pwact 0.2.0__py3-none-any.whl → 0.2.1__py3-none-any.whl

pwact/active_learning/init_bulk/init_bulk_run.py

@@ -5,12 +5,13 @@ from pwact.active_learning.init_bulk.relax import Relax
 from pwact.active_learning.init_bulk.duplicate_scale import do_pertub_work, do_post_pertub, pertub_done, \
     duplicate_scale_done, duplicate_scale, do_post_duplicate_scale
 from pwact.active_learning.init_bulk.aimd import AIMD
+from pwact.active_learning.init_bulk.explore import BIGMODEL
 from pwact.active_learning.init_bulk.relabel import Relabel
 from pwact.active_learning.user_input.init_bulk_input import InitBulkParam
 from pwact.active_learning.user_input.resource import Resource
 from pwact.active_learning.slurm.slurm import scancle_job
-from pwact.utils.constant import INIT_BULK, DFT_STYLE, TEMP_STRUCTURE, PWDATA
-from pwact.utils.file_operation import copy_file, copy_dir, search_files, del_file, del_file_list, write_to_file
+from pwact.utils.constant import INIT_BULK, DFT_STYLE, TEMP_STRUCTURE, PWDATA, LABEL_FILE_STRUCTURE
+from pwact.utils.file_operation import copy_file, copy_dir, search_files, del_file, del_file_list, write_to_file, del_file_list_by_patten
 from pwact.data_format.configop import extract_pwdata
 
 def init_bulk_run(resource: Resource, input_param:InitBulkParam):
@@ -42,13 +43,26 @@ def init_bulk_run(resource: Resource, input_param:InitBulkParam):
             aimd.do_aimd_jobs()
             aimd.do_post_process()
 
-    # do relabel
-    if input_param.is_scf:
-        relabel = Relabel(resource, input_param)
-        if not relabel.check_work_done():
-            relabel.make_scf_work()
-            relabel.do_scf_jobs()
-            relabel.do_post_process()
+    elif input_param.is_bigmodel:
+        bigmodel = BIGMODEL(resource, input_param)
+        if not bigmodel.check_work_done():
+            bigmodel.make_bigmodel_work()
+            bigmodel.do_bigmodel_jobs()
+            bigmodel.do_post_process()
+            
+        # do direct
+        if not bigmodel.check_direct_done():
+            bigmodel.make_direct_work()
+            bigmodel.do_direct_jobs() #  after 
+
+        # do relabel
+        if input_param.is_scf:
+            relabel = Relabel(resource, input_param)
+            if not relabel.check_work_done():
+                relabel.make_scf_work()
+                relabel.do_scf_jobs()
+                relabel.do_post_process()
+                
     do_collection(resource, input_param)
        
 def do_collection(resource: Resource, input_param:InitBulkParam):
@@ -59,6 +73,8 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
     pertub_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.pertub)
     aimd_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.aimd)
     collection_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.collection)
+    bigmodel_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.bigmodel)
+    direct_dir = os.path.join(bigmodel_dir, INIT_BULK.direct)
     
     real_collection_dir = os.path.join(input_param.root_dir, INIT_BULK.collection)
     real_relax_dir = os.path.join(input_param.root_dir, INIT_BULK.relax)
@@ -68,7 +84,9 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
 
     relabel_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.scf)
     real_relabel_dir = os.path.join(input_param.root_dir, INIT_BULK.scf)
-    
+    real_bigmodel_dir = os.path.join(input_param.root_dir, INIT_BULK.bigmodel)
+    real_direct_dir = os.path.join(input_param.root_dir, INIT_BULK.direct)
+     
         
