Danilo Burbano

@xusliebana Could you please elaborate more on the introduced changes in the PR template

@maziyarpanahi @katelynrwalker it seems that is not working when defining a prefix different than `dbfs:/` I tried with `/dbfs/` and got the error `java.io.FileNotFoundException: file or folder: /dbfs/danilo_tmp/graphs/ner-dl/ not found`...

@maziyarpanahi since there is no issue in 4.1 and 4.2 using `dbfs:/`, I close the issue.

Hi @jslim89, as @maziyarpanahi pointed out, WordSegmenter is not multi-lingual. All these models assume the document contents will be of one language only. So, suppose a sentence has a mix...

Hi @sujitpal, thanks for your feedback we are taking a look at Aho-Corasick algorithm to see if we can implement it for EntityRuler

Hi @jfernandrezj LGFM I think you can create a PR for this

Hi @maziyarpanahi I haven't been able to replicate the error. I tried in Google Colab with T4 but it is working for spark-np 5.2.0. Can you take a look at...

Hi, @karllessard. Sure, [here](https://github.com/danilojsl/spike_nn/blob/main/src/tensorflow/model_subclassing_lstm.py ) is the Python script. I assume LSTM Layer in Keras does many more things under the hood. So, I will follow @Craigacp suggestion to initialize...

I was able to get reasonable values for `BlockLSTM` output. Since I require to work with *TensorFlow Java Version 0.2.0* I used `TruncatedNormal` as an initializer for the weights. Now,...

So, only with `BlockLSTM` will be enough to have a behavior similar to Keras LSTM as in [this](https://github.com/danilojsl/spike_nn/blob/main/src/tensorflow/model_subclassing_lstm.py) Python script?