While admittedly say add_to is used in frameworks of the previous generation, let’s discuss about this: Do we really need it? Relay’s gradient pass just produces gradients without mutation and it looks perfectly fine, so in which case we have to rely on mutation?
Another thing that we could think about is that: is approach 1 always correct? If so, let’s simply do approach 1 instead. (assume we have turned off the pointer alias \_\_restrict__) --- [Visit Topic](https://discuss.tvm.ai/t/rfc-implement-add-to-semantic-in-tvm/3618/2) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai/email/unsubscribe/595268f8047a153fe1daf7310118bd82a432fa1f265868cf6a0a0f64b64c048d). Tianqi Chen, UW, Seattle, WA, 98105, United States http://tracking.discuss.tvm.ai/tracking/unsubscribe?msgid=jaeYk6usMy0jhAlskQRqjA2
