Multi-value filter

src/parse_results.py

@@ -4,7 +4,7 @@
 from typing import List, Optional
 
 from src.metrics import Metrics
-from src.utils import parse_config_args
+from src.utils import parse_config_args, parse_config_arg
 
 
 def get_arg_parser() -> ArgumentParser:
@@ -25,9 +25,9 @@ def get_arg_parser() -> ArgumentParser:
 
 DEFAULT_COLUMNS = (
     "Setting",
-    Metrics.BATCH_SIZE,
     Metrics.INPUT_LENGTH,
     Metrics.TOKENS_SAMPLE,
+    Metrics.BATCH_SIZE,
     Metrics.THROUGHPUT_E2E,
     Metrics.LATENCY_E2E,
 )
@@ -54,19 +54,19 @@ def read_data(input_file: Path):
     return data
 
 
+def parse_key(key: Optional[str]) -> Optional[str]:
+    if key is None:
+        return key
+    return getattr(Metrics, key.upper(), key)
+
+
 def make_table(data, cols):
     from markdownTable import markdownTable
 
     data = [Metrics.format_metrics({col: x[col] for col in cols}) for x in data]
     return markdownTable(data).getMarkdown()
 
 
-def parse_key(key: Optional[str]) -> Optional[str]:
-    if key is None:
-        return key
-    return getattr(Metrics, key.upper(), key)
-
-
 def make_compare_table(data, cols, compare_value, compare_col):
     from markdownTable import markdownTable
 
@@ -104,13 +104,20 @@ def make_compare_table(data, cols, compare_value, compare_col):
 def filter_data(data, filters):
     if filters is None:
         return data
-    filters = parse_config_args(filters)
-    filters = {parse_key(key): value for key, value in filters.items()}
+
+    parsed_filters = {}
+    for filter in filters:
+        key, value = parse_config_arg(filter)
+        key = parse_key(key)
+        if key not in parsed_filters:
+            parsed_filters[key] = []
+        parsed_filters[key].append(value)
+
     filtered_data = []
     for x in data:
         filter = True
-        for key, value in filters.items():
-            filter = filter and x[key] == value
+        for key, value in parsed_filters.items():
+            filter = filter and x[key] in value
         if filter:
             filtered_data.append(x)
     return filtered_data