:CPU:0"](map/TensorArray_1/_153, map/while/Exit_1/_155)]] [[Node: gradients/map/while/map/while/TensorArrayWrite/TensorArrayWriteV3_grad/tuple/control_dependency/_249 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_2317_...dependency", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

During handling of the above exception, another exception occurred:

NotFoundError Traceback (most recent call last) in () 6 for i_batch in range(TRAIN_STEPS): 7 fd = train_iterator.next_feed(BATCH_SIZE) ----> 8 train_step.run(fd) 9 if i_batch % LOSS_ITER == 0: 10 # Plotting loss

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py in run(self, feed_dict, session) 2375 none, the default session will be used. 2376 """ -> 2377 _run_using_default_session(self, feed_dict, self.graph, session) 2378 2379 _gradient_registry = registry.Registry("gradient")

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py in _run_using_default_session(operation, feed_dict, graph, session) 5213 "the operation's graph is different from the session's " 5214 "graph.") -> 5215 session.run(operation, feed_dict) 5216 5217

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata) 898 try: 899 result = self._run(None, fetches, feed_dict, options_ptr, --> 900 run_metadata_ptr) 901 if run_metadata: 902 proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata) 1133 if final_fetches or final_targets or (handle and feed_dict_tensor): 1134 results = self._do_run(handle, final_targets, final_fetches, -> 1135 feed_dict_tensor, options, run_metadata) 1136 else: 1137 results = []

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata) 1314 if handle is None: 1315 return self._do_call(_run_fn, feeds, fetches, targets, options, -> 1316 run_metadata) 1317 else: 1318 return self._do_call(_prun_fn, handle, feeds, fetches)

/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args) 1333 except KeyError: 1334 pass -> 1335 raise type(e)(node_def, op, message) 1336 1337 def _extend_graph(self):

NotFoundError: Resource __per_step_4/_tensor_arraysmap/TensorArray_1_85/N10tensorflow11TensorArrayE does not exist. [[Node: gradients/map/TensorArrayStack/TensorArrayGatherV3_grad/TensorArrayGrad/TensorArrayGradV3 = TensorArrayGradV3[_class=["loc:@map/TensorArray_1"], source="gradients", _device="/job:localhost/replica:0/task:0/device:CPU:0"](map/TensorArray_1/_153, map/while/Exit_1/_155)]] [[Node: gradients/map/while/map/while/TensorArrayWrite/TensorArrayWriteV3_grad/tuple/control_dependency/_249 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_2317_...dependency", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

Caused by op 'gradients/map/TensorArrayStack/TensorArrayGatherV3_grad/TensorArrayGrad/TensorArrayGradV3', defined at: File "/usr/lib/python3.5/runpy.py", line 184, in _run_module_as_main "main", mod_spec) File "/usr/lib/python3.5/runpy.py", line 85, in _run_code exec(code, run_globals) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel_launcher.py", line 16, in app.launch_new_instance() File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/traitlets/config/application.py", line 658, in launch_instance app.start() File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/kernelapp.py", line 486, in start self.io_loop.start() File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tornado/platform/asyncio.py", line 127, in start self.asyncio_loop.run_forever() File "/usr/lib/python3.5/asyncio/base_events.py", line 345, in run_forever self._run_once() File "/usr/lib/python3.5/asyncio/base_events.py", line 1312, in _run_once handle._run() File "/usr/lib/python3.5/asyncio/events.py", line 125, in _run self._callback(*self._args) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tornado/ioloop.py", line 759, in _run_callback ret = callback() File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tornado/stack_context.py", line 276, in null_wrapper return fn(*args, **kwargs) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 536, in self.io_loop.add_callback(lambda : self._handle_events(self.socket, 0)) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 450, in _handle_events self._handle_recv() File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 480, in _handle_recv self._run_callback(callback, msg) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 432, in _run_callback callback(*args, **kwargs) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tornado/stack_context.py", line 276, in null_wrapper return fn(*args, **kwargs) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 283, in dispatcher return self.dispatch_shell(stream, msg) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 233, in dispatch_shell handler(stream, idents, msg) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 399, in execute_request user_expressions, allow_stdin) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/ipkernel.py", line 208, in do_execute res = shell.run_cell(code, store_history=store_history, silent=silent) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/ipykernel/zmqshell.py", line 537, in run_cell return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2662, in run_cell raw_cell, store_history, silent, shell_futures) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2785, in _run_cell interactivity=interactivity, compiler=compiler, result=result) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2903, in run_ast_nodes if self.run_code(code, result): File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2963, in run_code exec(code_obj, self.user_global_ns, self.user_ns) File "", line 8, in train_step = optimizer.minimize(loss, name='train_step') File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/training/optimizer.py", line 414, in minimize grad_loss=grad_loss) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/training/optimizer.py", line 526, in compute_gradients colocate_gradients_with_ops=colocate_gradients_with_ops) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py", line 494, in gradients gate_gradients, aggregation_method, stop_gradients) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py", line 636, in _GradientsHelper lambda: grad_fn(op, *out_grads)) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py", line 385, in _MaybeCompile return grad_fn() # Exit early File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py", line 636, in lambda: grad_fn(op, *out_grads)) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_grad.py", line 161, in _TensorArrayGatherGrad .grad(source=grad_source, flow=flow)) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_ops.py", line 849, in grad return self._implementation.grad(source, flow=flow, name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_ops.py", line 241, in grad handle=self._handle, source=source, flow_in=flow, name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gen_data_flow_ops.py", line 6229, in tensor_array_grad_v3 name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper op_def=op_def) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 3392, in create_op op_def=op_def) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1718, in init self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

