Update 2024.06.17：**V1.1.0.7** Released
Major changes:
* adapted with cnstd>=1.2.4, thanks to [g1y5x3](https://github.com/g1y5x3) .
主要变更：
* 适配 cnstd>=1.2.4 ，感谢 [g1y5x3](https://github.com/g1y5x3) 。
Update 2024.06.04：**V1.1.0.6** Released
Major changes:
* Fix: The Text OCR incorrectly carried over the configuration from previous calls when it was called multiple times.
主要变更：
* 修复 bug：Text OCR 多次调用时错误沿用了之前的配置信息。
Update 2024.05.27：**V1.1.0.5** Released
Major changes:
* Fixed bugs such as that in `._parse_remaining`.
主要变更：
* 修复 `._parse_remaining` 等 bug。
Update 2024.05.20：**V1.1.0.4** Released
Major changes:
* set `table_as_image` as `True` if `self.table_ocr` is not available.
* fix typo: https://github.com/breezedeus/Pix2Text/pull/108 . Thanks to [billvsme](https://github.com/billvsme).
主要变更：
* 如果 `self.table_ocr` 不可用，将 `table_as_image` 设置为 `True`。
* 修复拼写错误：https://github.com/breezedeus/Pix2Text/pull/108 。感谢 [billvsme](https://github.com/billvsme)。
Update 2024.05.19：**V1.1.0.3** Released
Major changes:
* A new paid model, `mfr-plus`, has been added, which offers better recognition for multi-line formulas.
* When recognizing only English, CnOCR does not output Chinese.
* Bugs have been fixed.
主要变更：
* 加入新的付费模型：`mfr-plus`，对多行公式的识别效果更好。
* 在只识别英文时，CnOCR 不输出中文。
* 修复 bugs。
Update 2024.05.10：**V1.1.0.2** Released
Major changes:
* Fixed the error caused by empty lines in `merge_line_texts`.
主要变更：
* 修复 `merge_line_texts` 中空行导致的错误。
Update 2024.04.30：**V1.1.0.1** Released
Major changes:
* Fix the exception occurring when saving files on Windows.
主要变更：
* 修复 Windows 下存储文件时出现的异常。
Update 2024.04.28：**V1.1** Released
Major changes:
* Added layout analysis and table recognition models, supporting the conversion of images with complex layouts into Markdown format. See examples: [Pix2Text Online Documentation / Examples](https://pix2text.readthedocs.io/zh-cn/latest/examples_en/).
* Added support for converting entire PDF files to Markdown format. See examples: [Pix2Text Online Documentation / Examples](https://pix2text.readthedocs.io/zh-cn/latest/examples_en/).
* Enhanced the interface with more features, including adjustments to existing interface parameters.
* Launched the [Pix2Text Online Documentation](https://pix2text.readthedocs.io).
主要变更：
* 加入了版面分析和表格识别模型，支持把复杂排版的图片转换为 Markdown 格式，示例见：[Pix2Text 在线文档/Examples](https://pix2text.readthedocs.io/zh-cn/latest/examples/)。
* 支持把整个 PDF 文件转换为 Markdown 格式，示例见：[Pix2Text 在线文档/Examples](https://pix2text.readthedocs.io/zh-cn/latest/examples/)。
* 加入了更丰富的接口，已有接口的参数也有所调整。
* 上线了 [Pix2Text 在线文档](https://pix2text.readthedocs.io)。
Update 2024.03.30：**V1.0.2.3** Released
Major changes:
* Fixed the issue caused by `merge_line_texts`, see details at: https://github.com/breezedeus/Pix2Text/issues/84.
* Optimized the post-processing logic to handle some abnormal sequences.
主要变更：
* 修复 `merge_line_texts` 带来的错误，具体见：https://github.com/breezedeus/Pix2Text/issues/84 。
* 优化了后处理逻辑，处理部分不正常的序列。
Update 2024.03.18：**V1.0.2.2** Released
Major changes:
* The previously used `output_logits` argument is incompatible with transformers < 4.38.0, replaced by the `output_scores` argument. https://github.com/breezedeus/Pix2Text/issues/81
* Fixed a bug in `serve.py` that was not compatible with the new pix2text version.
主要变更：
* 之前使用的 `output_logits` 参数不兼容 transformers < 4.38.0，换为 `output_scores` 参数。 https://github.com/breezedeus/Pix2Text/issues/81
* 修复 `serve.py` 中未兼容新版接口的 bug。
Update 2024.03.15：**V1.0.2.1** Released
Major Changes:
* Fixed mishandling of LaTeX expressions during post-processing, such as replacing `\rightarrow` with `arrow`.
* Added `rec_config` parameter to `.recognize_text()` and `.recognize_formula()` methods for passing additional parameters for recognition.
主要变更：
* 修复对 LaTeX 表达式进行后处理时引入的误操作，如 `\rightarrow` 被替换为 `arrow`。
* 对 `.recognize_text()` 和 `.recognize_formula()` 加入了 `rec_config` 参数，以便传入用于识别的额外参数。
Update 2024.03.14：**V1.0.2** Released
Major Changes:
* Optimized the recognition process, improving the recognition of boundary punctuation that may have been missed before.
* Enhanced the LaTeX recognition results by restoring the formula tags to the formulas.
* Adjusted the output format of the recognition results, adding the `return_text` parameter to control whether to return only text or more detailed information. When returning more detailed information, confidence score `score` and position information `position` will also be provided. Thanks to [hiroi-sora](https://github.com/hiroi-sora) for the suggestion: https://github.com/breezedeus/Pix2Text/issues/67.
主要变更：
* 优化了识别的逻辑，以前可能漏识的边界标点现在可以比较好的识别。
* 对 Latex 识别结果进行了优化，把公式的 tag 还原到公式中。
* 调整了识别结果的输出格式，增加了参数 `return_text` 来控制结果是只返回文本还是更丰富的信息。当返回更丰富信息时，会返回置信度 `score` 以及位置信息 `position`。感谢 [hiroi-sora](https://github.com/hiroi-sora) 的建议：https://github.com/breezedeus/Pix2Text/issues/67 。