     for init_config in init_configs:
         init_config_name = "init_config_{}".format(init_config.config_index)
@@ -112,24 +130,66 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
                             data_shuffle=input_param.data_shuffle, 
                             interval=1
                 )
-
-        #6 convert relabel datas
-        if init_config.scf:
-            source_scf = search_files(os.path.join(relabel_dir, init_config_name),\
-                 "*/*/*/{}".format(DFT_STYLE.get_aimd_config(resource.scf_style)))
-                 # init/0-aimd/0-scf/OUTCAR
-            if len(source_aimd) == 0:
-                continue
-            source_scf = sorted(source_scf, key=lambda x:int(os.path.basename(os.path.dirname(x)).split('-')[0]), reverse=False)
-            #5. convert the aimd files (for vasp is outcar, for pwmat is movement) to npy format
-            extract_pwdata(input_data_list=source_scf, 
-                    intput_data_format= DFT_STYLE.get_format_by_postfix(os.path.basename(source_scf[0])),
-                    save_data_path  = os.path.join(collection_dir, init_config_name), 
+            # print the dir of pwdatas from aimd
+            if input_param.data_format == PWDATA.extxyz:
+                pwdatas = search_files(real_collection_dir, "*/{}".format(INIT_BULK.get_save_format(input_param.data_format)))
+            elif input_param.data_format == PWDATA.pwmlff_npy: # */PWdata/*.npy
+                tmp = search_files(real_collection_dir, "*/{}/*/position.npy".format(INIT_BULK.get_save_format(input_param.data_format)))
+                pwdatas = [os.path.dirname(_) for _ in tmp]
+            if len(pwdatas) > 0:
+                pwdatas = sorted(pwdatas)
+                result_lines = ["\"{}\",".format(_) for _ in pwdatas]
+                result_lines = "\n".join(result_lines)
+                # result_lines = result_lines[:-1] # Filter the last ','
+                result_save_path = os.path.join(collection_dir, INIT_BULK.npy_format_name)
+                write_to_file(result_save_path, result_lines, mode='w')
+                
+    #6 convert relabel datas
+    result_save_path = []
+    if input_param.is_scf:
+        source_scf = search_files(relabel_dir,\
+                "*/{}".format(DFT_STYLE.get_scf_config(resource.dft_style)))
+                # init/0-aimd/0-scf/OUTCAR
+        source_scf = sorted(source_scf, key=lambda x:int(os.path.basename(os.path.dirname(x))), reverse=False)
+        #5. convert the aimd files (for vasp is outcar, for pwmat is movement) to npy format
+        extract_pwdata(input_data_list=source_scf, 
+                intput_data_format= DFT_STYLE.get_format_by_postfix(os.path.basename(source_scf[0])),
+                save_data_path  = os.path.join(collection_dir, INIT_BULK.scf), 
+                save_data_name  = INIT_BULK.get_save_format(input_param.data_format), 
+                save_data_format= input_param.data_format,
+                data_shuffle=input_param.data_shuffle, 
+                interval=1
+            )
+        result_save_path.append(os.path.join(real_collection_dir, INIT_BULK.scf, INIT_BULK.get_save_format(input_param.data_format)))
+    #7 bigmodel infos
+    if input_param.is_bigmodel:
+        if not input_param.is_scf:
+            extract_pwdata(input_data_list=[os.path.join(direct_dir, INIT_BULK.direct_traj)], 
+                    intput_data_format= PWDATA.extxyz,
+                    save_data_path  = direct_dir, 
                     save_data_name  = INIT_BULK.get_save_format(input_param.data_format), 
                     save_data_format= input_param.data_format,
                     data_shuffle=input_param.data_shuffle, 
                     interval=1
                 )
+            result_save_path.append(os.path.join(real_collection_dir, INIT_BULK.bigmodel, INIT_BULK.direct, INIT_BULK.get_save_format(input_param.data_format)))
+        # copy bigmodel and direct files to realdir
+        copy_dir(bigmodel_dir, os.path.join(collection_dir, INIT_BULK.bigmodel))        
+    # delete logs
+    if len(result_save_path) > 0:
+        _path_path = []
+        for _data_path in result_save_path:
+            if input_param.data_format == PWDATA.extxyz:
+                _path_path.append(_data_path)
+            elif input_param.data_format == PWDATA.pwmlff_npy: # */PWdata/*.npy
+                tmp = search_files(_data_path, "*/position.npy")
+                _path_path.extend([os.path.dirname(_) for _ in tmp])
+            
+        result_lines = ["\"{}\",".format(_) for _ in _path_path]
+        result_lines = "\n".join(result_lines)
+        # result_lines = result_lines[:-1] # Filter the last ','
+        result_save_path = os.path.join(collection_dir, INIT_BULK.npy_format_name)
+        write_to_file(result_save_path, result_lines, mode='w')
 
     # delete link files
     del_file(real_relax_dir)
@@ -137,7 +197,10 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
     del_file(real_pertub_dir)
     del_file(real_aimd_dir)
     del_file(real_relabel_dir)
-
+    del_file(real_bigmodel_dir)
+    # del slurm logs and tags
+    del_file_list_by_patten(os.path.join(collection_dir, INIT_BULK.bigmodel), "slurm-*")
+    del_file_list_by_patten(os.path.join(collection_dir, INIT_BULK.bigmodel, INIT_BULK.direct), "slurm-*")
     # copy collection file to target
     copy_dir(collection_dir, real_collection_dir)
     if not input_param.reserve_work:
@@ -145,29 +208,6 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
         temp_work_dir = os.path.join(input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir)
         del_file_list([temp_work_dir])
 
