halib 0.1.57__tar.gz → 0.1.59__tar.gz

{halib-0.1.57 → halib-0.1.59}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: halib
-Version: 0.1.57
+Version: 0.1.59
 Summary: Small library for common tasks
 Author: Hoang Van Ha
 Author-email: hoangvanhauit@gmail.com
@@ -15,7 +15,7 @@ License-File: LICENSE.txt
 
 Helper package for coding and automation
 
-**Version 0.1.57**
+**Version 0.1.59**
 
 + add `util/perfcalc`: abstract class for performance calculation. This class need to be inherited and implemented with specific performance calculation logic.
 

{halib-0.1.57 → halib-0.1.59}/README.md

@@ -1,6 +1,6 @@
 Helper package for coding and automation
 
-**Version 0.1.57**
+**Version 0.1.59**
 
 + add `util/perfcalc`: abstract class for performance calculation. This class need to be inherited and implemented with specific performance calculation logic.
 

{halib-0.1.57 → halib-0.1.59}/halib/research/perfcalc.py

@@ -59,7 +59,7 @@ def valid_custom_fields(fn):
     return wrapper
 
 REQUIRED_COLS = ["experiment", "dataset"]
-CSV_FILE_POSTFIX = "__perf.csv"
+CSV_FILE_POSTFIX = "__perf"
 
 class PerfCalc(ABC): # Abstract base class for performance calculation
     @abstractmethod
@@ -124,6 +124,7 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
         all_metric_data = self.prepare_exp_data_for_metrics(
             metric_names, *args, **kwargs
         )
+        metric_col_names = []
         for metric in metric_names:
             if metric not in all_metric_data:
                 raise ValueError(f"Metric '{metric}' not found in provided data.")
@@ -140,8 +141,12 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
                 args = metric_data
             else:
                 raise TypeError(f"Unsupported data format for metric '{metric}'")
+
             # Call update and compute
-            tmetric.update(*args)
+            if len(expected_args) == 1:
+                tmetric.update(args)  # pass as single argument
+            else:
+                tmetric.update(*args)    # unpack multiple arguments
             computed_value = tmetric.compute()
             # ensure the computed value converted to a scala value or list array
             if isinstance(computed_value, torch.Tensor):
@@ -149,16 +154,16 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
                     computed_value = computed_value.item()
                 else:
                     computed_value = computed_value.tolist()
-            out_dict[metric] = computed_value
+            col_name = f"metric_{metric}" if 'metric_' not in metric else metric
+            metric_col_names.append(col_name)
+            out_dict[col_name] = computed_value
 
         # check if any kwargs named "outfile"
         csv_outfile = kwargs.get("outfile", None)
         if csv_outfile is not None:
-            # get the file path without the extension
             filePathNoExt, _ = os.path.splitext(csv_outfile)
-            # add the postfix to the file path
-            csvfilename = f"{now_str()}_{filePathNoExt}{CSV_FILE_POSTFIX}.csv"
-            csv_outfile = os.path.join(os.path.dirname(csv_outfile), csvfilename)
+            # pprint(f"CSV Outfile Path (No Ext): {filePathNoExt}")
+            csv_outfile = f'{filePathNoExt}{CSV_FILE_POSTFIX}.csv'
         elif "outdir" in kwargs:
             csvoutdir = kwargs["outdir"]
             csvfilename = f"{now_str()}_{self.get_dataset_name()}_{self.get_experiment_name()}_{CSV_FILE_POSTFIX}.csv"
@@ -166,7 +171,7 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
 
         # convert out_dict to a DataFrame
         df = pd.DataFrame([out_dict])
-        ordered_cols = REQUIRED_COLS + custom_fields + metric_names
+        ordered_cols = REQUIRED_COLS + custom_fields + metric_col_names
         df = df[ordered_cols]  # reorder columns
 
         if csv_outfile:
@@ -208,7 +213,7 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
                         temp_df[col] = None  # fill missing columns with None
                 df = pd.concat([df, temp_df], ignore_index=True)
             # assert that REQUIRED_COLS are present in the DataFrame
-            pprint(df.columns.tolist())
+            # pprint(df.columns.tolist())
             for col in REQUIRED_COLS:
                 if col not in df.columns:
                     raise ValueError(f"Required column '{col}' is missing from the DataFrame. REQUIRED_COLS = {REQUIRED_COLS}")
@@ -275,21 +280,36 @@ class PerfCalc(ABC): # Abstract base class for performance calculation
 
         assert os.path.exists(indir), f"Input directory {indir} does not exist."
 
