pwact 0.2.2.dev0__py3-none-any.whl → 0.2.3__py3-none-any.whl

pwact/active_learning/explore/select_image.py

@@ -90,6 +90,11 @@ def select_image(
     accurate_pd =None
     rand_candi =None
     remove_candi =None
+    
+    error_pd = pd.DataFrame(columns=EXPLORE_FILE_STRUCTURE.devi_columns)
+    accurate_pd = pd.DataFrame(columns=EXPLORE_FILE_STRUCTURE.devi_columns)
+    rand_candi = pd.DataFrame(columns=EXPLORE_FILE_STRUCTURE.devi_columns)
+    remove_candi = pd.DataFrame(columns=EXPLORE_FILE_STRUCTURE.devi_columns)
 
     for md in md_sys_dict.keys():
         sys_dict = md_sys_dict[md]
@@ -104,10 +109,10 @@ def select_image(
                 _lower = lower
                 _higer = higer
             tmp_error_pd, tmp_accurate_pd, tmp_rand_candi, tmp_remove_candi = select_pd(tmp_devi_pd, _lower, _higer, select_num)
-            error_pd = pd.concat([error_pd, tmp_error_pd]) if error_pd is not None else tmp_error_pd
-            accurate_pd = pd.concat([accurate_pd, tmp_accurate_pd]) if error_pd is not None else tmp_accurate_pd
-            rand_candi = pd.concat([rand_candi, tmp_rand_candi]) if error_pd is not None else tmp_rand_candi
-            remove_candi = pd.concat([remove_candi, tmp_remove_candi]) if error_pd is not None else tmp_remove_candi
+            error_pd = pd.concat([error_pd, tmp_error_pd])
+            accurate_pd = pd.concat([accurate_pd, tmp_accurate_pd])
+            rand_candi = pd.concat([rand_candi, tmp_rand_candi])
+            remove_candi = pd.concat([remove_candi, tmp_remove_candi])
     right_md, error_md = count_mdstop_info(model_devi_files)
     md_run_info = "A total of {} MD trajectories were run. with {} trajectories correctly executed and {} trajectories normally completed. \nFor detailed information, refer to File {}.".format(len(right_md) + len(error_md), len(right_md), len(error_md), EXPLORE_FILE_STRUCTURE.md_traj_error_record)
     

pwact/active_learning/init_bulk/explore.py

@@ -296,6 +296,5 @@ class BIGMODEL(object):
             write_to_file(slurm_job_file, group_slurm_script, "w")
 
     def do_post_process(self):
-        if os.path.exists(self.direct_dir):
-            link_file(self.direct_dir, self.real_direct_dir)
-        
+        if os.path.exists(self.bigmodel_dir):
+            link_file(self.bigmodel_dir, self.real_bigmodel_dir)

pwact/active_learning/init_bulk/init_bulk_run.py

@@ -61,8 +61,9 @@ def init_bulk_run(resource: Resource, input_param:InitBulkParam):
             if not relabel.check_work_done():
                 relabel.make_scf_work()
                 relabel.do_scf_jobs()
+                relabel.do_post_process()
                 
-        do_collection(resource, input_param)
+    do_collection(resource, input_param)
        
 def do_collection(resource: Resource, input_param:InitBulkParam):
     init_configs = input_param.sys_config
@@ -83,8 +84,8 @@ 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(real_collection_dir, INIT_BULK.bigmodel)
-    real_direct_dir = os.path.join(real_bigmodel_dir, INIT_BULK.direct)
+    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:
@@ -140,7 +141,7 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
                 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)
+                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
@@ -153,13 +154,13 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
         #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  = relabel_dir, 
+                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_relabel_dir, INIT_BULK.get_save_format(input_param.data_format)))
+        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:
@@ -171,12 +172,20 @@ def do_collection(resource: Resource, input_param:InitBulkParam):
                     data_shuffle=input_param.data_shuffle, 
                     interval=1
                 )
-            result_save_path.append(os.path.join(real_direct_dir, 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))        
+            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:
-        result_lines = ["\"{}\",".format(_) for _ in result_save_path]
+        _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)
@@ -188,9 +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(real_bigmodel_dir, "slurm-*")
-    del_file_list_by_patten(real_direct_dir, "slurm-*")
+    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:

pwact/active_learning/init_bulk/relabel.py

@@ -227,3 +227,7 @@ class Relabel(object):
             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):
+            link_file(self.scf_dir, self.real_scf_dir)

pwact/active_learning/label/labeling.py

@@ -380,16 +380,16 @@ class Labeling(object):
 
