rl4co 0.0.3.dev1__tar.gz → 0.0.3.dev3__tar.gz

{rl4co-0.0.3.dev1 → rl4co-0.0.3.dev3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rl4co
-Version: 0.0.3.dev1
+Version: 0.0.3.dev3
 Summary: RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark
 Author-email: Federico Berto <berto.federico2@gmail.com>, Chuanbo Hua <cbhua@kaist.ac.kr>, Junyoung Park <junyoungpark.ml@gmail.com>
 License:                                  Apache License
@@ -224,14 +224,15 @@ License-File: LICENSE
 <div align="center">
 
 # RL4CO
-    
-An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering. 
-        
+
+An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering.
+
 <a href="https://pytorch.org/get-started/locally/"><img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-ee4c2c?logo=pytorch&logoColor=white"></a>
 <a href="https://pytorchlightning.ai/"><img alt="Lightning" src="https://img.shields.io/badge/-Lightning-792ee5?logo=pytorchlightning&logoColor=white"></a>
 <a href="https://github.com/pytorch/rl"><img alt="base: TorchRL" src="https://img.shields.io/badge/base-TorchRL-red">
 <a href="https://hydra.cc/"><img alt="config: Hydra" src="https://img.shields.io/badge/config-Hydra-89b8cd"></a> [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
-![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)
+![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)[![PyPI](https://img.shields.io/pypi/v/rl4co?logo=pypi)](https://pypi.org/project/rl4co)
+[![Test](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml/badge.svg)](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml)
 <!-- ![testing](https://github.com/kaist-silab/ncobench/actions/workflows/tests.yml/badge.svg) -->
 
 </div>
@@ -247,6 +248,7 @@ RL4CO is built upon:
 
 
 ## Getting started
+<a href="https://colab.research.google.com/github/kaist-silab/rl4co/blob/main/notebooks/1-quickstart.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
 
 RL4CO is now available for installation on `pip`!
 ```bash
@@ -283,7 +285,7 @@ To get started, we recommend checking out our [quickstart notebook](notebooks/1-
 
 Train model with default configuration (AM on TSP environment):
 ```bash
-python run.py  
+python run.py
 ```
 
 
@@ -293,14 +295,14 @@ python run.py
 
 Train model with chosen experiment configuration from [configs/experiment/](configs/experiment/) (e.g. tsp/am, and environment with 42 cities)
 ```bash
-python run.py experiment=tsp/am env.num_loc=42 
+python run.py experiment=tsp/am env.num_loc=42
 ```
 </details>
 
 
 <details>
     <summary>Disable logging</summary>
-    
+
 ```bash
 python run.py experiment=test/am logger=none '~callbacks.learning_rate_monitor'
 ```
@@ -333,7 +335,7 @@ from rl4co.tasks.rl4co import RL4COLitModule
 config = DictConfig(
     {"data": {
             "train_size": 100000,
-            "val_size": 10000, 
+            "val_size": 10000,
             "batch_size": 512,
         },
     "optimizer": {"lr": 1e-4}}
@@ -368,7 +370,7 @@ pytest tests
 ```
 
 ## Contributing
-Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO! 
+Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO!
 
 ### Contributors
 <a href="https://github.com/kaist-silab/rl4co/graphs/contributors">

{rl4co-0.0.3.dev1 → rl4co-0.0.3.dev3}/README.md

@@ -1,14 +1,15 @@
 <div align="center">
 
 # RL4CO
-    
-An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering. 
-        
+
+An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering.
+
 <a href="https://pytorch.org/get-started/locally/"><img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-ee4c2c?logo=pytorch&logoColor=white"></a>
 <a href="https://pytorchlightning.ai/"><img alt="Lightning" src="https://img.shields.io/badge/-Lightning-792ee5?logo=pytorchlightning&logoColor=white"></a>
 <a href="https://github.com/pytorch/rl"><img alt="base: TorchRL" src="https://img.shields.io/badge/base-TorchRL-red">
 <a href="https://hydra.cc/"><img alt="config: Hydra" src="https://img.shields.io/badge/config-Hydra-89b8cd"></a> [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
-![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)
+![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)[![PyPI](https://img.shields.io/pypi/v/rl4co?logo=pypi)](https://pypi.org/project/rl4co)
+[![Test](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml/badge.svg)](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml)
 <!-- ![testing](https://github.com/kaist-silab/ncobench/actions/workflows/tests.yml/badge.svg) -->
 