+        csv_perf_files = []
         # Find experiment subdirectories
         exp_dirs = [
             os.path.join(indir, d)
             for d in os.listdir(indir)
             if os.path.isdir(os.path.join(indir, d))
         ]
-        assert exp_dirs, f"No experiment directories found in {indir}."
+        if len(exp_dirs) == 0:
+            csv_perf_files = glob.glob(
+                os.path.join(indir, f"*{exp_perf_csv_pattern}*.csv")
+            )
+            csv_perf_files = [
+                file_item
+                for file_item in csv_perf_files
+                if exp_perf_csv_pattern in file_item
+            ]
+        else:
+            # multiple experiment directories found
+            # Collect all matching CSV files in those subdirs
+            for exp_dir in exp_dirs:
+                # pprint(f"Searching in experiment directory: {exp_dir}")
+                matched = glob.glob(
+                    os.path.join(exp_dir, f"*{exp_perf_csv_pattern}*.csv")
+                )
+                csv_perf_files.extend(matched)
 
-        # Collect all matching CSV files in those subdirs
-        csv_perf_files = []
-        for exp_dir in exp_dirs:
-            pprint(f"Searching in experiment directory: {exp_dir}")
-            matched = glob.glob(os.path.join(exp_dir, f"*{exp_perf_csv_pattern}*.csv"))
+        assert (
+            len(csv_perf_files) > 0
+        ), f"No CSV files matching pattern '{exp_perf_csv_pattern}' found in the experiment directories."
 
-            csv_perf_files.extend(matched)
         assert len(csv_perf_files) > 0, f"No CSV files matching pattern '{exp_perf_csv_pattern}' found in the experiment directories."
 
         all_exp_perf_df = get_df_for_all_exp_perf(csv_perf_files, csv_sep=csv_sep)

halib-0.1.59/halib/utils/dict_op.py

@@ -0,0 +1,9 @@
+def flatten_dict(d, parent_key="", sep="."):
+    items = {}
+    for k, v in d.items():
+        key = f"{parent_key}{sep}{k}" if parent_key else k
+        if isinstance(v, dict):
+            items.update(flatten_dict(v, key, sep=sep))
+        else:
+            items[key] = v
+    return items

{halib-0.1.57 → halib-0.1.59}/halib.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: halib
-Version: 0.1.57
+Version: 0.1.59
 Summary: Small library for common tasks
 Author: Hoang Van Ha
 Author-email: hoangvanhauit@gmail.com
@@ -15,7 +15,7 @@ License-File: LICENSE.txt
 
 Helper package for coding and automation
 
-**Version 0.1.57**
+**Version 0.1.59**
 
 + add `util/perfcalc`: abstract class for performance calculation. This class need to be inherited and implemented with specific performance calculation logic.
 

{halib-0.1.57 → halib-0.1.59}/halib.egg-info/SOURCES.txt

@@ -37,5 +37,6 @@ halib/system/cmd.py
 halib/system/filesys.py
 halib/utils/__init__.py
 halib/utils/dataclass_util.py
+halib/utils/dict_op.py
 halib/utils/listop.py
 halib/utils/tele_noti.py

{halib-0.1.57 → halib-0.1.59}/setup.py

@@ -8,7 +8,7 @@ with open("requirements.txt") as f:
 
 setuptools.setup(
     name="halib",
-    version="0.1.57",
+    version="0.1.59",
     author="Hoang Van Ha",
     author_email="hoangvanhauit@gmail.com",
     description="Small library for common tasks",