空间站广场

论文

Notebooks

比赛

课程

Apps

我的主页

我的Notebooks

我的论文库

我的足迹

我的工作空间

任务

节点

文件

数据集

镜像

项目

数据库

公开

Uni-MOF: A Tool for Predicting the Full Adsorption Phase Diagram of MOF Materials

Yani Guan

更新于 2024-10-18

推荐镜像 :Uni-Mol:unimol-qsar:v0.5

推荐机型 :c2_m4_cpu

数据集

AI4SCUP-CNS-BBB(v1)

Uni-MOF: A Tool for Predicting the Full Adsorption Phase Diagram of MOF Materials

©️ Copyright 2023 @ Authors
Author: Zhifeng Gao 📨
Date: 2023-06-06
License: This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Quick Start: Click the Connect button above, select the unimol-qsar:0530 image and any GPU node configuration, then wait a moment to start running.

代码

文本

import Uni-MOF

代码

文本

[1]

# image：unimol-qsar:v0.5, GPU

# import unimol

from unimol import MOFPredictor

import os

/opt/conda/lib/python3.8/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

代码

文本

[2]

%%bash

# Download sample data, MOF core database. The data source is quite large and will take about 3 minutes to download. Once the download is complete, you can comment out this section.

rm -rf mof_database

rm -rf mof_database.tar.gz

wget -nv https://bohrium-example.oss-cn-zhangjiakou.aliyuncs.com/unimol-qsar/mof_database.tar.gz

tar --no-same-owner -xzf mof_database.tar.gz

find ./mof_database -type f > mof_name.txt

2024-07-23 15:20:47 URL:https://bohrium-example.oss-cn-zhangjiakou.aliyuncs.com/unimol-qsar/mof_database.tar.gz [28528154/28528154] -> "mof_database.tar.gz" [1]

代码

文本

[3]

!shuf -n 10 mof_name.txt

./mof_database/XAGCUE_clean.cif
./mof_database/WAGDAL_clean.cif
./mof_database/ic501898x_si_005_clean.cif
./mof_database/ACUTOI_clean.cif
./mof_database/AMOFAK_clean.cif
./mof_database/NEJSEC_clean.cif
./mof_database/LUCDES_clean.cif
./mof_database/DUQCAU_clean.cif
./mof_database/IYOVIC_clean.cif
./mof_database/QITDOO_clean.cif

代码

文本

MOF Performance Prediction Under Different Conditions

代码

文本

[4]

clf = MOFPredictor()

gas_list = ['CH4','CO2','Ar','Kr','Xe','O2','He','N2','H2']

gas = 'CH4'

cif_dir = './mof_database'

mof_name = 'FAQLUH_clean'

res = clf.predict_grid(cif_path=os.path.join(cif_dir, f'{mof_name}.cif'),

gas=gas,

temperature_list=[190,298],

pressure_bins=8)

2024-07-23 15:20:49 | unimol/models/unimol.py | 111 | INFO | Uni-Mol(QSAR) | Loading pretrained weights from /opt/conda/lib/python3.8/site-packages/unimol-0.0.2-py3.8.egg/unimol/weights/mof_pre_no_h_CORE_MAP_20230505.pt

代码

文本

[5]

res

	temperature	pressure	absorp_prediction	gas	mof
0	190	1.000000	0.058289	CH4	FAQLUH_clean.cif
1	190	5.179475	0.179443	CH4	FAQLUH_clean.cif
2	190	26.826958	1.232422	CH4	FAQLUH_clean.cif
3	190	138.949549	4.527344	CH4	FAQLUH_clean.cif
4	190	719.685673	30.718750	CH4	FAQLUH_clean.cif
5	190	3727.593720	144.000000	CH4	FAQLUH_clean.cif
6	190	19306.977289	208.250000	CH4	FAQLUH_clean.cif
7	190	100000.000000	239.875000	CH4	FAQLUH_clean.cif
8	298	1.000000	0.034760	CH4	FAQLUH_clean.cif
9	298	5.179475	0.019730	CH4	FAQLUH_clean.cif
10	298	26.826958	0.044922	CH4	FAQLUH_clean.cif
11	298	138.949549	0.057251	CH4	FAQLUH_clean.cif
12	298	719.685673	0.297852	CH4	FAQLUH_clean.cif
13	298	3727.593720	2.353516	CH4	FAQLUH_clean.cif
14	298	19306.977289	9.218750	CH4	FAQLUH_clean.cif
15	298	100000.000000	43.218750	CH4	FAQLUH_clean.cif

代码

文本

Prediction of MOF Adsorption Properties

Predicting adsorption performance based on the given MOF structure, gas molecules, temperature, and pressure.

代码

文本

[6]

clf = MOFPredictor()

gas_list = ['CH4','CO2','Ar','Kr','Xe','O2','He','N2','H2']

cif_dir = './mof_database'

gas = 'CH4'

mof_name = 'FAQLUH_clean'

pressure = 10000

temperature = 298

assert gas in gas_list

assert pressure<100000 and pressure>0

assert temperature>0 and temperature<400

s = clf.single_predict(cif_path=os.path.join(cif_dir, f'{mof_name}.cif'),

gas=gas,

pressure=pressure,

temperature=temperature)

print(f'gas:{gas}-pressure:{pressure}-temperature:{temperature}-absorb_predict: {s}')

2024-07-23 15:20:53 | unimol/models/unimol.py | 111 | INFO | Uni-Mol(QSAR) | Loading pretrained weights from /opt/conda/lib/python3.8/site-packages/unimol-0.0.2-py3.8.egg/unimol/weights/mof_pre_no_h_CORE_MAP_20230505.pt
gas:CH4-pressure:10000-temperature:298-absorb_predict: 3.977489017487298

代码

文本

[ ]

代码

文本

点个赞吧