...which was originally created as op 'map/TensorArrayStack/TensorArrayGatherV3', defined at: File "/usr/lib/python3.5/runpy.py", line 184, in _run_module_as_main "main", mod_spec) [elided 23 identical lines from previous traceback] File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2963, in run_code exec(code_obj, self.user_global_ns, self.user_ns) File "", line 9, in dtype=tf.float32) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/functional_ops.py", line 424, in map_fn results_flat = [r.stack() for r in r_a] File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/functional_ops.py", line 424, in results_flat = [r.stack() for r in r_a] File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_ops.py", line 893, in stack return self._implementation.stack(name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_ops.py", line 291, in stack return self.gather(math_ops.range(0, self.size()), name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/tensor_array_ops.py", line 305, in gather element_shape=element_shape) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/ops/gen_data_flow_ops.py", line 6018, in tensor_array_gather_v3 flow_in=flow_in, dtype=dtype, element_shape=element_shape, name=name) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper op_def=op_def) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 3392, in create_op op_def=op_def) File "/mnt/tensor2/python3/HR/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1718, in init self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

NotFoundError (see above for traceback): Resource __per_step_4/_tensor_arraysmap/TensorArray_1_85/N10tensorflow11TensorArrayE does not exist. [[Node: gradients/map/TensorArrayStack/TensorArrayGatherV3_grad/TensorArrayGrad/TensorArrayGradV3 = TensorArrayGradV3[_class=["loc:@map/TensorArray_1"], source="gradients", _device="/job:localhost/replica:0/task:0/device:CPU:0"](map/TensorArray_1/_153, map/while/Exit_1/_155)]] [[Node: gradients/map/while/map/while/TensorArrayWrite/TensorArrayWriteV3_grad/tuple/control_dependency/_249 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_2317_...dependency", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

Thanks Breta,

Now the output for your images is perfectly working.

However I am getting some issue with the following image.

I tried with and without page.detection(image) but it did not work. Output is like this:

I think there is some problem with bounding boxes. The co-ordinates returned are not proper. You can try it for yourself. I can send you the image on your email if you give it.

hey, i'm building a platform for exam papers classifications using django ... and i need this to detect names of the students unfortunately, i couldn't quite understand how to use it since i don't have any experience in building ml models

i would appreciate it if you could explain the steps needed to send an image through your algorithm and get a response .. i really need this for a school project

Now TF2 is suported

• OCR example got similar result as in the previous version
• Removed old garbage code used for debug
• add tf1 compat for tf2 on tfhelpers (forgoted)

Hey, I somehow started the training but now it is taking a lot of time can you tell an approximate or another method to speed up the training?

And also can you tell me what you have put the value for LOSS_ITER it is not mentioned here so I put it the same as TRAIN_ITER i.e. 150 in word_classifier_CTC.ipynb

check the train.csv, dev.csv, and test.csv I have generated are good to use or have some error.

dataset-processed-link: https://www.kaggle.com/datasets/rahuldhanola/handwriting-dataset-processed

Now I got a lot of attribute errors and Name errors while training CTC so I tried fixing through what I could. But now while training it trains for the 0 batch and doesn't go on, like it's running but I don't see anything else than Batch 0 can anyone give me the workable training code for ctc or help me fixing this

While running below code, I am getting an error "ModuleNotFoundError: No module named 'ocr'". Please help

import sys import numpy as np import pandas as pd import cv2 import matplotlib as plt import tensorflow as tf import os from imgaug import augmenters as iaa

sys.path.append('src') from ocr.datahelpers import load_words_data, char2idx, CHAR_SIZE from ocr.dataiterator import BucketDataIterator from ocr.helpers import img_extend, resize from ocr.mlhelpers import TrainingPlot from ocr.tfhelpers import create_cell

Hey , hope you are doing well,

I am currently working handwritten text recognition extracted from .tif files which has both printed & handwritten text. I have used pytesseract for printed text . But handwritten was not getting correctly extracted. I was referring your project where we need to use word_classifier_CTC.ipynb in OCR.ipynb file where there is code given. Can you please help in suggesting what these .csv file contain as these files are not there in your repo.

train_images, train_labels = load_words_data('data/sets/train.csv', is_csv=True) dev_images, dev_labels = load_words_data('data/sets/dev.csv', is_csv=True)

AttributeError Traceback (most recent call last) ~\AppData\Local\Temp/ipykernel_8152/2155200203.py in 10 # Plotting cost 11 tmpCost = cost.eval(feed_dict={x: trainBatch, y_: labelBatch, keep_prob: 1.0}) ---> 12 trainPlot.updateCost(tmpCost, i // COST_ITER) 13 14 if i%TEST_ITER == 0:

AttributeError: 'TrainingPlot' object has no attribute 'updateCost'