fix: created module level doc string.

.github/scripts/mr_generate_summary.py

@@ -1,11 +1,12 @@
-# Collect the results of the various model test runs which are done as part of
-# the model regression CI pipeline and dump them as a single file artifact.
-# This artifact will the then be published at the end of the tests.
-from collections import defaultdict
-import json
+"""Collect the results of the various model test runs which are done as part of
+the model regression CI pipeline and dump them as a single file artifact.
+This artifact will the then be published at the end of the tests."""
 import os
-from pathlib import Path
+import json
+
 from typing import Dict, List
+from collections import defaultdict
+from pathlib import Path
 
 
 def combine_result(

.github/scripts/mr_publish_results.py

@@ -5,6 +5,7 @@
 import datetime
 import json
 import os
+import re
 from typing import Any, Dict, List, Text, Tuple
 
 from datadog_api_client.v1 import ApiClient, Configuration
@@ -94,27 +95,21 @@ def transform_to_seconds(duration: Text) -> float:
         duration: Examples: '1m27s', '1m27.3s', '27s', '1h27s', '1h1m27s'
 
     Raises:
-        Exception: If the input is not supported.
+        Exception: If the value are not matched.
 
     Returns:
         Duration converted in seconds.
     """
-    h_split = duration.split("h")
-    if len(h_split) == 1:
-        rest = h_split[0]
-        hours = 0
-    else:
-        hours = int(h_split[0])
-        rest = h_split[1]
-    m_split = rest.split("m")
-    if len(m_split) == 2:
-        minutes = int(m_split[0])
-        seconds = float(m_split[1].rstrip("s"))
-    elif len(m_split) == 1:
-        minutes = 0
-        seconds = float(m_split[0].rstrip("s"))
-    else:
-        raise Exception(f"Unsupported duration: {duration}")
+    time_pattern = r'(?:(?P<hours>\d+)h)?(?:(?P<minutes>\d+)m)?(?:(?P<seconds>\d+(?:\.\d+)?)s)?'
+    match = re.match(time_pattern, duration)
+
+    if not match:
+        raise ValueError(f"Invalid duration format: {duration}")
+
+    hours = float(match.group("hours") or 0)
+    minutes = float(match.group("minuts") or 0)
+    seconds = float(match.group("seconds") or 0)
+
     overall_seconds = hours * 60 * 60 + minutes * 60 + seconds
     return overall_seconds
 
@@ -277,7 +272,7 @@ def generate_json(file: Text, task: Text, data: dict) -> dict:
 
 def create_report_file() -> None:
     data = {}
-    for dirpath, dirnames, files in os.walk(os.environ["RESULT_DIR"]):
+    for dirpath, _, files in os.walk(os.environ["RESULT_DIR"]):
         for f in files:
             if f not in TASK_MAPPING.keys():
                 continue

.github/tests/test_download_pretrained.py

@@ -18,14 +18,19 @@ def test_download_pretrained_lmf_exists_no_params():
     assert lmf_specs[0].model_weights == "rasa/LaBSE"
 
 
-def test_download_pretrained_lmf_exists_with_model_name():
+def _common_step_cofig():
     yaml = YAML(typ="safe")
     config = yaml.load(CONFIG_FPATH)
 
     steps = config.get("pipeline", [])
     step = list(  # noqa: RUF015
         filter(lambda x: x["name"] == download_pretrained.COMP_NAME, steps)
     )[0]
+    return step, steps, yaml, config
+
+
+def test_download_pretrained_lmf_exists_with_model_name():
+    step, _, yaml, config = _common_step_cofig()
     step["model_name"] = "roberta"
     step["cache_dir"] = "/this/dir"
 
@@ -39,13 +44,7 @@ def test_download_pretrained_lmf_exists_with_model_name():
 
 
 def test_download_pretrained_unknown_model_name():
-    yaml = YAML(typ="safe")
-    config = yaml.load(CONFIG_FPATH)
-
-    steps = config.get("pipeline", [])
-    step = list(  # noqa: RUF015
-        filter(lambda x: x["name"] == download_pretrained.COMP_NAME, steps)
-    )[0]
+    step, _, yaml, config = _common_step_cofig()
     step["model_name"] = "unknown"
 
     with tempfile.NamedTemporaryFile("w+") as fp:
@@ -56,13 +55,7 @@ def test_download_pretrained_unknown_model_name():
 
 
 def test_download_pretrained_multiple_model_names():
-    yaml = YAML(typ="safe")
-    config = yaml.load(CONFIG_FPATH)
-
-    steps = config.get("pipeline", [])
-    step = list(  # noqa: RUF015
-        filter(lambda x: x["name"] == download_pretrained.COMP_NAME, steps)
-    )[0]
+    step, steps, yaml, config = _common_step_cofig()
     step_new = deepcopy(step)
     step_new["model_name"] = "roberta"
     steps.append(step_new)
@@ -76,13 +69,7 @@ def test_download_pretrained_multiple_model_names():
 
 
 def test_download_pretrained_with_model_name_and_nondefault_weight():
-    yaml = YAML(typ="safe")
-    config = yaml.load(CONFIG_FPATH)
-
-    steps = config.get("pipeline", [])
-    step = list(  # noqa: RUF015
-        filter(lambda x: x["name"] == download_pretrained.COMP_NAME, steps)
-    )[0]
+    step, _, yaml, config = _common_step_cofig()
     step["model_name"] = "bert"
     step["model_weights"] = "bert-base-uncased"
 
@@ -95,13 +82,7 @@ def test_download_pretrained_with_model_name_and_nondefault_weight():
 
 
 def test_download_pretrained_lmf_doesnt_exists():
-    yaml = YAML(typ="safe")
-    config = yaml.load(CONFIG_FPATH)
-
-    steps = config.get("pipeline", [])
-    step = list(  # noqa: RUF015
-        filter(lambda x: x["name"] == download_pretrained.COMP_NAME, steps)
-    )[0]
+    step, steps, yaml, config = _common_step_cofig()
     steps.remove(step)
 
     with tempfile.NamedTemporaryFile("w+") as fp: