Type error:

[Ma01180724]

Traceback (most recent call last):
File "/home/mxt/dataset/deep-text-recognition-benchmark-master/train.py", line 283, in
train(opt)
File "/home/mxt/dataset/deep-text-recognition-benchmark-master/train.py", line 26, in train
train_dataset = Batch_Balanced_Dataset(opt)
File "/home/mxt/dataset/deep-text-recognition-benchmark-master/dataset.py", line 37, in init
_dataset = hierarchical_dataset(root=opt.train_data, opt=opt, select_data=[selected_d])
File "/home/mxt/dataset/deep-text-recognition-benchmark-master/dataset.py", line 102, in hierarchical_dataset
dataset = LmdbDataset(dirpath, opt)
File "/home/mxt/dataset/deep-text-recognition-benchmark-master/dataset.py", line 123, in init
nSamples = int(txn.get('num-samples'.encode()))
TypeError: int() argument must be a string, a bytes-like object or a number, not 'NoneType'

[hyungkwonko]

As I had the same issue, you may consider regenerating lmdb files which for me was the problem.

[Aldemaro14]

@hyungkwonko hey man, I'm having the same issue, also tried to regenerate the data, but still the same problem, any advice????

[hyungkwonko]

@Aldemaro14 in my case, the size of .lmdb file was 0. If you were in the same situation, you may reconsider checking the sanity of .lmdb files. Otherwise you may have different issue (library version, etc). Hope it helps.