Where can I get files named 'train_1000' and 'test_1000'？

[yeahQing]

I don`t understand why I should use split().

for dataset_root in config['train_dataset'].split(',')

[yeahQing]

How can I create a lmdb dataset for Chinese character?

[JingyeChen]

The link of the dataset is shown in http://www.nlpr.ia.ac.cn/databases/handwriting/Home.html

[yeahQing]

The link of the dataset is shown in http://www.nlpr.ia.ac.cn/databases/handwriting/Home.html

Hi, Chen, thanks for your reply. I have downloaded the dataset, but I don’t understand why a string is looped here. The code is:

def get_data_package():
    train_dataset = []
    # 'train_dataset': './data/mydata/train_1000' why loop this path?
    for dataset_root in config['train_dataset'].split(','):
        _, dataset = get_dataloader(dataset_root, shuffle=True)
        train_dataset.append(dataset)

[yeahQing]

What type of data set should I replace the path './data/mydata/train_1000'?

[JingyeChen]

The link of the dataset is shown in http://www.nlpr.ia.ac.cn/databases/handwriting/Home.html

Hi, Chen, thanks for your reply. I have downloaded the dataset, but I don’t understand why a string is looped here. The code is:

def get_data_package():
    train_dataset = []
    # 'train_dataset': './data/mydata/train_1000' why loop this path?
    for dataset_root in config['train_dataset'].split(','):
        _, dataset = get_dataloader(dataset_root, shuffle=True)
        train_dataset.append(dataset)

A loop is used to concatenate multiple datasets. For example, the dataset can be formulated in this way:

'train_dataset': './data/mydata/train_1000,./data/mydata/train_1500,./data/mydata/train_2000'

[JingyeChen]

What type of data set should I replace the path './data/mydata/train_1000'?

The format should be lmdb

[yeahQing]

Thank you very much, it has helped me a lot!

[cptbtptp125]

Hello, have you successfully converted LMDB format? I want to know how to convert, I have tried many methods without success

[yeahQing]

Hello, have you successfully converted LMDB format? I want to know how to convert, I have tried many methods without success

Hi, you can see in #57.

[cptbtptp125]

Hello, I am a little confused about the loop connection of multiple data sets, may I ask why this operation is carried out, and what is the difference between it and the direct single training? Thank you very much for your reply. I would appreciate it if you could help me.
'train_dataset': './data/mydata/train_1000,./data/mydata/train_1500,./data/mydata/train_2000'