好用的歪皮恩

Your first model

In this chapter you are going to build your first text classification model using AllenNLP.

超级微皮恩安卓破解版

This chapter will outline how to train your model and run prediction on new data

Next steps

Now that you have a working model, here are some things you can try with AllenNLP!

Reading data

This chapter provides a deep dive into AllenNLP abstractions that are essential for reading data, including fields and instances, dataset readers, vocabulary, and how batching is handled in AllenNLP

Building your model

This chapter gives a deep dive into one of the most important components of AllenNLP—Model—and an in-depth guide to building your own model.

Common architectures

In this chapter we'll introduce neural architectures and AllenNLP abstractions that are commonly used for building your NLP model.

Representing text as features: Tokenizers, TextFields, and TextFieldEmbedders

A deep dive into AllenNLP's core abstraction: how exactly we represent textual inputs, both on the data side and the model side.

Writing tests for your code

(Coming soon) This chapter gives our recommendations for testing practices in NLP code, and describes the utilities that AllenNLP provides to make it easier.

Making predictions and serving demos

(Coming soon) This chapter describes how AllenNLP data and model code is set up to make it easy for you serve a demo of your model.

Using config files: FromParams and Registrable

This chapter describes AllenNLP's simple dependency injection framework.

Writing your own scripts

(Coming soon) Here we talk about how to easily use allennlp without using our built in commands, if you have more advanced needs or find it easier to reason about python code than configuration files.

Debugging your AllenNLP code

Some tips and tricks for using an IDE debugger with AllenNLP.

Coming soon!

This guide is still under active development, and you should expect regular updates to the guide. This chapter lets you know what's on our mind for what to write next, and tells you how to let us know what you'd like to see.