tensorneko 0.3.19__tar.gz → 0.3.21__tar.gz

{tensorneko-0.3.19 → tensorneko-0.3.21}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: tensorneko
-Version: 0.3.19
+Version: 0.3.21
 Summary: Tensor Neural Engine Kompanion. An util library based on PyTorch and PyTorch Lightning.
 Home-page: https://github.com/ControlNet/tensorneko
 Author: ControlNet
@@ -22,7 +22,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         <div align="center">    
             <a href="https://www.python.org/"><img src="https://img.shields.io/pypi/pyversions/tensorneko?style=flat-square"></a>
             <a href="https://pytorch.org/"><img src="https://img.shields.io/badge/PyTorch-%3E%3D1.9.0-EE4C2C?style=flat-square&logo=pytorch"></a>
-            <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.0.*%20|%202.1.*%20|%202.2.*-792EE5?style=flat-square&logo=lightning"></a>
+            <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.*-792EE5?style=flat-square&logo=lightning"></a>
         </div>
         
         <div align="center">
@@ -59,7 +59,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         
         Then you can use the CLI tools `tensorneko` in the terminal.
         
-        ## Neko Layers, Modules and Architectures
+        ## Layers, Modules and Architectures
         
         Build an MLP with linear layers. The activation and normalization will be placed in the hidden layers.
         
@@ -148,7 +148,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         # torch.Size([1])
         ```
         
-        ## Neko IO
+        ## IO
         
         Easily load and save different modal data.
         
@@ -198,7 +198,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         Besides, the read/write for `mat` and `pickle` files is also supported.
         
         
-        ## Neko preprocessing
+        ## Preprocessing
         
         ```python
         import tensorneko as neko
@@ -225,7 +225,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         - `resample_video_fps`
         - `mp32wav`
         
-        ## Neko Visualization
+        ## Visualization
         
         ### Variable Web Watcher
         Start a web server to watch the variable status when the program (e.g. training, inference, data preprocessing) is running.
@@ -376,14 +376,14 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         trainer.fit(model, dm)
         ```
         
-        ## Neko Callbacks
+        ## Callbacks
         
         Some simple but useful pytorch-lightning callbacks are provided.
         
         - `DisplayMetricsCallback`
         - `EarlyStoppingLR`: Early stop training when learning rate reaches threshold.
         
-        ## Neko Notebook Helpers
+        ## Notebook Helpers
         Here are some helper functions to better interact with Jupyter Notebook.
         ```python
         import tensorneko as neko
@@ -395,7 +395,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         neko.notebook.display.code("path/to/code.java")
         ```
         
-        ## Neko Debug Tools
+        ## Debug Tools
         
         Get the default values from `ArgumentParser` args. It's convenient to use this in the notebook.
         ```python
@@ -411,7 +411,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         print(args.accumulate)  # <function sum at ...>
         ```
         
-        ## Neko Evaluation
+        ## Evaluation
         
         Some metrics function for evaluation are provided.
         
@@ -422,8 +422,36 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         - `ssim_video`
         - `ssim_image`
         
+        ## Message (Access to other services)
         
-        ## Neko Utilities
+        ### Gotify
+        
+        Send a message to the Gotify server.
+        
+        The title, URL and APP_TOKEN is the environment variable `GOTIFY_TITLE`, `GOTIFY_URL` and `GOTIFY_TOKEN`, or overwritten 
+        in the function arguments.
+        
+        ```python
+        from tensorneko.msg import gotify
+        gotify.push("This is a test message", "<URL>", "<APP_TOKEN>")
+        # then the message will be sent to the Gotify server.
+        # title = "<HOST_NAME>", message = "This is a test message", priority = 0
+        ```
+        
+        ### Postgres
+        
+        Require the `psycopg` package. Provide one single function to execute one SQL query with a temp connection.
+        
+        The database URL is the environment variable `DB_URL`, or overwritten in the function arguments.
+        ```python
+        from tensorneko.msg import postgres
+        result = postgres.execute("<SQL>", "<DB_URL>")
+        # also async version is provided
+        result = await postgres.execute_async("<SQL>", "<DB_URL>")
+        ```
+        
+        
+        ## Utilities
         
         ### Misc functions
         

{tensorneko-0.3.19 → tensorneko-0.3.21}/README.md

@@ -12,7 +12,7 @@
 <div align="center">    
     <a href="https://www.python.org/"><img src="https://img.shields.io/pypi/pyversions/tensorneko?style=flat-square"></a>
     <a href="https://pytorch.org/"><img src="https://img.shields.io/badge/PyTorch-%3E%3D1.9.0-EE4C2C?style=flat-square&logo=pytorch"></a>
-    <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.0.*%20|%202.1.*%20|%202.2.*-792EE5?style=flat-square&logo=lightning"></a>
+    <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.*-792EE5?style=flat-square&logo=lightning"></a>
 </div>
 
 <div align="center">
@@ -49,7 +49,7 @@ pipx install tensorneko_tool  # or `pip install tensorneko_tool`
 
 Then you can use the CLI tools `tensorneko` in the terminal.
 