         # scf files to pwdata format
         scf_configs = self.collect_scf_configs()
-        
-        extract_pwdata(input_data_list=scf_configs,
-                intput_data_format =DFT_STYLE.get_format_by_postfix(os.path.basename(scf_configs[0])),
-                save_data_path =self.result_dir,
-                save_data_name = INIT_BULK.get_save_format(self.input_param.data_format),
-                save_data_format = self.input_param.data_format,
-                data_shuffle     =self.input_param.train.data_shuffle
-        )
-        # copy to main dir
-        copy_dir(self.result_dir, self.real_result_dir)
+        if len(scf_configs) > 0:
+            extract_pwdata(input_data_list=scf_configs,
+                    intput_data_format =DFT_STYLE.get_format_by_postfix(os.path.basename(scf_configs[0])),
+                    save_data_path =self.result_dir,
+                    save_data_name = INIT_BULK.get_save_format(self.input_param.data_format),
+                    save_data_format = self.input_param.data_format,
+                    data_shuffle     =self.input_param.train.data_shuffle
+            )
+            # copy to main dir
+            copy_dir(self.result_dir, self.real_result_dir)
 
     def do_post_bigmodel(self):
         # copy the bigmodel labeled.xyz to result

pwact/active_learning/train/train_model.py

@@ -33,6 +33,7 @@ class ModelTrian(object):
         self.itername = itername
         self.resource = resource
         self.input_param = input_param
+        self.train_flag = True
         self.iter = get_iter_from_iter_name(self.itername)
         # train work dir
         self.train_dir = os.path.join(self.input_param.root_dir, self.itername, TEMP_STRUCTURE.tmp_run_iter_dir, AL_STRUCTURE.train)
@@ -68,11 +69,16 @@ class ModelTrian(object):
             if not os.path.exists(model_i_dir):
                 os.makedirs(model_i_dir)
             # make train.json file
-            train_dict = self.set_train_input_dict(work_dir=model_i_dir, model_index = model_index)
+            train_dict, train_tag = self.set_train_input_dict(work_dir=model_i_dir, model_index = model_index)
             train_json_file_path = os.path.join(model_i_dir, TRAIN_FILE_STRUCTUR.train_json)
             save_json_file(train_dict, train_json_file_path)
             train_list.append(model_i_dir)
-        self.make_train_slurm_job_files(train_list)
+        if train_tag:
+            self.make_train_slurm_job_files(train_list)
+        else:
+            pre_iter_name = make_iter_name(self.iter - 1)
+            pre_iter_dir = os.path.join(self.input_param.root_dir, pre_iter_name, AL_STRUCTURE.train)
+            copy_dir(pre_iter_dir, self.train_dir)
     
     def make_train_slurm_job_files(self, train_list:list[str]):
         # make train slurm script
@@ -158,6 +164,7 @@ class ModelTrian(object):
         # search train_feature_path in iter*/label/result/*/PWdata/*
         iter_index = get_iter_from_iter_name(self.itername)
         start_iter = 0
+        train_tag = True
         while start_iter < iter_index:
             if self.input_param.data_format == PWDATA.extxyz: # result/train.xyz
                 iter_data_list = search_files(self.input_param.root_dir, 
@@ -172,6 +179,9 @@ class ModelTrian(object):
                 train_feature_path.extend(iter_data_list)
             start_iter += 1
         
+        if start_iter > 0 and len(iter_data_list) == 0:
+            train_tag = False
+
         # reset seed
         train_json[TRAIN_INPUT_PARAM.seed] = get_seed_by_time()
         train_json[TRAIN_INPUT_PARAM.raw_files] = []
@@ -181,7 +191,7 @@ class ModelTrian(object):
         train_json[TRAIN_INPUT_PARAM.format] = self.input_param.data_format
         if self.input_param.strategy.uncertainty == UNCERTAINTY.kpu:
             train_json[TRAIN_INPUT_PARAM.save_p_matrix] = True
-        return train_json
+        return train_json, train_tag
 
     def do_train_job(self):
         mission = Mission()

pwact/active_learning/user_input/resource.py

@@ -126,7 +126,7 @@ class ResourceDetail(object):
         if self.gpu_per_node is None and self.cpu_per_node is None:
             raise Exception("ERROR! Both CPU and GPU resources are not specified!")
         # check param
-        if "$SLURM_NTASKS".lower() in command.lower():
+        if "$SLURM".lower() in command.lower():
             pass
         else:
             if "mpirun -np" in command:

pwact/main.py

@@ -79,7 +79,7 @@ def run_iter():
             if jj == 2 and not input_param.reserve_work: # delete temp_work_dir under current iteration after the labeling done
                 del_file_list([os.path.join(input_param.root_dir, iter_name, TEMP_STRUCTURE.tmp_run_iter_dir)])
 
