halib 0.2.19__tar.gz → 0.2.20__tar.gz

{halib-0.2.19 → halib-0.2.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: halib
-Version: 0.2.19
+Version: 0.2.20
 Summary: Small library for common tasks
 Author: Hoang Van Ha
 Author-email: hoangvanhauit@gmail.com
@@ -56,6 +56,9 @@ Dynamic: summary
 
 ## v0.2.x (Experiment & Core Updates)
 
+### **v0.2.20**
+- 🚀 **Improvement:**  `exp.perf.profiler` - allow to export *report dict* as csv files for further analysis
+
 ### **v0.2.19**
 - ✨ **New Feature:** Added `exp.core.param_gen` to facilitate fast generation of parameter combination sweeps (grid search) using YAML configurations.
 

{halib-0.2.19 → halib-0.2.20}/README.md

@@ -2,6 +2,9 @@
 
 ## v0.2.x (Experiment & Core Updates)
 
+### **v0.2.20**
+- 🚀 **Improvement:**  `exp.perf.profiler` - allow to export *report dict* as csv files for further analysis
+
 ### **v0.2.19**
 - ✨ **New Feature:** Added `exp.core.param_gen` to facilitate fast generation of parameter combination sweeps (grid search) using YAML configurations.
 

{halib-0.2.19 → halib-0.2.20}/halib/exp/perf/profiler.py

@@ -182,9 +182,9 @@ class zProfiler:
         │   │   │   [278091.230944486, 278091.251378469],
         │   }
         """
-        assert (
-            len(ctx_step_dict.keys()) > 0
-        ), "step_dict must have only one key (step_name) for detail."
+        assert len(ctx_step_dict.keys()) > 0, (
+            "step_dict must have only one key (step_name) for detail."
+        )
         normed_ctx_step_dict = {}
         for step_name, time_list in ctx_step_dict.items():
             if not isinstance(ctx_step_dict[step_name], list):
@@ -246,14 +246,126 @@ class zProfiler:
                 report_dict[ctx_name]["step_dict"]["summary"]["avg_time"][
                     f"avg_{step_name}"
                 ] = avg_time
-            report_dict[ctx_name]["step_dict"]["summary"][
-                "total_avg_time"
-            ] = total_avg_time
+            report_dict[ctx_name]["step_dict"]["summary"]["total_avg_time"] = (
+                total_avg_time
+            )
             report_dict[ctx_name]["step_dict"]["summary"] = dict(
                 sorted(report_dict[ctx_name]["step_dict"]["summary"].items())
             )
         return report_dict
 
+    def get_report_dataframes(self):
+        """
+        Returns two pandas DataFrames containing profiling data.
+
+        Returns:
+            tuple: (df_summary, df_detail)
+                - df_summary: Aggregated stats (Context, Step, Avg, %, Count)
+                - df_detail: Raw duration for every single iteration
+        """
+        try:
+            import pandas as pd
+        except ImportError:
+            logger.error(
+                "Pandas is required for get_pandas_dfs(). Please pip install pandas."
+            )
+            return None, None
+
+        # Get full data structure
+        data = self.get_report_dict(with_detail=True)
+
+        summary_rows = []
+        detail_rows = []
+
+        for ctx_name, ctx_data in data.items():
+            summary = ctx_data["step_dict"]["summary"]
+            detail_dict = ctx_data["step_dict"]["detail"]
+
+            # --- 1. Build Summary Data ---
+            # Iterate keys in 'avg_time' to ensure we capture all steps
+            for avg_key, avg_val in summary["avg_time"].items():
+                step_name = avg_key.replace("avg_", "")
+
+                # Get corresponding percent
+                per_key = f"per_{step_name}"
+                percent_val = summary["percent_time"].get(per_key, 0.0)
+
+                # Get sample count from detail list length
+                count = len(detail_dict.get(step_name, []))
+
+                summary_rows.append(
+                    {
+                        "context_name": ctx_name,
+                        "step_name": step_name,
+                        "avg_time_sec": avg_val,
+                        "percent_total": percent_val,
+                        "sample_count": count,
+                    }
+                )
+
+            # --- 2. Build Detail Data ---
+            for step_name, time_list in detail_dict.items():
+                # time_list format: [(idx, duration), (idx, duration)...]
+                for item in time_list:
+                    if len(item) == 2:
+                        idx, duration = item
+                        detail_rows.append(
+                            {
+                                "context_name": ctx_name,
+                                "step_name": step_name,
+                                "iteration_idx": idx,
+                                "duration_sec": duration,
+                            }
+                        )
+
+        # Create DataFrames
+        df_summary = pd.DataFrame(summary_rows)
+        df_detail = pd.DataFrame(detail_rows)
+
+        # Reorder columns for readability (optional but nice)
+        if not df_summary.empty:
+            df_summary = df_summary[
+                [
+                    "context_name",
+                    "step_name",
+                    "avg_time_sec",
+                    "percent_total",
+                    "sample_count",
+                ]
+            ]
+
+        if not df_detail.empty:
+            df_detail = df_detail[
+                ["context_name", "step_name", "iteration_idx", "duration_sec"]
+            ]
+
+        return df_summary, df_detail
+
+    def get_report_csv_files(self, outdir, tag="profiler"):
+        """
+        Exports profiling data to two CSV files:
+        1. {tag}_summary.csv: Aggregated stats (Avg time, %)
+        2. {tag}_detailed_logs.csv: Raw duration for every iteration
+
+        Args:
+            outdir (str): Directory to save files.
+            tag (str): Optional prefix for filenames.
+        """
+
+        if not os.path.exists(outdir):
+            os.makedirs(outdir)
+        tag_str = f"{tag}_" if tag else ""
+        summary_file = os.path.join(outdir, f"{tag_str}summary.csv")
+        detail_file = os.path.join(outdir, f"{tag_str}detailed_logs.csv")
+
+        df_summary, df_detail = self.get_report_dataframes()
+        if df_summary is not None:
+            df_summary.to_csv(summary_file, index=False, sep=";", encoding="utf-8")
+            logger.info(f"Saved summary CSV to: {summary_file}")
+        if df_detail is not None:
+            df_detail.to_csv(detail_file, index=False, sep=";", encoding="utf-8")
+            logger.info(f"Saved detailed logs CSV to: {detail_file}")
+
     @classmethod
     def plot_formatted_data(
         cls, profiler_data, outdir=None, file_format="png", do_show=False, tag=""
@@ -282,9 +394,13 @@ class zProfiler:
 