-    # print the dir of pwdatas from aimd
-    if input_param.data_format == PWDATA.extxyz:
-        pwdatas = search_files(real_collection_dir, "*/{}".format(INIT_BULK.get_save_format(input_param.data_format)))
-    elif input_param.data_format == PWDATA.pwmlff_npy: # */PWdata/*.npy
-        tmp = search_files(real_collection_dir, "*/{}/*/position.npy".format(INIT_BULK.get_save_format(input_param.data_format)))
-        pwdatas = [os.path.dirname(_) for _ in tmp]
-    if len(pwdatas) > 0:
-        pwdatas = sorted(pwdatas)
-        result_lines = ["\"{}\",".format(_) for _ in pwdatas]
-        result_lines = "\n".join(result_lines)
-        # result_lines = result_lines[:-1] # Filter the last ','
-        result_save_path = os.path.join(real_collection_dir, INIT_BULK.npy_format_name)
-        write_to_file(result_save_path, result_lines, mode='w')
-    
-    # print the dir of relabel_pwdatas from relabel
-    relebel_datas = search_files(real_collection_dir, "*/{}".format("scf_pwdata"))
-    if len(relebel_datas) > 0:
-        pwdatas = sorted(relebel_datas)
-        result_lines = ["\"{}\",".format(_) for _ in pwdatas]
-        result_lines = "\n".join(result_lines)
-        # result_lines = result_lines[:-1] # Filter the last ','
-        result_save_path = os.path.join(real_collection_dir, "scf_pwdata")
-        write_to_file(result_save_path, result_lines, mode='w')
 
 def scancel_jobs(work_dir):
     relax_job = os.path.join(work_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.relax)

pwact/active_learning/init_bulk/relabel.py

@@ -16,18 +16,27 @@
     
 """
 import os
-
+import glob
+import json
+import bisect
 from pwact.active_learning.user_input.resource import Resource
 from pwact.active_learning.user_input.init_bulk_input import InitBulkParam
 from pwact.active_learning.init_bulk.duplicate_scale import get_config_files_with_order
 
-from pwact.utils.constant import PWMAT, INIT_BULK, TEMP_STRUCTURE, SLURM_OUT, DFT_STYLE
+from pwact.utils.constant import PWMAT, INIT_BULK, TEMP_STRUCTURE, SLURM_OUT, DFT_STYLE, PWDATA, VASP
 from pwact.active_learning.slurm.slurm import SlurmJob, Mission
 from pwact.utils.slurm_script import get_slurm_job_run_info, split_job_for_group, set_slurm_script_content
     
-from pwact.utils.file_operation import write_to_file, link_file, search_files, del_file_list_by_patten
+from pwact.utils.file_operation import write_to_file, link_file, del_dir, del_file_list_by_patten, get_random_nums
 from pwact.utils.app_lib.common import link_pseudo_by_atom, set_input_script
-from pwact.data_format.configop import save_config, get_atom_type, load_config
+from pwact.data_format.configop import extract_pwdata, save_config, get_atom_type, load_config
+
+import pandas as pd
+from pwdata import Config
+
+# from pwact.utils.constant import DFT_TYPE, VASP, PWDATA, AL_STRUCTURE, TEMP_STRUCTURE,\
+#     LABEL_FILE_STRUCTURE, EXPLORE_FILE_STRUCTURE, LAMMPS, SLURM_OUT, DFT_STYLE, PWMAT, INIT_BULK
+# from pwact.utils.file_operation import write_to_file, copy_file, copy_dir, search_files, mv_file, add_postfix_dir, del_dir, del_file_list_by_patten, link_file
 
 class Relabel(object):
     def __init__(self, resource: Resource, input_param:InitBulkParam):
@@ -42,50 +51,104 @@ class Relabel(object):
         
         self.scf_dir = os.path.join(self.input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.scf)
         self.real_scf_dir = os.path.join(self.input_param.root_dir, INIT_BULK.scf)
-       
-    def make_scf_work(self):
-        scf_paths = []
-        use_dftb = False
-        for init_config in self.init_configs:
-            if init_config.scf is False:
-                continue
-            init_config_name = "init_config_{}".format(init_config.config_index)
-            #1. read construtures from aimd dir
-
-            #2. set relabel dir 
-            #       read trajs from ./aimd/init_config_0/relax/0_aimd/
-            #       make scf dir ./relabel/init_config_0/relax/0_aimd/10-scf/files
-            traj_list = search_files(os.path.join(self.aimd_dir, init_config_name), "*/*aimd")
-            for traj_dir in traj_list:
-                scf_dir = os.path.join(self.scf_dir, init_config_name, \
-                    os.path.basename(os.path.dirname(traj_dir)),\
-                    os.path.basename(traj_dir))
 
