Skip to content

Commit 9001a86

Browse files
authored
📑 Refactor TrlParser (huggingface#2412)
* refactor parser * Only document some methods * Update imports in cli_utils.py and remove config option in utils.py * add `test_parse_args_and_arg_override_config` and remove unnecessary mocks [ci skip] * fix comment [ci skip] * fix comment [ci skip] * Extra arg in config also returned * fix docstring [ci skip] * add mock back * use `deprecate_kwarg`
1 parent f6f4265 commit 9001a86

File tree

4 files changed

+298
-68
lines changed

4 files changed

+298
-68
lines changed

docs/source/script_utils.md

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -7,3 +7,6 @@
77
## TrlParser
88

99
[[autodoc]] TrlParser
10+
- parse_args_and_config
11+
- parse_args_into_dataclasses
12+
- set_defaults_with_config

tests/test_cli_utils.py

Lines changed: 165 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,165 @@
1+
# Copyright 2024 The HuggingFace Inc. team. All rights reserved.
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# http://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.
14+
15+
import unittest
16+
from dataclasses import dataclass
17+
from unittest.mock import mock_open, patch
18+
19+
from trl import TrlParser
20+
21+
22+
@dataclass
23+
class MyDataclass:
24+
arg1: int
25+
arg2: str = "default"
26+
27+
28+
@dataclass
29+
class InvalidDataclass:
30+
config: str # This should raise an error in the TrlParser
31+
32+
33+
class TestTrlParser(unittest.TestCase):
34+
def test_init_without_config_field(self):
35+
"""Test initialization without 'config' field in the dataclasses."""
36+
parser = TrlParser(dataclass_types=[MyDataclass])
37+
self.assertIsInstance(parser, TrlParser)
38+
39+
def test_init_with_config_field(self):
40+
"""Test initialization with a 'config' field in the dataclass (should raise ValueError)."""
41+
with self.assertRaises(ValueError) as context:
42+
TrlParser(dataclass_types=[InvalidDataclass])
43+
self.assertTrue("has a field named 'config'" in str(context.exception))
44+
45+
@patch("builtins.open", mock_open(read_data="env:\n VAR1: value1\n VAR2: value2\narg1: 2"))
46+
@patch("yaml.safe_load")
47+
@patch("os.environ", new_callable=dict) # Mock os.environ as a dictionary
48+
def test_parse_args_and_config_with_valid_config(self, mock_environ, mock_yaml_load):
49+
"""Test parse_args_and_config method with valid arguments and config."""
50+
mock_yaml_load.return_value = {"env": {"VAR1": "value1", "VAR2": "value2"}, "arg1": 2}
51+
52+
parser = TrlParser(dataclass_types=[MyDataclass])
53+
54+
args = ["--arg2", "value", "--config", "config.yaml"] # don't set arg1 to test default value
55+
56+
# Simulate the config being loaded and environment variables being set
57+
result_args = parser.parse_args_and_config(args)
58+
59+
# Set the environment variables using the mock
60+
mock_environ["VAR1"] = "value1"
61+
mock_environ["VAR2"] = "value2"
62+
63+
# Ensure that the environment variables were set correctly
64+
self.assertEqual(mock_environ.get("VAR1"), "value1")
65+
self.assertEqual(mock_environ.get("VAR2"), "value2")
66+
67+
# Check the parsed arguments
68+
self.assertEqual(len(result_args), 1)
69+
self.assertIsInstance(result_args[0], MyDataclass)
70+
self.assertEqual(result_args[0].arg1, 2)
71+
self.assertEqual(result_args[0].arg2, "value")
72+
73+
@patch("builtins.open", mock_open(read_data="arg1: 2"))
74+
@patch("yaml.safe_load")
75+
def test_parse_args_and_arg_override_config(self, mock_yaml_load):
76+
"""Test parse_args_and_config method and check that arguments override the config."""
77+
mock_yaml_load.return_value = {"arg1": 2} # this arg is meant to be overridden
78+
79+
parser = TrlParser(dataclass_types=[MyDataclass])
80+