             assert n_steps > 0, "No steps found for context: {}".format(ctx)
             # Generate dynamic colors
-            colors = px.colors.sample_colorscale(
-                "Viridis", [i / (n_steps - 1) for i in range(n_steps)]
-            ) if n_steps > 1 else [px.colors.sample_colorscale("Viridis", [0])[0]]
+            colors = (
+                px.colors.sample_colorscale(
+                    "Viridis", [i / (n_steps - 1) for i in range(n_steps)]
+                )
+                if n_steps > 1
+                else [px.colors.sample_colorscale("Viridis", [0])[0]]
+            )
             # pprint(f'{len(colors)} colors generated for {n_steps} steps')
             color_map = dict(zip(step_names, colors))
 
@@ -301,7 +417,7 @@ class zProfiler:
                 go.Bar(
                     x=step_names,
                     y=list(avg_times.values()),
-                    text=[f"{v*1000:.2f} ms" for v in avg_times.values()],
+                    text=[f"{v * 1000:.2f} ms" for v in avg_times.values()],
                     textposition="outside",
                     marker=dict(color=[color_map[s] for s in step_names]),
                     name="",  # unified legend
@@ -345,7 +461,9 @@ class zProfiler:
             # Save figure
             if outdir is not None:
                 file_prefix = ctx if len(tag_str) == 0 else f"{tag_str}_{ctx}"
-                file_path = os.path.join(outdir, f"{file_prefix}_summary.{file_format.lower()}")
+                file_path = os.path.join(
+                    outdir, f"{file_prefix}_summary.{file_format.lower()}"
+                )
                 fig.write_image(file_path)
                 pprint(f"Saved figure to: 🔽")
                 pprint_local_path(file_path)
@@ -368,6 +486,7 @@ class zProfiler:
         return self.plot_formatted_data(
             report, outdir=outdir, file_format=file_format, do_show=do_show, tag=tag
         )
+
     def meta_info(self):
         """
         Print the structure of the profiler's time dictionary.
@@ -375,7 +494,7 @@ class zProfiler:
         """
         for ctx_name, ctx_dict in self.time_dict.items():
             with ConsoleLog(f"Context: {ctx_name}"):
-                step_names = list(ctx_dict['step_dict'].keys())
+                step_names = list(ctx_dict["step_dict"].keys())
                 for step_name in step_names:
                     pprint(f"Step: {step_name}")
 

{halib-0.2.19 → halib-0.2.20}/halib.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: halib
-Version: 0.2.19
+Version: 0.2.20
 Summary: Small library for common tasks
 Author: Hoang Van Ha
 Author-email: hoangvanhauit@gmail.com
@@ -56,6 +56,9 @@ Dynamic: summary
 
 ## v0.2.x (Experiment & Core Updates)
 
+### **v0.2.20**
+- 🚀 **Improvement:**  `exp.perf.profiler` - allow to export *report dict* as csv files for further analysis
+
 ### **v0.2.19**
 - ✨ **New Feature:** Added `exp.core.param_gen` to facilitate fast generation of parameter combination sweeps (grid search) using YAML configurations.
 

{halib-0.2.19 → halib-0.2.20}/setup.py

@@ -8,7 +8,7 @@ with open("requirements.txt", "r", encoding="utf-8") as f:
 
 setuptools.setup(
     name="halib",
-    version="0.2.19",
+    version="0.2.20",
     author="Hoang Van Ha",
     author_email="hoangvanhauit@gmail.com",
     description="Small library for common tasks",