-    print("Active learning done! \nYou could use cmd 'al_pwmlff gather_pwdata' to collect all datas sampled from iterations.")
+    print("Active learning done! \nYou could use cmd 'pwact gather_pwdata' to collect all datas sampled from iterations.")
 
 def run_fp(itername:str, resource : Resource, input_param: InputParam):
     lab = Labeling(itername, resource, input_param)

pwact-0.2.3.dist-info/METADATA

@@ -0,0 +1,17 @@
+Metadata-Version: 2.1
+Name: pwact
+Version: 0.2.3
+Summary: PWACT is an open-source automated active learning platform based on PWMLFF for efficient data sampling.
+Home-page: https://github.com/LonxunQuantum/PWact
+Author: LonxunQuantum
+Author-email: lonxun@pwmat.com
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+
+# Dependencies
+
+Please refer to the [`user manual`](http://doc.lonxun.com/PWMLFF/pwact/)

{pwact-0.2.2.dev0.dist-info → pwact-0.2.3.dist-info}/RECORD

@@ -1,20 +1,20 @@
 pwact/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pwact/main.py,sha256=FubLfN4QaqPEe7aYGzZhqvT2T2GEvMyU3hy07jvXEWU,15379
+pwact/main.py,sha256=jCqNp0tn8P7R8EdPt81bqZlC-Ixjp_rnnFxMT7JAfXI,15375
 pwact/active_learning/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/environment.py,sha256=1vzOWnCNR2OXMFKz4DjQF2wb0aw1Xkjnz7cy1YGQrus,512
 pwact/active_learning/explore/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/explore/run_model_md.py,sha256=n3f6aSMr1Cpw_cSUrlH28JwoWW7VLwUMcF1XB0ao2sw,27448
-pwact/active_learning/explore/select_image.py,sha256=Qbdk4zhEfkvefY4MPtu7ErDL0XVfgfrBFPZUGDyiGF8,14468
+pwact/active_learning/explore/select_image.py,sha256=1zpTCs_15FRlffBYfy1HQ1r79p2dNM8u8ZHvPysQTvc,14597
 pwact/active_learning/init_bulk/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/init_bulk/aimd.py,sha256=XzDlX2vylaljQKoUnv6nrI2NfiOdHZpq8qr3DenA1F4,10465
 pwact/active_learning/init_bulk/direct.py,sha256=TBB7f13wbbURDgn4_IkgRZ-rwQ3p4gviNVYIAldi97U,9680
 pwact/active_learning/init_bulk/duplicate_scale.py,sha256=hlGAr27YJhOElMz5GMyN0wWoChia4Jj_tvENsJ3wSBs,9481
-pwact/active_learning/init_bulk/explore.py,sha256=ef1KA4Sii7sXSGncvomsLXTBtFM89H8Q5lYWGoAbN2U,15935
-pwact/active_learning/init_bulk/init_bulk_run.py,sha256=t-lCiraAtHEUtW-rdhfRuXYoWpP4eJndX3F3E6IxWt8,10871
-pwact/active_learning/init_bulk/relabel.py,sha256=I5vSmyDaCJtcJCWS6aAtlb72kAQSva4vCx4mn1BYXjg,11846
+pwact/active_learning/init_bulk/explore.py,sha256=IAxdOHPzCWjfvrHywyOuBQG2V76rpBZtsyNLQyIMThs,15932
+pwact/active_learning/init_bulk/init_bulk_run.py,sha256=QX0x1-DevQPSYP1PlWji3hZN55DbeSXJJxgCZZkJXBM,11514
+pwact/active_learning/init_bulk/relabel.py,sha256=gP_qqMWT_hMIcwFBdNmyVBHBpkQS52BzbQK80rGTvPQ,11974
 pwact/active_learning/init_bulk/relax.py,sha256=edyCZLEylUckIwOLBa55agUMo-aedj0dvoyG165YpuE,10450
 pwact/active_learning/label/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pwact/active_learning/label/labeling.py,sha256=JUdyaSvpspUwbMnhi65i6Cna0vWgMOAMNS1XTtqlygc,21740
+pwact/active_learning/label/labeling.py,sha256=Vq2hH9i3DH4X0XidrzxwE-MzgYiWQ0VkFWZKj1Lhc_I,21800
 pwact/active_learning/slurm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/slurm/slurm.py,sha256=-Dmc2B9y4_27THCR2tZFXnwJvMmifivJeZ2FX8iy2Tw,16881
 pwact/active_learning/slurm/slurm_tool.py,sha256=-4tc5dkpAUP0vmEdmqM8oYLcsUwixa4Z8h5_E9Gevdg,1249
@@ -22,12 +22,12 @@ pwact/active_learning/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJ
 pwact/active_learning/test/test.py,sha256=TXsk-gDDOHDLywVljpqA-zHK3-PrZbGAHnK1T7T6T-o,3587
 pwact/active_learning/train/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/train/dp_kpu.py,sha256=GkGKEGhLmOvPERqgTkf_0_vD9zOEPlBX2N7vuSQG_-c,9317
-pwact/active_learning/train/train_model.py,sha256=F6DUOuTsYGvinv-qz2iqa2j7DPy3hnCySVUkx8DpR_U,11605
+pwact/active_learning/train/train_model.py,sha256=Hh6_yq_86Y4pqP60sbEaeCJexD9a-BTqop07l2FAQSI,12021
 pwact/active_learning/user_input/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/active_learning/user_input/cmd_infos.py,sha256=g1QW5Wi3JvmC5xEY4F-7AO1xBuJSjHG8IhZjom8haxQ,3728
 pwact/active_learning/user_input/init_bulk_input.py,sha256=M8IlT4ABJUEwomb6eQGq8d3h1darMHtDlws1fcq077k,9422
 pwact/active_learning/user_input/iter_input.py,sha256=7k0M3BaP1dMWK2Bglvb2WXQ2Ugrfy0v7yMqPqWqNCdE,12744
-pwact/active_learning/user_input/resource.py,sha256=28rbsJ9qtkToeuXst7Q1Z189KpDd8YZ8bBK7cuDp7RU,7358
+pwact/active_learning/user_input/resource.py,sha256=OtsJEAWmMknLyrPB309C6ipyL1fCXHriF8sjXs5pjmg,7351
 pwact/active_learning/user_input/scf_param.py,sha256=QX_Dd6c9Q2sjV4lIKjfV7t-DgIwDKZp7fh2KY5HyKOU,11262
 pwact/active_learning/user_input/workdir.py,sha256=64J3wBbYzy6ztYHs513YHfEIF5d8zwv_WpF-U9HOEkA,8245
 pwact/active_learning/user_input/train_param/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -59,9 +59,9 @@ pwact/utils/app_lib/lammps.py,sha256=t3nkBqwkViO6UbnInGKAHiwVngHKayeat2pPEtaZZIU
 pwact/utils/app_lib/pwmat.py,sha256=PTRPkG_d00ibGhpCe2-4M7MW3dx2ZuAyb9hT2jl_LAs,18047
 pwact/utils/draw/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pwact/utils/draw/hist_model_devi.py,sha256=o1Try-ekith2A7S6u1mt3zuTqaQwyw5tdURReh8BeLY,4945
-pwact-0.2.2.dev0.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-pwact-0.2.2.dev0.dist-info/METADATA,sha256=RGt1qWZv8xFV4H4W3oFbtZGmMs_sPK73JRJFwm2ZJbQ,3663
-pwact-0.2.2.dev0.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-pwact-0.2.2.dev0.dist-info/entry_points.txt,sha256=p61auAnpbn8E2WjvHNBA7rb9_NRAOCew4zdcCj33cGc,42
-pwact-0.2.2.dev0.dist-info/top_level.txt,sha256=fY1_7sH5Lke4dC9L8MbYM4fT5aat5eCkAmpkIzY1SlM,6
-pwact-0.2.2.dev0.dist-info/RECORD,,
+pwact-0.2.3.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+pwact-0.2.3.dist-info/METADATA,sha256=CBxD2xEcgy3Lqsw7BUkG1BHvOxwQUv5e2OTa9r1GPVk,612
+pwact-0.2.3.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+pwact-0.2.3.dist-info/entry_points.txt,sha256=p61auAnpbn8E2WjvHNBA7rb9_NRAOCew4zdcCj33cGc,42
+pwact-0.2.3.dist-info/top_level.txt,sha256=fY1_7sH5Lke4dC9L8MbYM4fT5aat5eCkAmpkIzY1SlM,6
+pwact-0.2.3.dist-info/RECORD,,

