"RuntimeRrror: sizes must be non-negative" when using my own data

[zhongxiangboy]

The code works fine while using the data you provided , but when I use my own data for training, "RuntimeRrror: sizes must be non-negative".

My data generation follows the instruction from https://github.com/abisee/cnn-dailymail, which is also adopted by pointer-sum (https://github.com/atulkum/pointer_summarizer)

So, if it is convenient, can you provide the code to generate the training data?

[darsh10]

Hi, You need two inputs for every data point.

…

On Thu, Jan 27, 2022 at 6:34 AM 李鹏程 ***@***.***> wrote: The code works fine while using the data you provided , but when I use my own data for training, "RuntimeRrror: sizes must be non-negative". My data generation follows the instruction from https://github.com/abisee/cnn-dailymail, which is also adopted by pointer-sum (https://github.com/atulkum/pointer_summarizer) So, if it is convenient, can you provide the code to generate the training data? — Reply to this email directly, view it on GitHub <#2>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACXM572T4ZD55ICE5Z4QMOLUYFJXFANCNFSM5M6CCMSQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[zhongxiangboy]

I upgraded the torch version from 0.4.1 to 1.0.0 and it works :)