Umberto Griffo

Hi @abdullahshafin, you can't apply this to the case where there is multi-label classification problem with sigmoid output activation unit. Only support the case with softmax.

@abdullahshafin you're absolutely right. I meant that you can apply this to the **multi-class classification** problem with **softmax** output activation unit. **multi-label classification** isn't supported yet.

@abdullahshafin thanks for open this issue. Would be absolutely great if you help me to adapt the code for the **multi-label classification**. I'll try to find a task that you...

@abdullahshafin don't worry, let me know if I can help you somehow.

> @umbertogriffo My understanding is that with alpha = 1 and gamma = 0, then the focal loss should produce identical results to cross entropy. However, when I compile with...

Hi, @isaacgerg thanks for the suggestion. Could you open a PR?

Hey @dmonkoff, You are right, this is an older implementation. The issue has been already solved [here](https://github.com/umbertogriffo/focal-loss-keras/issues/4), so as I've already written in the latest version you need to specify...

@oguriq take a look at [https://github.com/umbertogriffo/focal-loss-keras](https://github.com/umbertogriffo/focal-loss-keras)

Hi @kszilagyi I'm sorry for not having written to you earlier and thanks for the contribution. Did you try to run `PerformanceSimilarityBigger.java`? because I'm not seeing great performance improvements. As...

@kszilagyi yep! It's understandable. If you like and if you wanted to, give me a little benchmark, and I'm going to merge speedy this time.