-## Neko Layers, Modules and Architectures
+## Layers, Modules and Architectures
 
 Build an MLP with linear layers. The activation and normalization will be placed in the hidden layers.
 
@@ -138,7 +138,7 @@ print(f(torch.rand(16)).shape)
 # torch.Size([1])
 ```
 
-## Neko IO
+## IO
 
 Easily load and save different modal data.
 
@@ -188,7 +188,7 @@ neko.io.write.json("path/to/json.json", json_obj)
 Besides, the read/write for `mat` and `pickle` files is also supported.
 
 
-## Neko preprocessing
+## Preprocessing
 
 ```python
 import tensorneko as neko
@@ -215,7 +215,7 @@ if `ffmpeg` is available, you can use below ffmpeg wrappers.
 - `resample_video_fps`
 - `mp32wav`
 
-## Neko Visualization
+## Visualization
 
 ### Variable Web Watcher
 Start a web server to watch the variable status when the program (e.g. training, inference, data preprocessing) is running.
@@ -366,14 +366,14 @@ trainer = neko.NekoTrainer(log_every_n_steps=100, gpus=1, logger=model.name, pre
 trainer.fit(model, dm)
 ```
 
-## Neko Callbacks
+## Callbacks
 
 Some simple but useful pytorch-lightning callbacks are provided.
 
 - `DisplayMetricsCallback`
 - `EarlyStoppingLR`: Early stop training when learning rate reaches threshold.
 
-## Neko Notebook Helpers
+## Notebook Helpers
 Here are some helper functions to better interact with Jupyter Notebook.
 ```python
 import tensorneko as neko
@@ -385,7 +385,7 @@ neko.notebook.display.audio("path/to/audio.wav")
 neko.notebook.display.code("path/to/code.java")
 ```
 
-## Neko Debug Tools
+## Debug Tools
 
 Get the default values from `ArgumentParser` args. It's convenient to use this in the notebook.
 ```python
@@ -401,7 +401,7 @@ print(args.integers)  # [1, 2, 3]
 print(args.accumulate)  # <function sum at ...>
 ```
 
-## Neko Evaluation
+## Evaluation
 
 Some metrics function for evaluation are provided.
 
@@ -412,8 +412,36 @@ Some metrics function for evaluation are provided.
 - `ssim_video`
 - `ssim_image`
 
+## Message (Access to other services)
 
-## Neko Utilities
+### Gotify
+
+Send a message to the Gotify server.
+
+The title, URL and APP_TOKEN is the environment variable `GOTIFY_TITLE`, `GOTIFY_URL` and `GOTIFY_TOKEN`, or overwritten 
+in the function arguments.
+
+```python
+from tensorneko.msg import gotify
+gotify.push("This is a test message", "<URL>", "<APP_TOKEN>")
+# then the message will be sent to the Gotify server.
+# title = "<HOST_NAME>", message = "This is a test message", priority = 0
+```
+
+### Postgres
+
+Require the `psycopg` package. Provide one single function to execute one SQL query with a temp connection.
+
+The database URL is the environment variable `DB_URL`, or overwritten in the function arguments.
+```python
+from tensorneko.msg import postgres
+result = postgres.execute("<SQL>", "<DB_URL>")
+# also async version is provided
+result = await postgres.execute_async("<SQL>", "<DB_URL>")
+```
+
+
+## Utilities
 
 ### Misc functions
 

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/dataset/list_dataset.py

@@ -1,6 +1,13 @@
 from typing import List
 
-from torch.utils.data.dataset import Dataset, T_co
+from torch.utils.data.dataset import Dataset
+
+try:
+    # for pytorch < 2.5
+    from torch.utils.data.dataset import T_co
+except ImportError:
+    # For pytorch >= 2.5
+    from torch.utils.data.dataset import _T_co as T_co
 
 
 class ListDataset(Dataset[T_co]):

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/dataset/nested_dataset.py

@@ -2,7 +2,14 @@ from abc import abstractmethod, ABC
 from typing import Tuple
 
 import numpy as np
-from torch.utils.data.dataset import Dataset, T_co
+from torch.utils.data.dataset import Dataset
+
+try:
+    # for pytorch < 2.5
+    from torch.utils.data.dataset import T_co
+except ImportError:
+    # For pytorch >= 2.5
+    from torch.utils.data.dataset import _T_co as T_co
 
 
 class NestedDataset(Dataset[T_co], ABC):

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/dataset/round_robin_dataset.py

@@ -2,7 +2,13 @@ import random
 from typing import List, Optional
 
 from torch.utils.data import Dataset
-from torch.utils.data.dataset import T_co
