Help!

[LifeOfCodeDesigner]

I'm so sorry to say that the code is not able to process directly ,there exist some mistake. Can you ensure the code can be runned in you computer?

[anishathalye]

I'm not sure I understand your question. Do you see some error message when you try running the code? What's the error?

[LifeOfCodeDesigner]

I'm very sorry to say that the code you offer can't run directly. Firstly,I notice some small mistake, for example，in the document of main.py, there is a code of "BATCH_SIZE = 50", but in the following code I find you use the code of "raise ValueError('batch size must be divisible by 2 * number of GPUs (batch_size=%d, gpus=%d)' % ( batch_size, len(gpus) ))" the name of variable you define and use is not the same. Secondly, in the document of logging_utils.py , I find three variable are used without definition . They are the variable named orig_class, target_class and label_to_name. Finally, in the document of inception_v3_imagenet.py, the code of "if image.dtype != tf.float32: image = tf.image.convert_image_dtype(image, dtype=tf.float32)" can't work successful. But when I delete the first code and put the second code in the last of the function ,it can work very well. Although I solve many small mistake of the code ,I can't run it successfully. My recent work need the code very much. I sincerely hope you can give me some help .

…

------------------ 原始邮件 ------------------ 发件人: "Anish Athalye"<[email protected]>; 发送时间: 2019年10月15日(星期二) 晚上9:16 收件人: "labsix/limited-blackbox-attacks"<[email protected]>; 抄送: "头很大~"<[email protected]>;"Author"<[email protected]>; 主题: Re: [labsix/limited-blackbox-attacks] Help! (#12) I'm not sure I understand your question. Do you see some error message when you try running the code? What's the error? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[LifeOfCodeDesigner]

I'm very sorry to say that the code you offer can't run directly. Firstly,I notice some small mistake, for example，in the document of main.py, there is a code of "BATCH_SIZE = 50", but in the following code I find you use the code of "raise ValueError('batch size must be divisible by 2 * number of GPUs (batch_size=%d, gpus=%d)' % ( batch_size, len(gpus) ))" the name of variable you define and use is not the same. Secondly, in the document of logging_utils.py , I find three variable are used without definition . They are the variable named orig_class, target_class and label_to_name. Finally, in the document of inception_v3_imagenet.py, the code of "if image.dtype != tf.float32: image = tf.image.convert_image_dtype(image, dtype=tf.float32)" can't work successful. But when I delete the first code and put the second code in the last of the function ,it can work very well. Although I solve many small mistake of the code ,I can't run it successfully. My recent work need the code very much. I sincerely hope you can give me some help .

…

------------------ 原始邮件 ------------------ 发件人: "Anish Athalye"<[email protected]>; 发送时间: 2019年10月15日(星期二) 晚上9:16 收件人: "labsix/limited-blackbox-attacks"<[email protected]>; 抄送: "头很大~"<[email protected]>;"Author"<[email protected]>; 主题: Re: [labsix/limited-blackbox-attacks] Help! (#12) I'm not sure I understand your question. Do you see some error message when you try running the code? What's the error? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.