diff --git a/ROADMAP.md b/ROADMAP.md index bc08ad3a86..9a3c6a263a 100644 --- a/ROADMAP.md +++ b/ROADMAP.md @@ -23,11 +23,12 @@ In the meanwhile, we are looking for contributions. An easy place to start is t * Native Binary high-performance format ### Featurization Improvements +We already provide text/NLP and image processing functionnalities that will be expanded * Text (*) - * Natural language text preprocessing such as tokenization, part-of-speech tagging, and sentence breaking - * Pre-trained text models that can be used for extracting of semantic or sentiment features from text + * Natural language text preprocessing such as improving tokenization features, adding part-of-speech tagging, and sentence boundary disambiguation + * Pre-trained text models (beyond current n-gram and pre-trained WordEmbedding text handling) that can further improve the extraction of semantic or sentiment features from text * Image (*) - * Image preprocessing such as loading, resizing, and normalization if images + * Image preprocessing such as loading, resizing, and normalization of images * Image featurization, including industry-standard pre-trained ImageNet neural models, such as ResNet and AlexNet ### Trained Model Management