+
+try:
+    # for pytorch < 2.5
+    from torch.utils.data.dataset import T_co
+except ImportError:
+    # For pytorch >= 2.5
+    from torch.utils.data.dataset import _T_co as T_co
 
 from ..util import circular_pad
 

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/dataset/sampler/sequential_iter_sampler.py

@@ -1,6 +1,13 @@
 from typing import Sized
 
-from torch.utils.data.sampler import Sampler, T_co
+from torch.utils.data.sampler import Sampler
+
+try:
+    # for pytorch < 2.5
+    from torch.utils.data.sampler import T_co
+except ImportError:
+    # For pytorch >= 2.5
+    from torch.utils.data.sampler import _T_co as T_co
 
 
 class SequentialIterSampler(Sampler[T_co]):

tensorneko-0.3.21/src/tensorneko/msg/__init__.py

@@ -0,0 +1,9 @@
+from tensorneko_util.msg import gotify
+
+__all__ = ["gotify"]
+
+try:
+    from tensorneko_util.msg import postgres
+    __all__.append("postgres")
+except ImportError:
+    pass

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/neko_model.py

@@ -214,12 +214,12 @@ class NekoModel(LightningModule, NekoModule):
         for key in outputs[0].keys():
             getter = summarize_dict_by(key, torch.mean)
             value = getter(outputs)
-            self.history[-1]["val_" + key] = value
+            self.history[-1]["val_" + key.split("/")[-1]] = value
             if self.logger is not None:
                 self.logger.log_metrics({key: value}, step=self.trainer.global_step)
             self.log(key, value, on_epoch=True, on_step=False, logger=False, sync_dist=self.distributed)
-            self.log(f"val_{key}", value, on_epoch=True, on_step=False, logger=False, prog_bar=True,
-                sync_dist=self.distributed)
+            self.log(f"val_{key.split('/')[-1]}", value, on_epoch=True, on_step=False, logger=False,
+                prog_bar=True, sync_dist=self.distributed)
 
     def log_on_training_step_end(self, output: STEP_OUTPUT) -> None:
         """Log the training step outputs"""

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko/neko_trainer.py

@@ -1,4 +1,5 @@
 import os
+import warnings
 from datetime import timedelta
 from time import time
 from typing import Optional, Union, List, Dict
@@ -70,6 +71,7 @@ class NekoTrainer(Trainer):
             callbacks = []
         # build checkpoint callback or from user defined
         if enable_checkpointing and len([c for c in callbacks if isinstance(c, Checkpoint)]) == 0:
+            warnings.warn("Checkpoint callback is not defined, using default checkpoint callback.")
             # use default checkpoint callback
             new_callback = ModelCheckpoint(
                 dirpath=os.path.join("logs", self.log_name, "checkpoints"),

tensorneko-0.3.21/src/tensorneko/version.txt

@@ -0,0 +1 @@
+0.3.21

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: tensorneko
-Version: 0.3.19
+Version: 0.3.21
 Summary: Tensor Neural Engine Kompanion. An util library based on PyTorch and PyTorch Lightning.
 Home-page: https://github.com/ControlNet/tensorneko
 Author: ControlNet
@@ -22,7 +22,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         <div align="center">    
             <a href="https://www.python.org/"><img src="https://img.shields.io/pypi/pyversions/tensorneko?style=flat-square"></a>
             <a href="https://pytorch.org/"><img src="https://img.shields.io/badge/PyTorch-%3E%3D1.9.0-EE4C2C?style=flat-square&logo=pytorch"></a>
-            <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.0.*%20|%202.1.*%20|%202.2.*-792EE5?style=flat-square&logo=lightning"></a>
+            <a href="https://www.pytorchlightning.ai/"><img src="https://img.shields.io/badge/Lightning-2.*-792EE5?style=flat-square&logo=lightning"></a>
         </div>
         
         <div align="center">
@@ -59,7 +59,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         
         Then you can use the CLI tools `tensorneko` in the terminal.
         
-        ## Neko Layers, Modules and Architectures
+        ## Layers, Modules and Architectures
         
         Build an MLP with linear layers. The activation and normalization will be placed in the hidden layers.
         
