You are viewing a plain text version of this content. The canonical link for it is here.

Posted to discuss-archive@mxnet.apache.org by twodragon via MXNet Forum <mx...@discoursemail.com.INVALID> on 2021/07/13 21:49:43 UTC

[MXNet Forum] [Gluon] Make ndarray or symbol output 0 when dividing by 0


Hi all,

tl;dr Is there a way to get 0 when dividing by 0 in Gluon F.broadcast_div() instead of ```nan``` or ```inf```?

I'm attempting to rewrite the Tensorflow seq2seq sequence_loss function for Gluon. The source is at https://github.com/tensorflow/addons/blob/v0.13.0/tensorflow_addons/seq2seq/loss.py#L24-L169. Line 160 has the following:

```
crossent = tf.math.divide_no_nan(crossent, total_size)
```

which outputs 0 when dividing by 0. On the other hand, Gluon's F.broadcast_div() will output something like ```nan``` or ```inf```. Does anyone have any suggestions for mimicking this behavior in Gluon?

Thanks!





---
[Visit Topic](https://discuss.mxnet.apache.org/t/make-ndarray-or-symbol-output-0-when-dividing-by-0/7005/1) or reply to this email to respond.

You are receiving this because you enabled mailing list mode.

To unsubscribe from these emails, [click here](https://discuss.mxnet.apache.org/email/unsubscribe/96a4a650bab0b01358c959d17d1dc4a657bec3bc7df06b93339f1d718267d26e).