import tensorflow as tf
import logging
from malaya.text.ngram import ngrams as generate_ngrams
from malaya.supervised import t5 as t5_load
from malaya.supervised import gpt2 as gpt2_load
from malaya.model.t5 import Generator
from herpetologist import check_type
from typing import List, Tuple
_accepted_pos = [
'ADJ',
'ADP',
'ADV',
'ADX',
'CCONJ',
'DET',
'NOUN',
'NUM',
'PART',
'PRON',
'PROPN',
'SCONJ',
'SYM',
'VERB',
'X',
]
_accepted_entities = [
'OTHER',
'law',
'location',
'organization',
'person',
'quantity',
'time',
'event',
]
_isi_penting_availability = {
't5': {'Size (MB)': 1250, 'Quantized Size (MB)': 481, 'Maximum Length': 1024},
'small-t5': {'Size (MB)': 355.6, 'Quantized Size (MB)': 195, 'Maximum Length': 1024},
}
_gpt2_availability = {
'117M': {'Size (MB)': 499, 'Quantized Size (MB)': 126, 'Perplexity': 6.232461},
'345M': {'Size (MB)': 1420, 'Quantized Size (MB)': 357, 'Perplexity': 6.1040115},
}
[docs]@check_type
def ngrams(
sequence,
n: int,
pad_left=False,
pad_right=False,
left_pad_symbol=None,
right_pad_symbol=None,
):
"""
generate ngrams.
Parameters
----------
sequence: List[str]
list of tokenize words.
n: int
ngram size
Returns
-------
result: List[Tuple[str, str]]
"""
return generate_ngrams(
sequence=sequence,
n=n,
pad_left=pad_left,
pad_right=pad_right,
left_pad_symbol=left_pad_symbol,
right_pad_symbol=right_pad_symbol,
)
[docs]@check_type
def pos_entities_ngram(
result_pos: List[Tuple[str, str]],
result_entities: List[Tuple[str, str]],
ngram: Tuple[int, int] = (1, 3),
accept_pos: List[str] = ['NOUN', 'PROPN', 'VERB'],
accept_entities: List[str] = [
'law',
'location',
'organization',
'person',
'time',
],
):
"""
generate ngrams.
Parameters
----------
result_pos: List[Tuple[str, str]]
result from POS recognition.
result_entities: List[Tuple[str, str]]
result of Entities recognition.
ngram: Tuple[int, int]
ngram sizes.
accept_pos: List[str]
accepted POS elements.
accept_entities: List[str]
accept entities elements.
Returns
-------
result: list
"""
if not all([i in _accepted_pos for i in accept_pos]):
raise ValueError(
'accept_pos must be a subset or equal of supported POS, please run malaya.describe_pos() to get supported POS'
)
if not all([i in _accepted_entities for i in accept_entities]):
raise ValueError(
'accept_entites must be a subset or equal of supported entities, please run malaya.describe_entities() to get supported entities'
)
words = []
sentences = []
for no in range(len(result_pos)):
if (
result_pos[no][1] in accept_pos
or result_entities[no][1] in accept_entities
):
words.append(result_pos[no][0])
for gram in range(ngram[0], ngram[1] + 1, 1):
gram_words = list(ngrams(words, gram))
for sentence in gram_words:
sentences.append(' '.join(sentence))
return list(set(sentences))
[docs]@check_type
def sentence_ngram(sentence: str, ngram: Tuple[int, int] = (1, 3)):
"""
generate ngram for a text
Parameters
----------
sentence: str
ngram: tuple
ngram sizes.
Returns
-------
result: list
"""
words = sentence.split()
sentences = []
for gram in range(ngram[0], ngram[1] + 1, 1):
gram_words = list(ngrams(words, gram))
for sentence in gram_words:
sentences.append(' '.join(sentence))
return list(set(sentences))
[docs]@check_type
def babble(
string: str,
model,
generate_length: int = 30,
leed_out_len: int = 1,
temperature: float = 1.0,
top_k: int = 100,
burnin: int = 15,
batch_size: int = 5,
):
"""
Use pretrained transformer models to generate a string given a prefix string.
https://github.com/nyu-dl/bert-gen, https://arxiv.org/abs/1902.04094
Parameters
----------
string: str
model: object
transformer interface object. Right now only supported BERT, ALBERT and ELECTRA.
generate_length: int, optional (default=256)
length of sentence to generate.
leed_out_len: int, optional (default=1)
length of extra masks for each iteration.
temperature: float, optional (default=1.0)
logits * temperature.
top_k: int, optional (default=100)
k for top-k sampling.
burnin: int, optional (default=15)
for the first burnin steps, sample from the entire next word distribution, instead of top_k.
batch_size: int, optional (default=5)
generate sentences size of batch_size.
Returns
-------
result: List[str]
"""
if not hasattr(model, 'samples'):
raise ValueError('model must have `samples` attribute')
if generate_length < 10:
raise ValueError('generate_length must bigger than 10')
if not 0 < temperature <= 1.0:
raise ValueError('temperature must, 0 < temperature <= 1.0')
if not top_k > 0:
raise ValueError('top_k must be bigger than 0')
if not burnin > 0:
raise ValueError('burnin must be bigger than 0')
if leed_out_len >= generate_length:
raise ValueError('leed_out_len must be smaller than generate_length')
if burnin >= generate_length:
raise ValueError('burnin must be smaller than generate_length')
from malaya.transformers.babble import sequential_generation
if tf.executing_eagerly():
logging.warning(
'malaya.generator.babble will disable eager execution.'
)
tf.compat.v1.disable_eager_execution()
return sequential_generation(
string,
model,
batch_size=batch_size,
max_len=generate_length,
leed_out_len=leed_out_len,
temperature=temperature,
top_k=top_k,
burnin=burnin,
)
[docs]def available_gpt2():
"""
List available gpt2 generator models.
"""
from malaya.function import describe_availability
return describe_availability(_gpt2_availability,
text='calculate perplexity on never seen malay karangan.')
[docs]@check_type
def gpt2(model: str = '345M', quantized: bool = False, **kwargs):
"""
Load GPT2 model to generate a string given a prefix string.
Parameters
----------
model: str, optional (default='345M')
Check available models at `malaya.generator.available_gpt2()`.
quantized: bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result: malaya.model.tf.GPT2 class
"""
model = model.upper()
if model not in _gpt2_availability:
raise ValueError(
'model not supported, please check supported models from `malaya.generator.available_gpt2()`.'
)
return gpt2_load.load(
model=model,
quantized=quantized,
**kwargs,
)
[docs]def available_isi_penting():
"""
List available transformer models for isi penting generator.
"""
from malaya.function import describe_availability
return describe_availability(_isi_penting_availability)
[docs]@check_type
def isi_penting(model: str = 't5', quantized: bool = False, **kwargs):
"""
Load Transformer model to generate a string given a isu penting.
Parameters
----------
model: str, optional (default='base')
Check available models at `malaya.generator.available_isi_penting()`.
quantized: bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result: malaya.model.t5.Generator class
"""
model = model.lower()
if model not in _isi_penting_availability:
raise ValueError(
'model not supported, please check supported models from `malaya.generator.available_isi_penting()`.'
)
return t5_load.load(
module='generator',
model=model,
model_class=Generator,
quantized=quantized,
**kwargs,
)