@@ -148,7 +148,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         # torch.Size([1])
         ```
         
-        ## Neko IO
+        ## IO
         
         Easily load and save different modal data.
         
@@ -198,7 +198,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         Besides, the read/write for `mat` and `pickle` files is also supported.
         
         
-        ## Neko preprocessing
+        ## Preprocessing
         
         ```python
         import tensorneko as neko
@@ -225,7 +225,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         - `resample_video_fps`
         - `mp32wav`
         
-        ## Neko Visualization
+        ## Visualization
         
         ### Variable Web Watcher
         Start a web server to watch the variable status when the program (e.g. training, inference, data preprocessing) is running.
@@ -376,14 +376,14 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         trainer.fit(model, dm)
         ```
         
-        ## Neko Callbacks
+        ## Callbacks
         
         Some simple but useful pytorch-lightning callbacks are provided.
         
         - `DisplayMetricsCallback`
         - `EarlyStoppingLR`: Early stop training when learning rate reaches threshold.
         
-        ## Neko Notebook Helpers
+        ## Notebook Helpers
         Here are some helper functions to better interact with Jupyter Notebook.
         ```python
         import tensorneko as neko
@@ -395,7 +395,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         neko.notebook.display.code("path/to/code.java")
         ```
         
-        ## Neko Debug Tools
+        ## Debug Tools
         
         Get the default values from `ArgumentParser` args. It's convenient to use this in the notebook.
         ```python
@@ -411,7 +411,7 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         print(args.accumulate)  # <function sum at ...>
         ```
         
-        ## Neko Evaluation
+        ## Evaluation
         
         Some metrics function for evaluation are provided.
         
@@ -422,8 +422,36 @@ Description: <h1 style="text-align: center">TensorNeko</h1>
         - `ssim_video`
         - `ssim_image`
         
+        ## Message (Access to other services)
         
-        ## Neko Utilities
+        ### Gotify
+        
+        Send a message to the Gotify server.
+        
+        The title, URL and APP_TOKEN is the environment variable `GOTIFY_TITLE`, `GOTIFY_URL` and `GOTIFY_TOKEN`, or overwritten 
+        in the function arguments.
+        
+        ```python
+        from tensorneko.msg import gotify
+        gotify.push("This is a test message", "<URL>", "<APP_TOKEN>")
+        # then the message will be sent to the Gotify server.
+        # title = "<HOST_NAME>", message = "This is a test message", priority = 0
+        ```
+        
+        ### Postgres
+        
+        Require the `psycopg` package. Provide one single function to execute one SQL query with a temp connection.
+        
+        The database URL is the environment variable `DB_URL`, or overwritten in the function arguments.
+        ```python
+        from tensorneko.msg import postgres
+        result = postgres.execute("<SQL>", "<DB_URL>")
+        # also async version is provided
+        result = await postgres.execute_async("<SQL>", "<DB_URL>")
+        ```
+        
+        
+        ## Utilities
         
         ### Misc functions
         

{tensorneko-0.3.19 → tensorneko-0.3.21}/src/tensorneko.egg-info/requires.txt

@@ -2,7 +2,7 @@ av>=8.0.3
 einops>=0.3.0
 numpy>=1.20.1
 pillow>=8.1
-tensorneko_util==0.3.19
+tensorneko_util==0.3.21
 torch>=1.9.0
 torchaudio>=0.9.0
 torchmetrics>=0.7.3
@@ -12,5 +12,5 @@ torchvision>=0.10.0
 pysoundfile>=0.9.0
 
 [lightning]
-lightning<2.2,>=2.0
+lightning<3,>=2.0
 tensorboard>=2.0.0

tensorneko-0.3.19/src/tensorneko/msg/__init__.py

@@ -1,3 +0,0 @@
-from tensorneko_util.msg import push_gotify
-
-__all__ = ["push_gotify"]

tensorneko-0.3.19/src/tensorneko/version.txt

@@ -1 +0,0 @@
-0.3.19