New versions of dependencies.

pyproject.toml

@@ -4,29 +4,28 @@ build-backend = "hatchling.build"
 
 [project]
 name = "lambo"
-version = "2.0.0"
+version = "2.1.0"
 authors = [
   { name="Piotr Przybyła", email="piotr.przybyla@ipipan.waw.pl" },
 ]
 description = "A multi-lingual text segmentation tool/"
 readme = "README.md"
-requires-python = ">=3.6.9"
+requires-python = ">=3.10"
 classifiers = [
     "Programming Language :: Python :: 3",
     "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
     "Operating System :: OS Independent",
 ]
 dependencies = [
-  'torch >= 1.7.1, < 2',
-  'importlib-resources >= 3.3.0, < 6',
-  'requests >= 2.22.0, < 3',
-  'tqdm >= 4.43.0, < 5',
-  'urllib3 >= 1.22, <2',
-  'numpy >= 1.21.0, < 2',
+  'torch >= 2.1, < 3',
+  'requests >= 2.31, < 3',
+  'tqdm >= 4.31, < 5',
+  'urllib3 >= 1.26.18, <2',
+  'numpy >= 1.23, < 2',
 ]
 
 [project.optional-dependencies]
-spacy = [ 'spacy >= 3.0.3, < 4', ]
+spacy = [ 'spacy >= 3.5.3, < 4', ]
 
 
 [project.urls]

src/lambo/examples/run_evaluation.py

@@ -3,7 +3,7 @@ Evaluation of available model by comparing to UD gold standard
 """
 from pathlib import Path
 
-import importlib_resources as resources
+import importlib.resources as resources
 
 from lambo.evaluation.evaluate import evaluate_segmenter
 from lambo.segmenter.lambo import Lambo

src/lambo/examples/run_pretraining.py

@@ -4,7 +4,7 @@ Script from pretraining models using OSCAR corpora
 import gzip
 from urllib.error import HTTPError
 
-import importlib_resources as resources
+import importlib.resources as resources
 from pathlib import Path
 
 import torch

src/lambo/examples/run_training.py

@@ -3,7 +3,7 @@ Script for training LAMBO models using UD data
 """
 import sys
 
-import importlib_resources as resources
+import importlib.resources as resources
 from pathlib import Path
 
 import torch

src/lambo/examples/run_training_pretrained.py

@@ -4,7 +4,7 @@ Script for training LAMBO models using UD data from pretrained
 import sys
 from pathlib import Path
 
-import importlib_resources as resources
+import importlib.resources as resources
 import torch
 
 from lambo.learning.train import train_new_and_save, train_pretrained_and_save

src/lambo/examples/run_training_splitting.py

@@ -4,7 +4,7 @@ Script for training LAMBO subword splitting models using UD data from pretrained
 import time, sys
 from pathlib import Path
 
-import importlib_resources as resources
+import importlib.resources as resources
 import torch
 
 from lambo.segmenter.lambo import Lambo

src/lambo/segmenter/lambo.py

 import re
 
-import importlib_resources as resources
+import importlib.resources as resources
 import torch
 
 from lambo.data.document import Document

src/lambo/utils/special_tokens.py

@@ -5,7 +5,7 @@ Detects special tokens, i.e. text fragments that should always be included in a
 """
 import re
 
-import importlib_resources as resources
+import importlib.resources as resources
 
 emoji_file_str = resources.read_text('lambo.resources', 'emoji.tab', encoding='utf-8', errors='strict')
 emojis = [line.split('\t')[0] for line in emoji_file_str.split('\n') if len(line) > 0]