væt

væt package tools .

This tools used to connect aenet and VASP.

1. Running MD，you can get XDATCAR and OUTCAR

   • PoscarFromXdatacarAse.py

   you can use for get poscar from xdatcar according to you set number and space (need ASE)

   • PoscarFromXdatcar_noASE.py

   you can use for get poscar from xdatcar according to you set number and space ( Don't need ASE)

   • ConvertCoordinate.py

   you can convert fraction poscar to cartesian or reverse. ( need ASE)

2. When you get all POSCAR，you can running static calculate，and for each POSCAR you will get a OUTCAR in a file. So you can use following scripts to get xsf from different type of OUTCAR file(Maybe all md results in only one OUTCAR or static results in each one OUTCAR) ：

   • GenerateXsfFromEachOutcar.py

   For static calculate，you can use this script get XSF

   • GenerateXsfFromTotalMd.py

   For MD，you can use it to get all XSF file from MD's OUTCAR(Don't need XDATCAR)In this process, you can set begain number and end number and space and you'll get "gogal" file which contain all xsf file

3. ANN generate

   • AtomeFingerprint.py

   get “XXX.fingerprint.stp" file for generate

   • GenerateIn.py

   get "generate.in" file for generate Note: You must have “XXX.fingerprint.stp" file in current path

4. ANN train

   • TrainErrorData.py

   By testing the neural network by script, the results can be divided into multiple files, and the data in the train and test process can be extracted for analysis.

5. ANN predict

   • XsfForPredict.py TODO

   • PredictData.py

   You can return the final forecast to two files, force or energy

   • PredictError.py

   Plot figure (TODO)

6. ANN predict

   • ReadAseTrajectory.py

   The result of the ase MD simulation turns out to be poscar. which can be analyzed by default, converted to vasp5 format by default, but it is a bit slow. (you can delete the intermediate vasp4 style function, and this line of ase defaults to vasp4 format, but post-processing is cumbersome, and it is not recommended to do so.)

   • CombinePoscarToXdatcar.py

   The structure generated by the ase simulation MD can be written as a single poscar. this script is used to combine it into xdatcar for analysis of aadrings software and rings software to support the xdatcar format of vasp4, which must be input coordinates from line 8. Therefore, you need to manually delete a line) poscar with vasp4 needs to change 8 of line 104 to 7)

   • CombinePoscarToXdatcarSetNumberSpaceDon'trecommend_nottest.py

   Xdatcar-based on the name of poscar, you can set the output every few times, it is not recommended to use this script.

TODO

7. Else

   • new_vasp2force.py

     Overwritten vasp2force. you don't have to traverse the entire outcar file every time

   • FindDir_fileNumber.py

     Find the number of files in the current path to determine how many xsf files are generated

   • GetTFromOSZICAR.py