Julio Antonio Soto
Julio Antonio Soto
Yup, still down
Hi, By looking at the source code, it looks like the user/item features can take any real number as their value, and it is multiplied by the estimated embedding vector...
Ok some updates on this: I realized that you can include `sample weight`s on the interactions data (thus, effectively on the sampled triplets as the weights for the zero-valued negative...
Hi @gpleiss, thank you for your answer. However, it is a little bit confusing to me whether to include the noise or not **when training** the model (this is, where...
> > However, it is a little bit confusing to me whether to include the noise or not when training the model (this is, where to do loss = -mll(model(...))...
@n-riesco TBH I'm not that familiar with how Falcon works itself (I do not have much spare time neither...) I will try to poke around a little bit
Hi, unfortunately I did not have any time at all :( Also, I found out that my js is extremely rusty 😅 So yeah, @kndungu can go ahead anytime :)
Hi, This may happen if you don't "reload" the notebook after installing the library. Moreover, the recommended way to install Python libraries in Databricks is by attaching them rather than...
I see. What Spark version are you using? Did you compile Spark with Hive support? That function comes from Hive's builtin UDFs...
I see, great point! Well, my guess is that the best way to deal with it is that your dates "make sense for the time being", literally. But feel free...