81+
args = ["--arg1", "3", "--config", "config.yaml"] # override arg1 default with 3
82+
83+
# Simulate the config being loaded and arguments being passed
84+
result_args = parser.parse_args_and_config(args)
85+
86+
# Check the parsed arguments
87+
self.assertEqual(len(result_args), 1)
88+
self.assertIsInstance(result_args[0], MyDataclass)
89+
self.assertEqual(result_args[0].arg1, 3)
90+
91+
@patch("builtins.open", mock_open(read_data="env: not_a_dict"))
92+
@patch("yaml.safe_load")
93+
def test_parse_args_and_config_with_invalid_env(self, mock_yaml_load):
94+
"""Test parse_args_and_config method when the 'env' field is not a dictionary."""
95+
mock_yaml_load.return_value = {"env": "not_a_dict"}
96+
97+
parser = TrlParser(dataclass_types=[MyDataclass])
98+
99+
args = ["--arg1", "2", "--arg2", "value", "--config", "config.yaml"]
100+
101+
with self.assertRaises(ValueError) as context:
102+
parser.parse_args_and_config(args)
103+
104+
self.assertEqual(str(context.exception), "`env` field should be a dict in the YAML file.")
105+
106+
def test_parse_args_and_config_without_config(self):
107+
"""Test parse_args_and_config without the `--config` argument."""
108+
parser = TrlParser(dataclass_types=[MyDataclass])
109+
110+
args = ["--arg1", "2", "--arg2", "value"]
111+
112+
# Simulate no config, just parse args normally
113+
result_args = parser.parse_args_and_config(args)
114+
115+
# Check that the arguments are parsed as is
116+
self.assertEqual(len(result_args), 1)
117+
self.assertIsInstance(result_args[0], MyDataclass)
118+
self.assertEqual(result_args[0].arg1, 2)
119+
self.assertEqual(result_args[0].arg2, "value")
120+
121+
def test_set_defaults_with_config(self):
122+
"""Test set_defaults_with_config updates the defaults."""
123+
parser = TrlParser(dataclass_types=[MyDataclass])
124+
125+
# Update defaults
126+
parser.set_defaults_with_config(arg1=42)
127+
128+
# Ensure the default value is updated
129+
result_args = parser.parse_args_and_config([])
130+
self.assertEqual(len(result_args), 1)
131+
self.assertIsInstance(result_args[0], MyDataclass)
132+
self.assertEqual(result_args[0].arg1, 42)
133+
134+
def test_parse_args_and_config_with_remaining_strings(self):
135+
parser = TrlParser(dataclass_types=[MyDataclass])
136+
137+
args = ["--arg1", "2", "--arg2", "value", "remaining"]
138+
139+
# Simulate no config, just parse args normally
140+
result_args = parser.parse_args_and_config(args, return_remaining_strings=True)
141+
142+
# Check that the arguments are parsed as is
143+
self.assertEqual(len(result_args), 2)
144+
self.assertIsInstance(result_args[0], MyDataclass)
145+
self.assertEqual(result_args[0].arg1, 2)
146+
self.assertEqual(result_args[0].arg2, "value")
147+
self.assertEqual(result_args[1], ["remaining"])
148+
149+
@patch("builtins.open", mock_open(read_data="remaining_string_in_config: abc"))
150+
@patch("yaml.safe_load")
151+
def test_parse_args_and_config_with_remaining_strings_in_config_and_args(self, mock_yaml_load):
152+
mock_yaml_load.return_value = {"remaining_string_in_config": "abc"}
153+
154+
parser = TrlParser(dataclass_types=[MyDataclass])
155+
156+
args = ["--arg1", "2", "--remaining_string_in_args", "def", "--config", "config.yaml"]
157+
158+
# Simulate the config being loaded and arguments being passed
159+
result_args = parser.parse_args_and_config(args, return_remaining_strings=True)
160+
161+
# Check that the arguments are parsed as is
162+
self.assertEqual(len(result_args), 2)
163+
self.assertIsInstance(result_args[0], MyDataclass)
164+
self.assertEqual(result_args[0].arg1, 2)
165+
self.assertEqual(result_args[1], ["--remaining_string_in_config", "abc", "--remaining_string_in_args", "def"])

0 commit comments

Comments
 (0)