 </div>
@@ -24,6 +25,7 @@ RL4CO is built upon:
 
 
 ## Getting started
+<a href="https://colab.research.google.com/github/kaist-silab/rl4co/blob/main/notebooks/1-quickstart.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
 
 RL4CO is now available for installation on `pip`!
 ```bash
@@ -60,7 +62,7 @@ To get started, we recommend checking out our [quickstart notebook](notebooks/1-
 
 Train model with default configuration (AM on TSP environment):
 ```bash
-python run.py  
+python run.py
 ```
 
 
@@ -70,14 +72,14 @@ python run.py
 
 Train model with chosen experiment configuration from [configs/experiment/](configs/experiment/) (e.g. tsp/am, and environment with 42 cities)
 ```bash
-python run.py experiment=tsp/am env.num_loc=42 
+python run.py experiment=tsp/am env.num_loc=42
 ```
 </details>
 
 
 <details>
     <summary>Disable logging</summary>
-    
+
 ```bash
 python run.py experiment=test/am logger=none '~callbacks.learning_rate_monitor'
 ```
@@ -110,7 +112,7 @@ from rl4co.tasks.rl4co import RL4COLitModule
 config = DictConfig(
     {"data": {
             "train_size": 100000,
-            "val_size": 10000, 
+            "val_size": 10000,
             "batch_size": 512,
         },
     "optimizer": {"lr": 1e-4}}
@@ -145,7 +147,7 @@ pytest tests
 ```
 
 ## Contributing
-Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO! 
+Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO!
 
 ### Contributors
 <a href="https://github.com/kaist-silab/rl4co/graphs/contributors">

rl4co-0.0.3.dev3/rl4co/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.0.3.dev3"

{rl4co-0.0.3.dev1 → rl4co-0.0.3.dev3}/rl4co/envs/tsp.py

@@ -173,8 +173,7 @@ class TSPEnv(RL4COEnvBase):
     @staticmethod
     def render(td, actions=None, ax=None):
         import matplotlib.pyplot as plt
-
-        from matplotlib import cm
+        import numpy as np
 
         if ax is None:
             # Create a plot of the nodes
@@ -197,51 +196,20 @@ class TSPEnv(RL4COEnvBase):
             locs = gather_by_index(locs, actions, dim=0)
 
         # Cat the first node to the end to complete the tour
+        locs = torch.cat((locs, locs[0:1]))
         x, y = locs[:, 0], locs[:, 1]
 
-        # SECTION: new
-        # plot visited nodes
-        ax.scatter(
-            x,
-            y,
-            edgecolors=cm.Set2(0),
-            facecolors="none",
-            s=100,
-            linewidths=2,
-            marker="o",
-            alpha=1,
-        )
+        # Plot the visited nodes
+        ax.scatter(x, y, color="tab:blue")
 
-        # plot actions
-        for action_idx in range(len(actions) - 1):
-            from_loc = locs[actions[action_idx]]
-            to_loc = locs[actions[action_idx + 1]]
-            ax.plot(
-                [from_loc[0], to_loc[0]],
-                [from_loc[1], to_loc[1]],
-                color=cm.tab20c(5),
-            )
-            ax.annotate(
-                "",
-                xy=(to_loc[0], to_loc[1]),
-                xytext=(from_loc[0], from_loc[1]),
-                arrowprops=dict(arrowstyle="->", color=cm.tab20c(5)),
-                annotation_clip=False,
-            )
-
-        # setup
-        ax.set_xlim(-0.05, 1.05)
-        ax.set_ylim(-0.05, 1.05)
-        ax.grid(axis="both", color="black", ls="--", alpha=0.1)
-
-        # plt.tick_params(axis='both', which='both', bottom=False, top=False, labelbottom=False, right=False, left=False, labelleft=False)
+        # Add arrows between visited nodes as a quiver plot
+        dx, dy = np.diff(x), np.diff(y)
+        ax.quiver(
+            x[:-1], y[:-1], dx, dy, scale_units="xy", angles="xy", scale=1, color="k"
+        )
 
