Ignore tag order while checking their occurrences (strict_tags) (#36)

* ignore tag order * keep the event file sort since input_dirs could be already sorted * remove sorted() when setting all_dirs_tags_list (now named tags_in_each_dir) refactor missing_tags_report * update pre-commit hooks --------- Co-authored-by: HeinrichAD <[email protected]> Co-authored-by: Janosh Riebesell <[email protected]>
janosh · Sep 21, 2023 · cca47fa · cca47fa
1 parent 1ace02f
commit cca47fa
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 14 deletions.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -7,18 +7,18 @@ default_install_hook_types: [pre-commit, commit-msg]
 
 repos:
   - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: v0.0.284
+    rev: v0.0.290
     hooks:
       - id: ruff
         args: [--fix]
 
   - repo: https://github.com/psf/black
-    rev: 23.7.0
+    rev: 23.9.1
     hooks:
       - id: black-jupyter
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.5.0
+    rev: v1.5.1
     hooks:
       - id: mypy
 

diff --git a/tensorboard_reducer/load.py b/tensorboard_reducer/load.py
@@ -71,22 +71,21 @@ def load_tb_events(
     # strict_tags=False.
     if strict_tags:
         # generate list of scalar tags for all event files each in alphabetical order
-        all_dirs_tags_list = sorted(
-            accumulator.scalar_tags for accumulator in accumulators
-        )
-        first_tags = all_dirs_tags_list[0]
-
-        all_runs_same_tags = all(first_tags == tags for tags in all_dirs_tags_list)
+        tags_in_each_dir = [
+            set(accumulator.scalar_tags) for accumulator in accumulators
+        ]
 
-        tags_set = {tag for tags in all_dirs_tags_list for tag in tags}
+        all_tags = {tag for tags in tags_in_each_dir for tag in tags}
 
+        # generate report of missing tags for each run directory
+        # will be empty string if no tags are missing
         missing_tags_report = "".join(
-            f"- {in_dir} missing tags: {', '.join(tags_set - {*tags})}\n"
-            for in_dir, tags in zip(input_dirs, all_dirs_tags_list)
-            if len(tags_set - {*tags}) > 0
+            f"- {in_dir} missing tags: {', '.join(all_tags - run_tags)}\n"
+            for in_dir, run_tags in zip(input_dirs, tags_in_each_dir)
+            if len(all_tags - run_tags) > 0
         )
 
-        if not all_runs_same_tags:
+        if missing_tags_report:
             raise ValueError(
                 f"Some tags are in some logs but not others:\n{missing_tags_report}"
                 "\nIf intentional, pass CLI flag --lax-tags or strict_tags=False "