-                traj_file_name = DFT_STYLE.get_aimd_config(self.resource.dft_style)
-
-                scf_lsit = self.make_scf_file(
-                    scf_dir      =scf_dir, 
-                    traj_file    =os.path.join(traj_dir, traj_file_name), 
-                    traj_format  =DFT_STYLE.get_format_by_postfix(traj_file_name), 
-                    interval     = self.input_param.interval,
-                    target_format=DFT_STYLE.get_pwdata_format(self.input_param.scf_style, is_cp2k_coord=True),
-                    input_file   =init_config.scf_input_file,
-                    kspacing     =init_config.scf_kspacing, 
-                    flag_symm    =init_config.scf_flag_symm,
-                    is_dftb      = False,
-                    in_skf       =None)
+        self.bigmodel_dir = os.path.join(self.input_param.root_dir, TEMP_STRUCTURE.tmp_init_bulk_dir, INIT_BULK.bigmodel)
+        self.real_bigmodel_dir = os.path.join(self.input_param.root_dir, INIT_BULK.bigmodel)
+        
+        self.direct_dir = os.path.join(self.bigmodel_dir, INIT_BULK.direct)
+        self.real_direct_dir = os.path.join(self.real_bigmodel_dir, INIT_BULK.direct)
 
-                scf_paths.extend(scf_lsit)
-        # make slurm script and slurm job
-        self.make_scf_slurm_job_files(scf_paths, use_dftb)
-    
     def check_work_done(self):
         slurm_remain, slurm_success = get_slurm_job_run_info(self.scf_dir, \
             job_patten="*-{}".format(INIT_BULK.scf_job), \
             tag_patten="*-{}".format(INIT_BULK.scf_tag))
         slurm_done = True if len(slurm_remain) == 0 and len(slurm_success) > 0 else False # len(slurm_remain) > 0 exist slurm jobs need to do
         return slurm_done
-            
+
+    def make_scf_work(self):
+        def find_position_binary(prefix_sum, N):
+            idx = bisect.bisect_right(prefix_sum, N)
+            if idx == 0:
+                return 0
+            elif idx >= len(prefix_sum):
+                return (len(prefix_sum)-1)
+            else:
+                return idx
+
+        def compute_prefix_sum(arr):
+            prefix_sum = []
+            current_sum = 0
+            for num in arr:
+                current_sum += num
+                prefix_sum.append(current_sum)
+            return prefix_sum
+                            
+        candidate = Config(data_path=os.path.join(self.direct_dir, INIT_BULK.direct_traj), format=PWDATA.extxyz)
+        # from idx get config idx
+        candidate_idx = json.load(open(os.path.join(self.direct_dir, INIT_BULK.candidate_idx)))
+        candidate_idx_sum = compute_prefix_sum([candidate_idx[_]['num'] for _ in candidate_idx.keys()])
+        _tmp = Config(data_path=os.path.join(self.direct_dir, INIT_BULK.direct_traj), format=PWDATA.extxyz)
+        scf_dir_list = []
+        if self.input_param.dft_input.scf_max_num is not None:
+            random_list = get_random_nums(0, len(candidate.images), self.input_param.dft_input.scf_max_num, seed=2024)
+        else:
+            random_list = None
+        for index, image in enumerate(candidate.images):
+            if random_list is not None and index not in random_list:
+                continue
+            _idx = find_position_binary(candidate_idx_sum, index)
+            config_idx = candidate_idx["{}".format(_idx)]['idx']
+            scf_dir = os.path.join(self.scf_dir, "{}".format(index))
+            if not os.path.exists(scf_dir):
+                os.makedirs(scf_dir)
+                
+            _tmp.images = [image]
+            _tmp.to(data_path=scf_dir, data_name=PWMAT.atom_config,
+                         format=PWDATA.pwmat_config)
+            self.make_scf_file(
+                    scf_dir      =scf_dir, 
+                    traj_file    =os.path.join(scf_dir, PWMAT.atom_config), 
+                    traj_format  =PWDATA.pwmat_config, 
+                    target_format=DFT_STYLE.get_pwdata_format(self.input_param.dft_style, is_cp2k_coord=True),
+                    input_file   =self.init_configs[config_idx].scf_input_file,
+                    kspacing     =self.init_configs[config_idx].scf_kspacing, 
+                    flag_symm    =self.init_configs[config_idx].scf_flag_symm,
+                    is_dftb      = False,
+                    in_skf       =None)
+
+            scf_dir_list.append(scf_dir)
+        
+        self.make_scf_slurm_job_files(scf_dir_list)
+
+    def make_scf_slurm_job_files(self, scf_dir_list:list[str]):
+        del_file_list_by_patten(self.scf_dir, "*{}".format(INIT_BULK.scf_job))
