Use the links in the table below to download the pre-trained models for the Apache OpenNLP.
|
Important
|
All models are zip compressed (like a jar file), they must not be uncompressed. |
| Component | Language | Compatibility | Description | README and Reports | File | Signatures |
|---|---|---|---|---|---|---|
Language Detector |
Detects 103 languages |
>= 1.8.3 |
Detects 103 languages in ISO 693-3 standard. Works well with longer texts that have at least 2 sentences or more from the same language. |
|||
Sentence |
fr |
>= 1.0.0 |
Sentence detection model for French |
|||
Sentence |
de |
>= 1.0.0 |
Sentence detection model for German |
|||
Sentence |
en |
>= 1.0.0 |
Sentence detection model for English |
|||
Sentence |
it |
>= 1.0.0 |
Sentence detection model for Italian |
|||
Sentence |
nl |
>= 1.0.0 |
Sentence detection model for Dutch |
|||
Parts of Speech |
de |
>= 1.0.0 |
Parts of speech model for German |
|||
Parts of Speech |
en |
>= 1.0.0 |
Parts of speech model for English |
|||
Parts of Speech |
fr |
>= 1.0.0 |
Parts of speech model for French |
|||
Parts of Speech |
it |
>= 1.0.0 |
Parts of speech model for Italian |
|||
Parts of Speech |
nl |
>= 1.0.0 |
Parts of speech model for Dutch |
|||
Tokens |
de |
>= 1.0.0 |
Tokenizer model for German |
|||
Tokens |
en |
>= 1.0.0 |
Tokenizer model for English |
|||
Tokens |
fr |
>= 1.0.0 |
Tokenizer model for French |
|||
Tokens |
it |
>= 1.0.0 |
Tokenizer model for Italian |
|||
Tokens |
nl |
>= 1.0.0 |
Tokenizer model for Dutch |
The md5, sha1, sha512, and asc files are signature files and can be used to verify the integrity of the downloaded distribution package.
Use the following commands to verify the integrity:
gpg --print-md MD5 fileName.zip
gpg --print-md SHA1 fileName.tar.gz
gpg --verify fileName.tar.gz.asc
It might be necessary to import the KEYS file to verify the integrity of the asc files.
That can easily be done with:
gpg --import KEYS
More information about release signing and verifying signatures can be found here.
The models on Sourceforge for 1.5.0 are found here. and are fully compatible with Apache OpenNLP 2.4.0.
The models can be used for testing or getting started. Please train your own models for all other use cases.