This tutorial section goes over the NLP capabilities available through AdaptNLP and how to use them.
You should ideally follow the tutorials along with the provided notebooks in the tutorials directory at the top
level of the AdaptNLP library.
You could also run the code snippets in these tutorials straight through the python interpreter as well.
Install and Setup
You will first need to install AdaptNLP with Python 3.6+ using the following command:
pip install adaptnlp
AdaptNLP is largely built on top of Flair, Transformers, and PyTorch, and dependencies will be handled on install.
AdaptNLP can be used with or without GPUs. AdaptNLP will automatically make use of GPU VRAM in environment with CUAD-compatible NVIDIA GPUs and NVIDIA drivers installed. GPU-less environments will run AdaptNLP modules fine as well.
Overview of NLP Capabilities
An overview of some of the AdaptNLP capabilities via. code snippets.
A good way to make sure AdaptNLP is installed correctly is by running each of these code snippets.
Note this may take a while if models have not already been downloaded.
Named Entity Recognition with EasyTokenTagger
from adaptnlp import EasyTokenTagger
## Example Text
example_text = "Novetta's headquarters is located in Mclean, Virginia."
## Load the token tagger module and tag text with the NER model
tagger = EasyTokenTagger()
sentences = tagger.tag_text(text=example_text, model_name_or_path="ner")
## Output tagged token span results in Flair's Sentence object model
for sentence in sentences:
for entity in sentence.get_spans("ner"):
print(entity)
Output
```python Span [1]: "Novetta" [− Labels: ORG (0.9925)] Span [7]: "Mclean" [− Labels: LOC (0.9993)] Span [9]: "Virginia" [− Labels: LOC (1.0)] ```English Sentiment Classifier EasySequenceClassifier
from adaptnlp import EasySequenceClassifier
## Example Text
example_text = "Novetta is a great company that was chosen as one of top 50 great places to work!"
## Load the sequence classifier module and classify sequence of text with the english sentiment model
classifier = EasySequenceClassifier()
sentences = classifier.tag_text(text=example_text, model_name_or_path="en-sentiment")
## Output labeled text results in Flair's Sentence object model
for sentence in sentences:
print(sentence.labels)
Output
```python [POSITIVE (0.9977)] ```Language Model Embeddings EasyWordEmbeddings
from adaptnlp import EasyWordEmbeddings
## Example Text
example_text = "Albert Einstein used to work at Novetta."
## Load the embeddings module and embed the tokens within the text
embeddings = EasyWordEmbeddings()
sentences = embeddings.embed_text(example_text, model_name_or_path="gpt2")
# Iterate through tokens in the sentence to access embeddings
for sentence in sentences:
for token in sentence:
print(token.get_embedding())
Output
```python tensor([-1.8757, 0.6195, -1.3108, ..., -1.3787, -0.6885, 1.6934]) tensor([-0.0617, -2.3885, 2.2028, ..., 0.2774, 0.8424, -1.5328]) tensor([-0.0480, -0.7461, -0.5282, ..., 0.1554, 0.2542, 0.8199]) tensor([ 1.0621, -0.3834, 1.5259, ..., -0.0937, -0.0337, 1.0316]) tensor([-0.0027, -1.6549, -1.6274, ..., 0.3001, 0.0146, -0.1931]) tensor([ 0.6624, -0.9889, 0.6716, ..., -0.4907, 0.5692, 0.9456]) tensor([ 0.5633, 0.4789, -0.2232, ..., -0.1454, 0.2486, 0.5163]) ```Span-based Question Answering EasyQuestionAnswering
from adaptnlp import EasyQuestionAnswering
from pprint import pprint
## Example Query and Context
query = "What is the meaning of life?"
context = "Machine Learning is the meaning of life."
top_n = 5
## Load the QA module and run inference on results
qa = EasyQuestionAnswering()
best_answer, best_n_answers = qa.predict_qa(query=query, context=context, n_best_size=top_n)
## Output top answer as well as top 5 answers
print(best_answer)
pprint(best_n_answers)