trainur 0.0.0__tar.gz

trainur-0.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Nam Kha Nguyen
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

trainur-0.0.0/PKG-INFO

@@ -0,0 +1,36 @@
+Metadata-Version: 2.4
+Name: trainur
+Version: 0.0.0
+Summary: A simple Pytorch trainer
+Author-email: Nam Kha Nguyen <namkha1032@gmail.com>
+License: MIT License
+        
+        Copyright (c) 2026 Nam Kha Nguyen
+        
+        Permission is hereby granted, free of charge, to any person obtaining a copy
+        of this software and associated documentation files (the "Software"), to deal
+        in the Software without restriction, including without limitation the rights
+        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+        copies of the Software, and to permit persons to whom the Software is
+        furnished to do so, subject to the following conditions:
+        
+        The above copyright notice and this permission notice shall be included in all
+        copies or substantial portions of the Software.
+        
+        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+        SOFTWARE.
+        
+Project-URL: Homepage, https://github.com/namkha1032/trainur
+Project-URL: Repository, https://github.com/namkha1032/trainur
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
+# trainur
+A simple Pytorch trainer

trainur-0.0.0/README.md

@@ -0,0 +1,2 @@
+# trainur
+A simple Pytorch trainer

trainur-0.0.0/pyproject.toml

@@ -0,0 +1,21 @@
+[project]
+name = "trainur" 
+version = "0.0.0"
+description = "A simple Pytorch trainer"
+readme = "README.md"
+authors = [
+    { name = "Nam Kha Nguyen", email = "namkha1032@gmail.com" }
+]
+license = { file = "LICENSE" }
+requires-python = ">=3.10"
+
+[project.urls]
+Homepage = "https://github.com/namkha1032/trainur"
+Repository = "https://github.com/namkha1032/trainur"
+
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+packages = ["trainur"]

trainur-0.0.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

trainur-0.0.0/tests/test.py

@@ -0,0 +1,21 @@
+from trainur import Trainur
+from logab import log_wrap
+from fire import Fire
+from dataclasses import dataclass
+
+class CustomTrainer(Trainur):
+    train_var:any=1
+    def train(self):
+        pass
+    
+    def inference(self):
+        pass
+
+class CustomTester(CustomTrainer):
+    test_var:any="some str"
+    def run_test(self):
+        pass
+
+if __name__ == "__main__":
+    with log_wrap():
+        Fire(CustomTester)

trainur-0.0.0/trainur/__init__.py

@@ -0,0 +1 @@
+from .trainer_utils import Trainur

trainur-0.0.0/trainur/trainer_utils.py

@@ -0,0 +1,88 @@
+import torch
+from torch.utils.data import DataLoader
+import sys
+from dataclasses import dataclass, asdict, fields
+import random
+import numpy as np
+from logab import log_init
+
+@dataclass
+class Trainur:
+    device: any ="cuda:0"
+    dtype: int = 32
+    epoch: int = 1
+    batch_size: int = 2
+    effective_batch_size: int = 32
+    accumulated_steps: int = None
+    num_workers: int = 4
+    prefetch_factor: int = 8
+
+    def __post_init__(self):
+        self.accumulated_steps = self.effective_batch_size // self.batch_size
+        attr_old = asdict(self)
+        
+        result_dict = dict()
+
+        for cls in reversed(self.__class__.__mro__):
+            if hasattr(cls, '__annotations__'):
+                result_dict[cls.__name__] = dict()
+                for attr in cls.__annotations__:
+                    if attr in attr_old:
+                        result_dict[cls.__name__][attr] = attr_old[attr]
+                        attr_old.pop(attr, None)
+
+        for cls_name, attr_list in result_dict.items():
+            print(f"{cls_name}:")
+            for attr_name, attr_value in attr_list.items():
+                print(f"\t{attr_name}: {attr_value}")
+            print(f"{'-'*50}\n")
+
+        self.dtype = torch.float32 if self.dtype == 32 else torch.bfloat16
+        self.logger = log_init()
+
+    def __init_subclass__(cls, **kwargs):
+        super().__init_subclass__(**kwargs)
+        dataclass(cls)
+        pass
+
+    def fix_seed(self, seed=42):
+        random.seed(seed)
+        np.random.seed(seed)
+        torch.manual_seed(seed)
+        torch.cuda.manual_seed(seed)
+        torch.cuda.manual_seed_all(seed)
+
+    def count_trainable(self, model):
+        total = 0
+        trainable = 0
+        for p in model.parameters():
+            total += p.numel()
+            if p.requires_grad:
+                trainable += p.numel()
+        result = f"Trainable {trainable} / {total} ({trainable/total*100:.4f})%"
+        return result
+
+    def transfer_tensor(self, tensor):
+        return {key: value.to(device=self.device, dtype=self.dtype if value.dtype == torch.float32 else value.dtype) if isinstance(value, torch.Tensor) else value for key, value in tensor.items()}
+    
+    def create_dataloader(self, dataset, drop_last=True, is_shuffle=True):
+        if 'debugpy' in sys.modules:
+            num_workers = 0
+            prefetch_factor = None
+            persistent_workers = False
+        else:
+            num_workers = self.num_workers
+            prefetch_factor = self.prefetch_factor
+            persistent_workers = True if num_workers > 0 else False
+        dataloader = DataLoader(
+            dataset, 
+            shuffle=is_shuffle, 
+            batch_size=self.batch_size,
+            num_workers=num_workers,
+            prefetch_factor=prefetch_factor,
+            persistent_workers=persistent_workers,
+            pin_memory=True,
+            drop_last=drop_last,
+        )
+        return dataloader
+    

trainur-0.0.0/trainur.egg-info/PKG-INFO

@@ -0,0 +1,36 @@
+Metadata-Version: 2.4
+Name: trainur
+Version: 0.0.0
+Summary: A simple Pytorch trainer
+Author-email: Nam Kha Nguyen <namkha1032@gmail.com>
+License: MIT License
+        
+        Copyright (c) 2026 Nam Kha Nguyen
+        
+        Permission is hereby granted, free of charge, to any person obtaining a copy
+        of this software and associated documentation files (the "Software"), to deal
+        in the Software without restriction, including without limitation the rights
+        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+        copies of the Software, and to permit persons to whom the Software is
+        furnished to do so, subject to the following conditions:
+        
+        The above copyright notice and this permission notice shall be included in all
+        copies or substantial portions of the Software.
+        
+        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+        SOFTWARE.
+        
+Project-URL: Homepage, https://github.com/namkha1032/trainur
+Project-URL: Repository, https://github.com/namkha1032/trainur
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
+# trainur
+A simple Pytorch trainer

trainur-0.0.0/trainur.egg-info/SOURCES.txt

@@ -0,0 +1,10 @@
+LICENSE
+README.md
+pyproject.toml
+tests/test.py
+trainur/__init__.py
+trainur/trainer_utils.py
+trainur.egg-info/PKG-INFO
+trainur.egg-info/SOURCES.txt
+trainur.egg-info/dependency_links.txt
+trainur.egg-info/top_level.txt

trainur-0.0.0/trainur.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

trainur-0.0.0/trainur.egg-info/top_level.txt

@@ -0,0 +1 @@
+trainur