Urdu Handwritten Text Dataset

该数据集包含了用于训练和评估OCR模型的乌尔都语手写文本样本。尽管该数据集在规模上有限，这影响了模型的性能，导致由于数据不足，错误率超过了85%。这项任务主要是光学字符识别（OCR）。

This dataset contains Urdu handwritten text samples for training and evaluating OCR models. Although the dataset is limited in scale, which compromises model performance, the error rate exceeds 85% due to insufficient training data. The primary task involved is optical character recognition (OCR).