da03 / Attention-OCR

Visual Attention based OCR

da03/Attention-OCR Issues

DVAN
Updated 4 months ago
OSS License compatibility question
Updated a year ago
training and testing as README introduced, but got a bad accuracy on provided evaluation dataset. What may cause that?
Updated 2 years ago1
module 'tensorflow.contrib.rnn.python.ops.rnn_cell' has no attribute '_linear'
Updated 2 years ago15
when i running the testing code. i am getting the below error
Updated 2 years ago7
generating attention mask based on the SavedModel output
Updated 3 years ago
printed mathematical expression recognition
Updated 3 years ago
Does it support special ascii characters ?
Updated 3 years ago2
I don't understand the meaning of the code: ord(c) - 97 + 13 if ord(c) > 96 else ord(c) - 48 + 3)
Updated 3 years ago1
Training is not starting.
Closed 4 years ago2
Which loss function is used?
Updated 4 years ago
Stop writing of events.out.tf.events file during test phase
Closed 4 years ago1
Does it support special asscii charater?
Updated 4 years ago1
using multi gpu
Updated 4 years ago1
About the running requirement
Updated 4 years ago2
Output Node
Updated 4 years ago5
How to test a model that I choose
Updated 5 years ago2
Cant find sample.txt file.
Closed 5 years ago
Testing trained images with bad accuracy.
Updated 5 years ago
What dependencies' versions to run this?
Updated 5 years ago1
toy跑不起来
Updated 5 years ago1
difference result in difference running test
Updated 5 years ago1
I want to know where can i get the paper of this code????
Updated 5 years ago2
bucket_specs
Updated 5 years ago2
fine-tune model, train my data
Updated 5 years ago1
What is the upper limit of the number of characters recognized by this model?
Updated 5 years ago1
testing the result in a single image
Updated 5 years ago1
which paper is implementd by this repository?"what you get"..or "coarse to fine"
Updated 5 years ago1
output_node_names
Updated 5 years ago1
cub segmented reduce error invalid configuration argument
Updated 5 years ago
icdar 2003 lexicon
Updated 5 years ago1
Can widechar be supported? Such as chinese and japanese characters.
Updated 5 years ago1
Problem when training on the ICDAR2013 dataset
Closed 5 years ago
100% incorrect
Closed 6 years ago2
train my data
Closed 6 years ago
How to train the model on coco-dataset :D
Updated 6 years ago
During the execution of "Test and visualize attention results" i am getting error:
Updated 6 years ago4
I switched to tensorflow 0.12.1 and Keras 1.1.1 but now i am getting this error
Updated 6 years ago6
Hi Can I use this code for English letter recognition in an image for different fonts and different font sizes. Can you plz help me. Thanks
Updated 6 years ago
ImportError: cannot import name ContextualZipFile
Closed 6 years ago1
Issue with GPU training (Windows)
Updated 6 years ago
Uppcase Letter
Updated 6 years ago1
Which paper does it be refered to?
Closed 6 years ago1
How to improve the accuracy?
Updated 6 years ago
麻烦问个问题，这个输出的字符个数是怎么对应的呢？
Closed 6 years ago
could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED
Updated 6 years ago
why do you set TARGET_EMBEDDING_SIZE=10？how do you represent TARGET_VOCAB_SIZE=39 with TARGET_EMBEDDING_SIZE=10？
Closed 6 years ago
fixed already
Updated 6 years ago
Magic line for visualising attention
Updated 6 years ago
Why are there not bias terms in the CNN model?
Updated 6 years ago3