Skip to content

使用阿里的ocr服务,是否需要在配置里面加入OCR_ACCESS_KEY_ID和OCR_ACCESS_KEY_SECRET #263

@sunnyingit

Description

@sunnyingit

执行:python run.py --instruction="Open Chrome and search the PC-Agent paper." --mac 1
报错:缺失了OCR_ACCESS_KEY_ID配置

之后,我在config.json中增加了OCR_ACCESS_KEY_ID, OCR_ACCESS_KEY_SECRET。开通了OCR服务,并通过阿里云平台验证这两个参数可以解析图片。

但运行python run.py --instruction="Open Chrome and search the PC-Agent paper." --mac 1,依然连接不上,是还有什么配置问题吗?

task_1/screenshot_1.png
Traceback (most recent call last):
File "/Users/lisun/workplace/MobileAgent/PC-Agent/PCAgent/text_localization.py", line 62, in ocr
outputs = Sample.main(image)
File "/Users/lisun/workplace/MobileAgent/PC-Agent/PCAgent/text_localization.py", line 33, in main
output = client.recognize_all_text_with_options(recognize_all_text_request, runtime)
File "/Users/lisun/workplace/MobileAgent/PC-Agent/.venv/lib/python3.10/site-packages/alibabacloud_ocr_api20210707/client.py", line 354, in recognize_all_text_with_options
self.call_api(params, req, runtime)
File "/Users/lisun/workplace/MobileAgent/PC-Agent/.venv/lib/python3.10/site-packages/alibabacloud_tea_openapi/client.py", line 2015, in call_api
return self.do_request(params, request, runtime)
File "/Users/lisun/workplace/MobileAgent/PC-Agent/.venv/lib/python3.10/site-packages/alibabacloud_tea_openapi/client.py", line 1492, in do_request
raise UnretryableException(_last_request, _last_exception)
Tea.exceptions.UnretryableException: ('Connection aborted.', TimeoutError('The write operation timed out'))

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/Users/lisun/workplace/MobileAgent/PC-Agent/run.py", line 647, in
perception_infos, width, height = get_perception_infos(screenshot_file, screenshot_som_file, font_path=args.font_path)
File "/Users/lisun/workplace/MobileAgent/PC-Agent/run.py", line 335, in get_perception_infos
sub_text, sub_coordinates = ocr(img) # for api
File "/Users/lisun/workplace/MobileAgent/PC-Agent/PCAgent/text_localization.py", line 64, in ocr
raise OCRError(e.message)
PCAgent.text_localization.OCRError: RetryError("('Connection aborted.', TimeoutError('The write operation timed out'))")

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions