# Kinetic Monte Carlo of Systems (kmcos)

[toc]

- [Guide]((https://github.com/kmcos/intro2kmcos)) I followed
- Check for [issues]((https://github.com/kmcos/kmcos/issues)) in [kmcos repo](https://github.com/kmcos/kmcos)
> Note that **kmcos** is the newer version for Python3, while [**kmos**](https://github.com/mhoffman/kmos) is for Python2 (many bugs!!)

## Installation (Ubuntu, Serial)

1. Creating a Python Virtual Environment

   cd ~
   sudo apt-get update
   sudo apt-get install python3
   sudo apt-get install virtualenv
   virtualenv -p /usr/bin/python3 ~/VENV/kmcos
   source ~/VENV/kmcos/bin/acti

2. Fetch latest version of kmcos

   sudo apt install git
   git clone https://github.com/kmcos/kmcos-installers
   cd kmcos-installers
   bash install-kmcos-linux-venv.bas

他的Viewer GUI現在問題還很多，這部分安裝就先跳過

## Run Interactively

1. 先用`intro2kmcos`的資料試跑

   source ~/VENV/kmcos/bin/activate #確定有在VENV
   git clone https://github.com/kmcos/intro2kmcos
   cd ~/intro2kmcos/task_material/COoxRuO2_local_smart

2. Save + compile the model

   python3 COoxRuO2__build.py

This creates:

- folder `COoxRuO2_local_smart` : compiled model
- `COoxRuO2.xml` : contains data defining the model (copy it to `COoxRuO2_local_smart`)

[More about COoxRuO2__build.py](https://hackmd.io/@princesswinnie/B14ey8MJA)

> 如果compile的時候有error `Error: Symbol ‘get_seed’ at (1) has no IMPLICIT type; did you mean ‘get_seedf2pywrap’?`
是因為NumPy的f2py有問題，再裝一次舊版本就可以解決：`pip install numpy==1.24.2`

3. Launch API

   cd COoxRuO2_local_smart
   python3 kmc_settings.py run

4. Run directly !!

   NSTEPS = 1e6
   model.do_steps(NSTEPS)
   
   model.show()


## Useful commands

- Parameters

  model #查看所有參數和其當前值
  
  #重設參數
  model.parameters.T = 550 #溫度(k)
  model.parameters.p_COgas = 0.5 #壓力(bar)

- TOFs (in the form of NumPy arrays)
Reaction rate is often expressed as **turnover frequency (TOF)**, the average rate of molecule production per second per surface site (or surface area). In this case we refer to the production rate of $CO_2$ .

  #To get the headers of these arrays
  model.get_occupation_header()
  model.get_tof_heade
  
  #Sample an average model
  model.get_std_sampled_data(samples, sample_size, tof_method='integ',
  output='str')
  
  #To get all info in a list
  data = model.get_std_sampled_data(samples=10, sample_size=1e6)
  L = data.split(' ')
  print(L)
  
  #Return info in the form of Python dictionary
  output_dict = model.get_std_sampled_data(samples=10, sample_size=1e6,
  output = 'dict')
  print(output_dict)
  TOF_CO = output_dict['CO_oxidation']
  print(TOF_CO)

- Print details

  #current occupations (averaged coverages)
  model.print_coverages () 
  
  #number of times each process has been executed (since the simulation starts)
  model.print_procstat () 

- Analyze & View model

  atoms = model.get_atoms ()
  model.show ()
  
  #Some additional data in the "atoms" object
  atoms.occupation
  atoms.tof_data
  atoms.kmc_time
  atoms.kmc_ste

## Build models