+        group_list = split_job_for_group(self.resource.dft_resource.group_size, scf_dir_list, self.resource.dft_resource.parallel_num)
+        for group_index, group in enumerate(group_list):
+            if group[0] == "NONE":
+                continue
+            jobname = "scf{}".format(group_index)
+            tag_name = "{}-{}".format(group_index, INIT_BULK.scf_tag)
+            tag = os.path.join(self.scf_dir, tag_name)
+            run_cmd = self.resource.dft_resource.command
+            group_slurm_script = set_slurm_script_content(gpu_per_node=self.resource.dft_resource.gpu_per_node, 
+                number_node = self.resource.dft_resource.number_node, 
+                cpu_per_node = self.resource.dft_resource.cpu_per_node,
+                queue_name = self.resource.dft_resource.queue_name,
+                custom_flags = self.resource.dft_resource.custom_flags,
+                env_script = self.resource.dft_resource.env_script,
+                job_name = jobname,
+                run_cmd_template = run_cmd,
+                group = group,
+                job_tag = tag,
+                task_tag = INIT_BULK.scf_tag, 
+                task_tag_faild = INIT_BULK.scf_tag_failed,
+                parallel_num=self.resource.dft_resource.parallel_num,
+                check_type=None
+                )
+            slurm_script_name = "{}-{}".format(group_index, INIT_BULK.scf_job)
+            slurm_job_file =  os.path.join(self.scf_dir, slurm_script_name)
+            write_to_file(slurm_job_file, group_slurm_script, "w")
+
     def do_scf_jobs(self):
         mission = Mission()
         slurm_remain, slurm_success = get_slurm_job_run_info(self.scf_dir, \
@@ -111,89 +174,59 @@ class Relabel(object):
                 mission.all_job_finished(error_type=SLURM_OUT.dft_out)
                 # mission.move_slurm_log_to_slurm_work_dir()
 
-    '''
-    description: 
-        input_file is scf input control file, for vasp is incar, for pwmat is etot.input
-    return {*}
-    author: wuxingxing
-    '''    
-    def make_scf_file(self, scf_dir:str, traj_file:str, traj_format:str, interval:int, target_format:str, \
-                input_file:str, kspacing:float=None, flag_symm:int=None, is_dftb:bool=False, in_skf:str=None):
-        config = load_config(format=traj_format, config=traj_file)
-        index_list = list(range(0, len(config), interval))
-        scf_lsit = []
-        for index in index_list:
-            save_dir = os.path.join(scf_dir, "{}-{}".format(index, INIT_BULK.scf))
-            if not os.path.exists(save_dir):
-                os.makedirs(save_dir)
-            #1. set config file
-            target_config = save_config(config=config[index], 
-                                        input_format=traj_format,# or None, the same
+    def make_scf_file(self, 
+                    scf_dir, 
+                    traj_file    , 
+                    traj_format  , # the input is pwmat/config
+                    target_format,
+                    input_file   ,
+                    kspacing     =None, 
+                    flag_symm    =None,
+                    is_dftb      =None,
+                    in_skf       =None,
+                    atom_names:list[str]=None):
+        if DFT_STYLE.pwmat == self.resource.dft_style:
+            target_config = traj_file
+            pass
+        else:
+            if DFT_STYLE.vasp == self.resource.dft_style: # when do scf, the vasp input file name is 'POSCAR'
+                save_name = VASP.poscar
+            else:
+                save_name="{}".format(DFT_STYLE.get_normal_config(self.resource.dft_style))# for cp2k this param will be set as coord.xzy
+            target_config = save_config(config=traj_file,
+                                        input_format=traj_format,
                                         wrap = False, 
                                         direct = True, 
                                         sort = True, 
-                                        save_format=target_format, 
-                                        save_path=save_dir, 
-                                        save_name=DFT_STYLE.get_normal_config(self.input_param.scf_style))
+                                        save_name = save_name,
+                                        save_format=DFT_STYLE.get_pwdata_format(dft_style=self.resource.dft_style, is_cp2k_coord=True),
+                                        save_path=scf_dir, 
+                                        atom_names=atom_names)
 
