311 release candidate

[maziyarpanahi]

3.1.1

---

New Features

• Migrate AlbertEmbeddings to TensorFlow v2. This allows the importing of HuggingFace ALBERT models to Spark NLP
• Migrate AlbertEmbeddings to BatchAnnotate to allow better performance on accelerated hardware such as GPU
• Enable stdout/stderr in real-time for child processes sparknlp.start(). Thanks to PySpark 3.x, this is now possible with sparknlp.start(real_time_output=True) to have the outputs of Spark NLP (such as metrics during training) right in your Jupyter, Colab, and Kaggle notebooks.
• Complete examples for all annotators in Scaladoc APIs WIP: Doc Examples update #5668

---

Bug Fixes & Enhancements

• Fix YakeModel issue with empty token Fixing yake bug, was throwing exception on empty token input #5683 thanks to @shaddoxac
• Fix getAnchorDateMonth method in DateMatcher and MultiDateMatcher Bugfix/get anchor date month denormalized #5693
• Fix the broken PubTutor class in Python python pubtator fix #5702
• Fix relative dates in DateMatcher and MultiDateMatcher such as day after tomorrow or day before yesterday fix datematecher utils for relatives dates #5706
• Add isPaddedToken param to PubTutor python pubtator fix #5702
• Fix issue with logger inside session on some setup Fix logger issue in some cluster / add signatures to the sessions #5715
• Add signatures to TF session to handle inputs/outputs more dynamically in BertEmbeddings, DistilBertEmbeddings, RoBertaEmbeddings, and XlmRoBertaEmbeddings Fix logger issue in some cluster / add signatures to the sessions #5715
• Fix XlmRoBertaEmbeddings issue with init_all_tables Fix logger issue in some cluster / add signatures to the sessions #5715
• Add missing YakeModel from annotators
• Add missing random seed param to ClassifierDLApproach, MultiClassifierDLApproach, and SentimentDLApproach Missing getter/setter for random seed #5697
• Make the Java Exceptions appear before Py4J exceptions for ease of debugging in Python Make the Java error appear first #5709
• Make sure batchSize set in NerDLModel is the same internally to feed TensorFlow Use the same batchSize internally in NerDLModel #5716

---

Backward compatibility

We have migrated AlbertEmbeddings to TensorFlow v2, the earlier models prior to 3.1.1 won't work after this release. We have already updated the models and uploaded them on Models Hub. You can use pretrained() that takes care of it automatically or please make sure you download the new models manually.

• Fix typo in concepts.md #5664 thanks to @roger-yu-ds
• WIP: Doc Examples update #5668
• Feature/allow stdout sterr pipes #5677
• Fixing yake bug, was throwing exception on empty token input #5683 thanks to @shaddoxac
• Bugfix/get anchor date month denormalized #5693
• Missing getter/setter for random seed #5697
• python pubtator fix #5702
• fix datematecher utils for relatives dates #5706
• Make the Java error appear first #5709
• Migrate AlbertEmbeddings to TF v2 and BatchAnnotate #5714
• Fix logger issue in some cluster / add signatures to the sessions #5715
• Use the same batchSize internally in NerDLModel #5716