FangShancheng/ABINet
Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition
Jupyter NotebookNOASSERTION
Issues
- 1
- 0
ABINet self-traing部分代码
#111 opened by kekexu123 - 6
callbacks.py中断言错误:AssertionError: tensor([0, 0], device='cuda:0') != tensor([26, 26], device='cuda:0')
#102 opened by anthonyAndchen - 2
- 7
- 0
- 0
- 5
请问训练中断之后怎么加载已经保存的模型继续训练呀?没找到修改的参数
#105 opened by wu-yz - 2
Reproduce training issue
#107 opened by htuannn - 1
- 0
- 7
想问一下有没有中文模型相关的csv
#83 opened by 1193700079 - 0
RecursionError
#104 opened by yywwwwww - 0
- 3
- 3
可视化结果
#63 opened by tangbinchn - 2
Error(s) in loading state_dict for BCNLanguage,Missing key(s) in state_dict
#58 opened by VictorYang097 - 0
assert os.path.isfile(file)
#100 opened by leduy-it - 1
I can't start training the model and get an error. AttributeError: 'NoneType' object has no attribute 'update'
#93 opened by kos94ok - 0
- 0
论文细节求解释下
#97 opened by menglin0320 - 1
Evaluation dataset contain corrupted image
#94 opened by Yosu26 - 0
Hello, may I ask how this visualized attention map is generated? Can you provide the code please?
#95 opened by 1193700079 - 2
- 9
- 1
语言预训练模型单独验证
#90 opened by kai1980 - 0
- 0
Unable to perform case sensitive recognition
#89 opened by shivpoojansaini - 0
想问这个MJ和ST这两个数据集对应的txt的格式是什么?
#88 opened by Johnny-xx - 1
ABINet has language bias?
#85 opened by light42 - 0
question about evaluation
#87 opened by vanche - 3
关于demo和evaluation预处理问题
#69 opened by Topdu - 1
Dataparallel for training on mutliple GPU
#86 opened by Preethse - 7
- 0
奇怪的热图
#81 opened by icecream-Tnak - 1
Looking forward to ABINet++
#84 opened by Yuliang-Liu - 1
testing the model on multiple images.
#82 opened by Preethse - 1
- 1
训练期间fastai debug的问题
#74 opened by 1193700079 - 4
Thank you ! it works amazing
#77 opened by sneedgers - 1
Less VRAM U_U 14GBvRAM in USE
#76 opened by sneedgers - 2
Fine-Tuning for captcha?
#73 opened by sneedgers - 0
.mdb文件相关问题
#72 opened by trra1988 - 0
About inference time
#71 opened by Strand2013 - 1
Detection of upper case letters and punctuation
#70 opened by sjscotti - 1
- 2
- 1
CTCLoss is nan
#65 opened by dagongji10 - 0
- 1
About the method of rectify the image first, and then perform scene text recognition
#60 opened by HHeracles