pwact 0.1.22__tar.gz → 0.1.24__tar.gz

{pwact-0.1.22 → pwact-0.1.24}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pwact
-Version: 0.1.22
+Version: 0.1.24
 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

{pwact-0.1.22 → pwact-0.1.24}/pwact/active_learning/label/labeling.py

@@ -245,7 +245,7 @@ class Labeling(object):
             for sub_md_sys in sub_md_sys_dir_list:
                 out_mlmd_list =search_files(sub_md_sys, "*-{}/{}".format(LABEL_FILE_STRUCTURE.scf, DFT_STYLE.get_scf_config(self.resource.dft_style)))
                 # do a sorted?
-                md_sys_mlmd.append(out_mlmd_list)
+                md_sys_mlmd.extend(out_mlmd_list)
         return md_sys_mlmd
 
     def get_aimd_list(self):
@@ -279,14 +279,23 @@ class Labeling(object):
 
         # scf files to pwdata format
         scf_configs = self.collect_scf_configs()
-        for scf_md in scf_configs:
-            datasets_path_name = os.path.basename(os.path.dirname(os.path.dirname(scf_md[0])))#md.001.sys.001.t.000.p.000
-            extract_pwdata(data_list=scf_md,
-                data_format      =DFT_STYLE.get_format_by_postfix(os.path.basename(scf_md[0])),
-                datasets_path    =os.path.join(self.result_dir, datasets_path_name),
+
+        extract_pwdata(data_list=scf_configs,
+                data_format      =DFT_STYLE.get_format_by_postfix(os.path.basename(scf_configs[0])),
+                datasets_path    =self.result_dir,
                 train_valid_ratio=self.input_param.train.train_valid_ratio, 
                 data_shuffle     =self.input_param.train.data_shuffle, 
                 merge_data       =True
-            )
+        )
+
+        # for id, scf_md in enumerate(scf_configs):
+        #     datasets_path_name = os.path.basename(os.path.dirname(os.path.dirname(scf_md[0])))#md.001.sys.001.t.000.p.000
+        #     extract_pwdata(data_list=scf_md,
+        #         data_format      =DFT_STYLE.get_format_by_postfix(os.path.basename(scf_md[0])),
+        #         datasets_path    =os.path.join(self.result_dir, "{}-{}".format(id, datasets_path_name)),
+        #         train_valid_ratio=self.input_param.train.train_valid_ratio, 
+        #         data_shuffle     =self.input_param.train.data_shuffle, 
+        #         merge_data       =True
+        #     )
         # copy to main dir
         copy_dir(self.result_dir, self.real_result_dir)

{pwact-0.1.22 → pwact-0.1.24}/pwact/active_learning/user_input/cmd_infos.py

@@ -22,6 +22,7 @@ def cmd_infos(cmd_type=None):
         cmd_info += "'pwact kill run' for 'run' tasks\n\n"
 
         cmd_info += "to_pwdata".lower() + "\n"
+        cmd_info += "(This method has been abandoned, new conversion methods refer to the pwdata documentation http://doc.lonxun.com/PWMLFF/Appendix-2/)\n\n"
         cmd_info += "you could use this method to change outcars or movements to pwdata format.\nFor more detail for this command, you could use 'PWact to_pwdata -h'\n\n"
 
         cmd_info += "gather_pwdata".lower() + "\n"

{pwact-0.1.22 → pwact-0.1.24}/pwact/data_format/configop.py

@@ -150,57 +150,32 @@ def extract_pwdata(data_list:list[str],
     # if data_format == DFT_STYLE.cp2k:
     #     raise Exception("not relized cp2k pwdata convert")
 
-    data_name = None
-    if merge_data:
-        data_name = os.path.basename(datasets_path)
-        if not os.path.isabs(datasets_path):
-            # data_name = datasets_path
-            datasets_path = os.path.dirname(os.path.join(os.getcwd(), datasets_path))
+    if not os.path.isabs(datasets_path):
+        # data_name = datasets_path
+        datasets_path = os.path.join(os.getcwd(), datasets_path)
+    image_data = None
+    for data_path in data_list:
+        if image_data is not None:
+            tmp_config = Config(data_format, data_path)
+            # if not isinstance(tmp_config, list):
+            #     tmp_config = [tmp_config]
+            image_data.images.extend(tmp_config.images)
         else:
-            datasets_path = os.path.dirname(datasets_path)
-        image_data = None
-        for data_path in data_list:
-            if image_data is not None:
-                tmp_config = Config(data_format, data_path)
-                # if not isinstance(tmp_config, list):
-                #     tmp_config = [tmp_config]
-                image_data.images.extend(tmp_config.images)
-            else:
-                image_data = Config(data_format, data_path)
-                
-                if not isinstance(image_data.images, list):
-                    image_data.images = [image_data.images]
+            image_data = Config(data_format, data_path)
             