         # Set plot title and axis labels
-        title_font = {"fontsize": 14}
-        label_font = {"fontsize": 12}
-        ax.set_title("TSP Solution", fontdict=title_font)
-        ax.set_xlabel("X Coordinate", fontdict=label_font)
-        ax.set_ylabel("Y Coordinate", fontdict=label_font)
-
-        # plt.tight_layout()
+        ax.set_title("TSP Solution")
+        ax.set_xlabel("x-coordinate")
+        ax.set_ylabel("y-coordinate")
         plt.show()

{rl4co-0.0.3.dev1 → rl4co-0.0.3.dev3}/rl4co.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rl4co
-Version: 0.0.3.dev1
+Version: 0.0.3.dev3
 Summary: RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark
 Author-email: Federico Berto <berto.federico2@gmail.com>, Chuanbo Hua <cbhua@kaist.ac.kr>, Junyoung Park <junyoungpark.ml@gmail.com>
 License:                                  Apache License
@@ -224,14 +224,15 @@ License-File: LICENSE
 <div align="center">
 
 # RL4CO
-    
-An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering. 
-        
+
+An extensive Reinforcement Learning (RL) for Combinatorial Optimization (CO) benchmark. Our goal is to provide a unified framework for RL-based CO algorithms, and to facilitate reproducible research in this field, decoupling the science from the engineering.
+
 <a href="https://pytorch.org/get-started/locally/"><img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-ee4c2c?logo=pytorch&logoColor=white"></a>
 <a href="https://pytorchlightning.ai/"><img alt="Lightning" src="https://img.shields.io/badge/-Lightning-792ee5?logo=pytorchlightning&logoColor=white"></a>
 <a href="https://github.com/pytorch/rl"><img alt="base: TorchRL" src="https://img.shields.io/badge/base-TorchRL-red">
 <a href="https://hydra.cc/"><img alt="config: Hydra" src="https://img.shields.io/badge/config-Hydra-89b8cd"></a> [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
-![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)
+![license](https://img.shields.io/badge/license-Apache%202.0-green.svg?)[![PyPI](https://img.shields.io/pypi/v/rl4co?logo=pypi)](https://pypi.org/project/rl4co)
+[![Test](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml/badge.svg)](https://github.com/kaist-silab/rl4co/actions/workflows/tests.yml)
 <!-- ![testing](https://github.com/kaist-silab/ncobench/actions/workflows/tests.yml/badge.svg) -->
 
 </div>
@@ -247,6 +248,7 @@ RL4CO is built upon:
 
 
 ## Getting started
+<a href="https://colab.research.google.com/github/kaist-silab/rl4co/blob/main/notebooks/1-quickstart.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
 
 RL4CO is now available for installation on `pip`!
 ```bash
@@ -283,7 +285,7 @@ To get started, we recommend checking out our [quickstart notebook](notebooks/1-
 
 Train model with default configuration (AM on TSP environment):
 ```bash
-python run.py  
+python run.py
 ```
 
 
@@ -293,14 +295,14 @@ python run.py
 
 Train model with chosen experiment configuration from [configs/experiment/](configs/experiment/) (e.g. tsp/am, and environment with 42 cities)
 ```bash
-python run.py experiment=tsp/am env.num_loc=42 
+python run.py experiment=tsp/am env.num_loc=42
 ```
 </details>
 
 
 <details>
     <summary>Disable logging</summary>
-    
+
 ```bash
 python run.py experiment=test/am logger=none '~callbacks.learning_rate_monitor'
 ```
@@ -333,7 +335,7 @@ from rl4co.tasks.rl4co import RL4COLitModule
 config = DictConfig(
     {"data": {
             "train_size": 100000,
-            "val_size": 10000, 
+            "val_size": 10000,
             "batch_size": 512,
         },
     "optimizer": {"lr": 1e-4}}
@@ -368,7 +370,7 @@ pytest tests
 ```
 
 ## Contributing
-Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO! 
+Have a suggestion, request, or found a bug? Feel free to [open an issue](https://github.com/kaist-silab/rl4co/issues) or [submit a pull request](https://github.com/kaist-silab/rl4co/pulls). We welcome contributions to RL4CO!
 
 ### Contributors
 <a href="https://github.com/kaist-silab/rl4co/graphs/contributors">

rl4co-0.0.3.dev1/rl4co/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.0.3.dev1"