-            atom_type_list, _ = get_atom_type(config[index])
-            #2. set pseudo files
-            # if not is_dftb:
-            pseudo_names = link_pseudo_by_atom(
-                pseudo_list     = self.input_param.dft_input.scf_pseudo, 
-                target_dir      = save_dir, 
-                atom_order      = atom_type_list, 
-                dft_style       = self.resource.scf_style,
+        #2.
+        atomic_name_list, atomic_number_list = get_atom_type(traj_file, traj_format)
+        #1. set pseudo files
+        pseudo_names = link_pseudo_by_atom(
+                pseudo_list = self.input_param.dft_input.pseudo, 
+                target_dir = scf_dir, 
+                atom_order = atomic_name_list, 
+                dft_style = self.resource.dft_style,
                 basis_set_file  =self.input_param.dft_input.basis_set_file,
                 potential_file  =self.input_param.dft_input.potential_file
                 )
-            # else:
-            #     # link in.skf path to aimd dir
-            #     pseudo_names = []
-            #     target_dir = os.path.join(aimd_dir, PWMAT.in_skf)
-            #     link_file(in_skf, target_dir)
-            #3. make dft input file
-            set_input_script(
-                input_file=input_file,
-                config=target_config,
-                dft_style=self.resource.scf_style,
-                kspacing=kspacing, 
-                flag_symm=flag_symm, 
-                save_dir = save_dir,
-                pseudo_names=pseudo_names,
-                gaussian_base_param=self.input_param.dft_input.gaussian_base_param,# these for cp2k
-            )
-            scf_lsit.append(save_dir)
-        return scf_lsit
-
-    def make_scf_slurm_job_files(self, scf_dir_list:list[str],use_dftb: bool=False):
-        del_file_list_by_patten(self.scf_dir, "*{}".format(INIT_BULK.scf_job))
-        group_list = split_job_for_group(self.resource.scf_resource.group_size, scf_dir_list, self.resource.scf_resource.parallel_num)
-        for group_index, group in enumerate(group_list):
-            if group[0] == "NONE":
-                continue
-            jobname = "scf{}".format(group_index)
-            tag_name = "{}-{}".format(group_index, INIT_BULK.scf_tag)
-            tag = os.path.join(self.scf_dir, tag_name)
-            run_cmd = self.resource.scf_resource.command
-            group_slurm_script = set_slurm_script_content(gpu_per_node=self.resource.scf_resource.gpu_per_node, 
-                number_node = self.resource.scf_resource.number_node, 
-                cpu_per_node = self.resource.scf_resource.cpu_per_node,
-                queue_name = self.resource.scf_resource.queue_name,
-                custom_flags = self.resource.scf_resource.custom_flags,
-                env_script = self.resource.scf_resource.env_script,
-                job_name = jobname,
-                run_cmd_template = run_cmd,
-                group = group,
-                job_tag = tag,
-                task_tag = INIT_BULK.scf_tag, 
-                task_tag_faild = INIT_BULK.scf_tag_failed,
-                parallel_num=self.resource.scf_resource.parallel_num,
-                check_type=self.resource.scf_style
-                )
-            slurm_script_name = "{}-{}".format(group_index, INIT_BULK.scf_job)
-            slurm_job_file =  os.path.join(self.scf_dir, slurm_script_name)
-            write_to_file(slurm_job_file, group_slurm_script, "w")
+    
+        #2. make etot.input file
+        set_input_script(
+            input_file=input_file,
+            config=target_config,
+            dft_style=self.resource.dft_style,
+            kspacing=kspacing, 
+            flag_symm=flag_symm, 
+            save_dir = scf_dir,
+            pseudo_names=pseudo_names,
+            gaussian_base_param=self.input_param.dft_input.gaussian_base_param,# these for cp2k
+            is_scf = True
+        )
 
     def do_post_process(self):
         if os.path.exists(self.scf_dir):