fix: incorrect accuracy setup calculation (#85)

Author: lukaszkolodziejczyk

mostlyai/qa/_sampling.py

@@ -110,15 +110,16 @@ def pull_data_for_accuracy(
     df = pd.merge(df, df_tgt, on=key, how="left")
     df = pd.merge(df, df_nxt, on=key, how="left")
     df = df.drop(columns=[key])
-
-    # remove records with sequence length equal to 0
     count_column = f"{TGT_COLUMN_PREFIX}{COUNT_COLUMN}"
     df[count_column] = df[count_column].fillna(0).astype("Int64")
-    df = df.loc[df[count_column] > 0].reset_index(drop=True)
 
+    # determine setup if not provided
     if setup is None:
         setup = "1:1" if (df[count_column] == 1).all() else "1:N"
 
+    # remove records with sequence length equal to 0
+    df = df.loc[df[count_column] > 0].reset_index(drop=True)
+
     # for 1:1 ctx/tgt setups, drop nxt and count columns; ensure at least one column remains
     if setup == "1:1":
         df = df.drop(columns=[c for c in df.columns if c.startswith(NXT_COLUMN_PREFIX)])

mostlyai/qa/reporting_from_statistics.py

@@ -113,6 +113,9 @@ def report_from_statistics(
             ctx_primary_key=ctx_primary_key,
             tgt_context_key=tgt_context_key,
             max_sample_size=max_sample_size_accuracy,
+            # always pull Sequence Length and nxt columns for synthetic data
+            # and let downstream functions decide if they are needed
+            setup="1:N",
         )
         _LOG.info(f"sample synthetic data finished ({syn.shape=})")
         progress.update(completed=20, total=100)