Update README.md

README.md

@@ -15,11 +15,11 @@ tags:
 
 ## Introduction
 
-  ***Combinatorial Optimization Problems(COPs)*** has long been an active field of research. Generally speaking, there exists two main approachs for solving COPs, each of them having pros and cons. On the one hand, the *exact algorithms* can find the optimal solution, but they may be prohibitive for solving large instances because of the exponential increate of the execution time.
-  On the other hand, *heuristic algorithms* can compute solutions efficiently, but are not able to prove the optimality of solutions.
+  ***Combinatorial Optimization Problems (COPs)*** has long been an active field of research. Generally speaking, there exists two main approaches for solving COPs, each of them having pros and cons. On one hand, the *exact algorithms* can find the optimal solution, but they may be prohibitive for solving large instances because of the exponential increase of the execution time.
+  On the other hand, *heuristic algorithms* can compute solutions efficiently, but are not able to guarantee the optimality of solutions.
 
-  In the realistic business scenarios, COPs are usually large-scale(>=1000 nodes), which have very strict requirements for the execution time and performance of solutions. To better solve these problems, we
-  proposes a generic and complete solver, named **🤠GreedRL**, based on **Deep Reinforcement Learning(DRL)**, which achieves better speed and performance of solutions than *heuristic algorithms* .
+  In the realistic business scenarios, COPs are usually large-scale (>=1000 nodes), which have very strict requirements for the execution time and performance of solutions. To better solve these problems, we
+  propose a generic and complete solver, named **🤠GreedRL**, based on **Deep Reinforcement Learning (DRL)**, which achieves improved speed and performance of solutions than *heuristic algorithms* .
 
 ## 🏆Award
 
@@ -34,11 +34,11 @@ tags:
 
 * **HIGH-PERFORMANCE**
 
-  🤠GreedRL have improved the DRL environment(Env) simulation speed by **CUDA and C++ implementations**. At the same time, we have also implemented some **Operators** to replace the native operators of PyTorch, like *Masked Matrix Multiplication* and *Masked Additive Attention*, to achive the ultimate computing performance.
+  🤠GreedRL have improved the DRL environment (Env) simulation speed by **CUDA and C++ implementations**. At the same time, we have also implemented some **Operators** to replace the native operators of PyTorch, like *Masked Matrix Multiplication* and *Masked Additive Attention*, to achive the ultimate computing performance.
   
 * **USER-FRIENDLY**
 
-  🤠GreedRL have **warped commonly used modules**, such as Neural Network(NN) components, RL training algothrim and COPs constraints implementations, which makes it easy to use.
+  🤠GreedRL have **warped commonly used modules**, such as Neural Network (NN) components, RL training algorithms and COPs constraints implementations, which makes it easy to use.
 
 ## Architecture
   ![](./images/GREEDRL-Framwork_en.png)
@@ -46,8 +46,7 @@ tags:
 ## COPs Modeling examples
 
 
-### 
-Capacitated Vehicle Routing Problem (CVRP)
+### Capacitated Vehicle Routing Problem (CVRP)
 <details>
     <summary>CVRP</summary>
 
@@ -564,14 +563,14 @@ class Objective:
 
 We are delighted to release 🤠GreedRL Community Edition, as well as pretrained models, which are specialized to CVRP with problem size ranging from 100 to 5000 nodes. 
 
-The model is trained using a deep reinforcement learning(DRL) algorithm known as REINFORCE. The model consists of two main components, an Encoder and a Decoder. The encoder produces embedding of all input nodes. The decoder then generates a solution sequence autoregressively. Feasibility of the solution is ensured by a *mask* procedure that prevents the model from selecting nodes that would result in a violation of constraints, e.g. exceeding the vehicle capacity.
+The model is trained using a deep reinforcement learning (DRL) algorithm known as REINFORCE. The model consists of two main components, an Encoder and a Decoder. The encoder produces embedding of all input nodes. The decoder then generates a solution sequence autoregressively. Feasibility of the solution is ensured by a *mask* procedure that prevents the model from selecting nodes that would result in a violation of constraints, e.g. exceeding the vehicle capacity.
 
 
 ## Intended uses & limitations
 
-You can use these default models for solving the Capacitated VRP(CVRP) with deep reinforcement learning(DRL).
+You can use these default models for solving the Capacitated VRP (CVRP) with deep reinforcement learning(DRL).
 
-These model is limited by its training dataset, this may not generalize well for all use cases in different domains.
+These models are limited by the training dataset, which may not generalize well for all use cases in different domains.
 
 ## How to use