SKILL BENCHMARK

NLP with Deep Learning Competency (Intermediate Level)

  • 16m
  • 16 questions
Explore all Skills Benchmarks
The NLP with Deep Learning Competency (Intermediate Level) benchmark measures your ability to identify the structure of neural networks, train a Deep Neural Network (DNN) model, and generate term frequency-inverse document frequency (TF-IDF) encodings for text. You will be evaluated on your ability to train models using pre-trained word vector embeddings, recognize the structure of a recurrent neural network (RNN), and train an RNN for sentiment analysis and with long short-term memory (LSTM). A learner who scores high on this benchmark demonstrates that they have good knowledge and experience in developing NLP applications using deep learning models.

Topics covered

  • clean and visualize text data
  • create word embeddings for model training
  • generate count vector representations for text using the TextVectorization layer
  • generate term frequency-inverse document frequency (TF-IDF) encodings for text
  • load and analyze text data
  • perform hyperparameter tuning using the Keras Tuner
  • provide an overview of how to use RNNs for processing text data
  • train a CNN for sentiment analysis
  • train a DNN model
  • train a DNN on GloVe embeddings
  • train a DNN on TF-IDF encodings
  • train a DNN on word embeddings
  • train a simple RNN memory cell
  • train RNNs with long short-term memory (LSTM) and bidirectional LSTM
  • view the TensorBoard callback output
  • visualize word embeddings with the projector plug-in