File size: 3,752 Bytes
e4b9a7b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 | # -*- coding: utf-8 -*-
# SPDX-FileCopyrightText: 2016-2025 PyThaiNLP Project
# SPDX-FileType: SOURCE
# SPDX-License-Identifier: Apache-2.0
import gzip
import json
from typing import List, Tuple
import numpy as np
class GzipModel:
"""
This class is a re-implementation of
“Low-Resource” Text Classification: A Parameter-Free Classification Method with Compressors
(Jiang et al., Findings 2023)
:param list training_data: list [(text_sample,label)]
:param str model_path: Path for loading model (if you saved the model)
"""
def __init__(self, training_data: List[Tuple[str, str]] = None, model_path: str = None):
if model_path is not None:
self.load(model_path)
else:
self.training_data = np.array(training_data)
self.Cx2_list = self.train()
def train(self):
Cx2_list = []
for i in range(len(self.training_data)):
Cx2_list.append(
len(gzip.compress(self.training_data[i][0].encode("utf-8")))
)
return Cx2_list
def predict(self, x1: str, k: int = 1) -> str:
"""
:param str x1: the text that we want to predict label for.
:param str k: k
:return: label
:rtype: str
:Example:
::
from pythainlp.classify import GzipModel
training_data = [
("รายละเอียดตามนี้เลยค่าา ^^", "Neutral"),
("กลัวพวกมึงหาย อดกินบาบิก้อน", "Neutral"),
("บริการแย่มากก เป็นหมอได้ไง😤", "Negative"),
("ขับรถแย่มาก", "Negative"),
("ดีนะครับ", "Positive"),
("ลองแล้วรสนี้อร่อย... ชอบๆ", "Positive"),
("ฉันรู้สึกโกรธ เวลามือถือแบตหมด", "Negative"),
("เธอภูมิใจที่ได้ทำสิ่งดี ๆ และดีใจกับเด็ก ๆ", "Positive"),
("นี่เป็นบทความหนึ่ง", "Neutral")
]
model = GzipModel(training_data)
print(model.predict("ฉันดีใจ", k=1))
# output: Positive
"""
Cx1 = len(gzip.compress(x1.encode("utf-8")))
disance_from_x1 = []
for i in range(len(self.Cx2_list)):
x2 = self.training_data[i][0]
Cx2 = self.Cx2_list[i]
x1x2 = "".join([x1, x2])
Cx1x2 = len(gzip.compress(x1x2.encode("utf-8")))
# normalized compression distance
ncd = (Cx1x2 - min(Cx1, Cx2)) / max(Cx1, Cx2)
disance_from_x1.append(ncd)
sorted_idx = np.argsort(np.array(disance_from_x1))
top_k_class = self.training_data[sorted_idx[:k], 1]
_, counts = np.unique(top_k_class, return_counts=True)
predict_class = top_k_class[counts.argmax()]
return predict_class
def save(self, path: str):
"""
:param str path: path for save model
"""
with open(path, "w") as f:
json.dump({
"training_data": self.training_data.tolist(),
"Cx2_list": self.Cx2_list
}, f, ensure_ascii=False)
def load(self, path: str):
with open(path, "r") as f:
data = json.load(f)
self.Cx2_list = data["Cx2_list"]
self.training_data = np.array(data["training_data"])
|