delete all keys not in database

Author: CISC

scripts/compare-llama-bench.py

@@ -22,14 +22,14 @@
 logger = logging.getLogger("compare-llama-bench")
 
 # All llama-bench SQLite3 fields
-DB_FIELDS = [
+DB_FIELDS = {
     "build_commit", "build_number", "cpu_info",       "gpu_info",   "backends",     "model_filename",
     "model_type",   "model_size",   "model_n_params", "n_batch",    "n_ubatch",     "n_threads",
     "cpu_mask",     "cpu_strict",   "poll",           "type_k",     "type_v",       "n_gpu_layers",
     "split_mode",   "main_gpu",     "no_kv_offload",  "flash_attn", "tensor_split", "tensor_buft_overrides",
     "use_mmap",     "embeddings",   "no_op_offload",  "n_prompt",   "n_gen",        "n_depth",
     "test_time",    "avg_ns",       "stddev_ns",      "avg_ts",     "stddev_ts",
-]
+}
 
 # Properties by which to differentiate results per commit:
 KEY_PROPERTIES = [
@@ -306,10 +306,8 @@ def __init__(self, data_file: str):
                 for i, line in enumerate(fp):
                     parsed = json.loads(line)
 
-                    if "samples_ns" in parsed:
-                        del parsed["samples_ns"]
-                    if "samples_ts" in parsed:
-                        del parsed["samples_ts"]
+                    for k in parsed.keys() - DB_FIELDS:
+                        del parsed[k]
 
                     if (missing_keys := self._check_keys(parsed.keys())):
                         raise RuntimeError(f"Missing required data key(s) at line {i + 1}: {', '.join(missing_keys)}")
@@ -328,10 +326,8 @@ def __init__(self, data_files: list[str]):
                 parsed = json.load(fp)
 
                 for i, entry in enumerate(parsed):
-                    if "samples_ns" in entry:
-                        del entry["samples_ns"]
-                    if "samples_ts" in entry:
-                        del entry["samples_ts"]
+                    for k in entry.keys() - DB_FIELDS:
+                        del entry[k]
 
                     if (missing_keys := self._check_keys(entry.keys())):
                         raise RuntimeError(f"Missing required data key(s) at entry {i + 1}: {', '.join(missing_keys)}")
@@ -348,7 +344,12 @@ def __init__(self, data_files: list[str]):
         for data_file in data_files:
             with open(data_file, "r", encoding="utf-8") as fp:
                 for i, parsed in enumerate(csv.DictReader(fp)):
-                    if (missing_keys := self._check_keys(set(parsed.keys()))):
+                    keys = set(parsed.keys())
+
+                    for k in keys - DB_FIELDS:
+                        del parsed[k]
+
+                    if (missing_keys := self._check_keys(keys)):
                         raise RuntimeError(f"Missing required data key(s) at line {i + 1}: {', '.join(missing_keys)}")
 
                     # FIXME: Convert float/int columns from str!