pwact-0.2.2.dev0.dist-info/METADATA

@@ -1,107 +0,0 @@
-Metadata-Version: 2.1
-Name: pwact
-Version: 0.2.2.dev0
-Summary: PWACT is an open-source automated active learning platform based on PWMLFF for efficient data sampling.
-Home-page: https://github.com/LonxunQuantum/PWact
-Author: LonxunQuantum
-Author-email: lonxun@pwmat.com
-Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
-Classifier: Operating System :: OS Independent
-Requires-Python: >=3.9
-Description-Content-Type: text/markdown
-License-File: LICENSE
-
-# Dependencies
-
-Please refer to the [`user manual`](http://doc.lonxun.com/PWMLFF/active%20learning/)
-
-1. AL-PWMLFF job scheduling uses the [SLURM](https://slurm.schedmd.com/documentation.html) cluster management and job scheduling system. SLURM must be installed on your computing cluster.
-
-2. DFT calculations in AL-PWMLFF support [PWmat](https://www.pwmat.com/gpu-download), [VASP](https://www.vasp.at/), [CP2K](https://www.cp2k.org/) and DFTB. We have integrated DFTB in PWmat. You can find detailed usage instructions in the `DFTB_DETAIL section` of the [`PWmat Manual`](http://www.pwmat.com/pwmat-resource/Manual.pdf).
-
-3. AL-PWMLFF model training is based on [`PWMLFF`](https://github.com/LonxunQuantum/PWMLFF). Refer to the [`PWMLFF documentation`](http://doc.lonxun.com/PWMLFF/Installation) for installation instructions ([`Download address for PWmat version integrated with DFTB`](https://www.pwmat.com/modulefiles/pwmat-resource/mstation-download/cuda-11.6-mstation-beta.zip)).
-
-4. AL-PWMLFF Lammps molecular dynamics simulation is based on [Lammps_for_pwmlff](https://github.com/LonxunQuantum/Lammps_for_PWMLFF/tree/libtorch). Refer to the [`Lammps_for_pwmlff documentation`](https://github.com/LonxunQuantum/Lammps_for_PWMLFF/blob/libtorch/README) for installation instructions.
-
-# Installation Process
-You can install it through the pip command or the github source code installation.
-
-## install by pip
-
-```bash
-    pip install pwact
-```
-## from github
-
-### Code Download
-
-    git clone https://github.com/LonxunQuantum/PWact.git
-
-Then import environment variable.
-
-```
-export PATH=/data/home/wuxingxing/codespace/al_pwmlff/bin:$PATH
-```
-AL-PWMLFF is developed in Python and supports Python 3.9 and above. It is recommended to use the Python runtime environment provided by PWMLFF.
-
-If you need to create a virtual environment for AL-PWMLFF separately, you only need to install the following dependent packages (compatible with your Python version, Python >= 3.9).
-```bash
-pip install numpy pandas tqdm pwdata
-```
-
-    
-# Command List
-
-AL-PWMLFF includes the following commands, which are not case sensitive. The starting command is `pwact`
-
-### 1. Display the available command list
-
-```bash
-pwact  [ -h / --help / help ]
-```
-
-### 2. Display the parameter list for cmd_name:
-
-```bash
-pwact cmd_name -h
-```
-
-### 3. Initial Training Set Preparation
-
-```bash
-pwact init_bulk param.json resource.json
-```
-
-### 4. Active Learning
-
-```bash
-pwact run param.json resource.json
-```
-
-For the 3-th and 4-th command above, the names of the JSON files can be modified by the user, but it is required that the input order of [`param.json`](#paramjson) and [`resouce.json`](#resourcejson) cannot be changed.
-
-### 5. Tool Commands
-
-Convert MOVEMENT or OUTCAR to PWdata format
-
-```bash
-pwact to_pwdata
-```
-
-Search for labeled datasets in the active learning directory
-
-```bash
-pwact gather_pwdata
-```
-
-### examples download
-from github
-``` 
-https://github.com/LonxunQuantum/PWact/tree/main/pwact/example
-```
-
-from BaiduNetdisk included the calculation results of examples
-```
-https://pan.baidu.com/s/14E0u_7cpntiBZgg-C1S5XA?pwd=pwmt
-```