Introduction:

I will be talking about this processor at Apache Con @ Home 2020 in my “Apache Deep Learning 301” talk with Dr. Ian Brooks.

Sometimes you want your Deep Learning Easy and in Java, so let’s do that with DJL in custom Apache NiFi processors running in CDP Data Hubs, Private Cloud, and in laptop deployments.

Let’s run this flow with both processors getting data from the same news REST source.

For Sentiment, we get the probability and percentage for both Negative and Positive outputs as well as the raw classification results if you want them. These are NiFi attributes and won’t affect the ingested flow file data. You can change this in my class, rebuild and redeploy if you wish.

To install, please grab the prebuilt NARs referenced below and copy them to your NiFi nodes lib directories and restart those nodes. That is different for other NiFi runtimes like Stateless and K8.

I highly recommend you increase your NiFi RAM, by default you only have 512MB?!?!?!

#machine learning #deep learning #cloudera #apache nifi #pipelines #real-time streaming #djl