-                # if not isinstance(image_data, list):
-                #     image_data = [image_data]
-        if interval > 1:
-            tmp = []
-            for i in range(0, len(image_data.images)):
-                if i % interval == 0:
-                    tmp.append(image_data.images[i])
-            image_data.images = tmp
+            if not isinstance(image_data.images, list):
+                image_data.images = [image_data.images]
+        
+            # if not isinstance(image_data, list):
+            #     image_data = [image_data]
+    if interval > 1:
+        tmp = []
+        for i in range(0, len(image_data.images)):
+            if i % interval == 0:
+                tmp.append(image_data.images[i])
+        image_data.images = tmp
 
-        image_data.to(
-                    output_path=datasets_path,
-                    save_format=PWDATA.pwmlff_npy,
-                    data_name=data_name,
-                    train_ratio = train_valid_ratio, 
-                    train_data_path="train", 
-                    valid_data_path="valid", 
-                    random=data_shuffle,
-                    seed = 2024, 
-                    retain_raw = False
-                    )
-    else:
-        for data_path in data_list:
-            image_data = Config(data_format, data_path)
-            if interval > 1:
-                tmp = []
-                for i in range(0, len(image_data.images)):
-                    if i % interval == 0:
-                        tmp.append(image_data.images[i])
-                image_data.images = tmp
-            image_data.to(
+    image_data.to(
                 output_path=datasets_path,
                 save_format=PWDATA.pwmlff_npy,
                 train_ratio = train_valid_ratio, 

{pwact-0.1.22 → pwact-0.1.24}/pwact/main.py

@@ -348,6 +348,7 @@ def main():
         gather_pwmata(sys.argv[2:])
 
     elif "to_pwdata".upper() == sys.argv[1].upper():#these function may use pwdata command
+        print("\n\nWarning! This method has been abandoned, new conversion methods refer to the pwdata documentation http://doc.lonxun.com/PWMLFF/Appendix-2/\n\n")
         to_pwdata(sys.argv[2:])
  
     elif "run".upper() == sys.argv[1].upper():

pwact-0.1.24/pwact/utils/tmp.py

@@ -0,0 +1,29 @@
+from pwdata import Config
+import os, sys, glob
+import numpy as np
+
+def count_pwdata():
+    work_dir = "/data/home/wuxingxing/datas/debugs/dengjiapei/run_iter/iter.0000/label/result_bk"
+    dir_list = glob.glob(os.path.join(work_dir, "*"))
+    res = []
+    for dir in dir_list:
+        # train
+        train_num = np.load(os.path.join(dir, "train/energies.npy")).shape[0]
+        res.append(train_num)
+
+        if os.path.exists(os.path.join(dir, "valid/energies.npy")):
+            test_num = np.load(os.path.join(dir, "valid/energies.npy")).shape[0]
+            res.append(test_num)
+            print("{} {} {}".format( dir, train_num, test_num))
+        else:
+            print("{} {}".format(dir, train_num))
+    print(np.sum(res))
+
+def count_outmlmd():
+    work_dir = "/data/home/wuxingxing/datas/debugs/dengjiapei/run_iter/iter.0000/label/scf"
+    mlmds = glob.glob(os.path.join(work_dir, "*/*/*/OUT.MLMD"))
+    print(len(mlmds))
+
+if __name__=="__main__":
+    count_pwdata()
+    count_outmlmd()

{pwact-0.1.22 → pwact-0.1.24}/pwact.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pwact
-Version: 0.1.22
+Version: 0.1.24
 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

{pwact-0.1.22 → pwact-0.1.24}/pwact.egg-info/SOURCES.txt

@@ -54,6 +54,7 @@ pwact/utils/json_operation.py
 pwact/utils/pre_al_data_util.py
 pwact/utils/process_tool.py
 pwact/utils/slurm_script.py
+pwact/utils/tmp.py
 pwact/utils/app_lib/__init__.py
 pwact/utils/app_lib/common.py
 pwact/utils/app_lib/cp2k.py

{pwact-0.1.22 → pwact-0.1.24}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r") as fh:
 
 setuptools.setup(
     name="pwact", 
-    version="0.1.22",
+    version="0.1.24",
     author="LonxunQuantum",
     author_email="lonxun@pwmat.com",
     description="PWACT is an open-source automated active learning platform based on PWMLFF for efficient data sampling.",