Trainer support simultaneously parse JSON files and cmd arguments.

llm/finetune_generation.py

@@ -58,8 +58,10 @@ def read_local_dataset(path):
 def main():
     # Arguments
     parser = PdArgumentParser((GenerateArgument, QuantArgument, ModelArgument, DataArgument, TrainingArguments))
-    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
-        gen_args, quant_args, model_args, data_args, training_args = parser.parse_json_file(
+    # Support format as "args.json --args1 value1 --args2 value2.”
+    # In case of conflict, command line arguments take precedence.
+    if len(sys.argv) >= 2 and sys.argv[1].endswith(".json"):
+        gen_args, quant_args, model_args, data_args, training_args = parser.parse_json_file_and_cmd_lines(
             json_file=os.path.abspath(sys.argv[1])
         )
     else:

llm/run_pretrain.py

@@ -372,8 +372,12 @@ def _get_train_sampler(self) -> Optional[paddle.io.Sampler]:
 
 def main():
     parser = PdArgumentParser((ModelArguments, DataArguments, PreTrainingArguments))
-    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
-        model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    # Support format as "args.json --args1 value1 --args2 value2.”
+    # In case of conflict, command line arguments take precedence.
+    if len(sys.argv) >= 2 and sys.argv[1].endswith(".json"):
+        model_args, data_args, training_args = parser.parse_json_file_and_cmd_lines(
+            json_file=os.path.abspath(sys.argv[1])
+        )
     else:
         model_args, data_args, training_args = parser.parse_args_into_dataclasses()
 

paddlenlp/trainer/argparser.py

@@ -247,6 +247,52 @@ def parse_json_file(self, json_file: str) -> Tuple[DataClass, ...]:
             outputs.append(obj)
         return (*outputs,)
 
+    def parse_json_file_and_cmd_lines(self, json_file: str) -> Tuple[DataClass, ...]:
+        """
+        Extend the functionality of `parse_json_file` to handle command line arguments in addition to loading a JSON
+        file.
+
+        When there is a conflict between the command line arguments and the JSON file configuration,
+        the command line arguments will take precedence.
+
+        This method combines data from a JSON file and command line arguments to populate instances of dataclasses.
+
+        Args:
+            json_file :
+                The path to the JSON formatted file should be at index position 1 in the command line
+                arguments array (sys.argv[1]).
+                Any JSON file path at other positions will be considered invalid.
+
+        Returns:
+            Tuple consisting of:
+
+                - the dataclass instances in the same order as they were passed to the initializer.abspath
+        """
+        json_args = json.loads(Path(json_file).read_text())
+        del sys.argv[1]
+        output_dir_arg = next(
+            (arg for arg in sys.argv if arg == "--output_dir" or arg.startswith("--output_dir=")), None
+        )
+        if output_dir_arg is None:
+            if "output_dir" in json_args.keys():
+                sys.argv.extend(["--output_dir", json_args["output_dir"]])
+            else:
+                raise ValueError("The following arguments are required: --output_dir")
+        cmd_args = vars(self.parse_args())
+        merged_args = {}
+        for key in json_args.keys() | cmd_args.keys():
+            if any(arg == f"--{key}" or arg.startswith(f"--{key}=") for arg in sys.argv):
+                merged_args[key] = cmd_args.get(key)
+            elif json_args.get(key):
+                merged_args[key] = json_args.get(key)
+        outputs = []
+        for dtype in self.dataclass_types:
+            keys = {f.name for f in dataclasses.fields(dtype) if f.init}
+            inputs = {k: v for k, v in merged_args.items() if k in keys}
+            obj = dtype(**inputs)
+            outputs.append(obj)
+        return (*outputs,)
+
     def parse_dict(self, args: dict) -> Tuple[DataClass, ...]:
         """
         Alternative helper method that does not use `argparse` at all, instead uses a dict and populating the dataclass

tests/trainer/test_argparser.py

@@ -0,0 +1,125 @@
+# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import json
+import os
+import sys
+import tempfile
+import unittest
+from unittest.mock import patch
+
+from llm.run_pretrain import PreTrainingArguments
+from paddlenlp.trainer.argparser import PdArgumentParser
+
+
+def parse_args():
+    parser = PdArgumentParser((PreTrainingArguments,))
+    # Support format as "args.json --args1 value1 --args2 value2.”
+    # In case of conflict, command line arguments take precedence.
+    if len(sys.argv) >= 2 and sys.argv[1].endswith(".json"):
+        model_args = parser.parse_json_file_and_cmd_lines(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args = parser.parse_args_into_dataclasses()
+    return model_args
+
+
+def create_json_from_dict(data_dict, file_path):
+    with open(file_path, "w") as f:
+        json.dump(data_dict, f)
+
+
+class ArgparserTest(unittest.TestCase):
+    script_name = "test_argparser.py"
+    args_dict = {
+        "max_steps": 3000,
+        "amp_master_grad": False,
+        "adam_beta1": 0.9,
+        "adam_beta2": 0.999,
+        "adam_epsilon": 1e-08,
+        "bf16": False,
+        "enable_linear_fused_grad_add": False,
+        "eval_steps": 3216,
+        "flatten_param_grads": False,
+        "fp16": 1,
+        "log_on_each_node": True,
+        "logging_dir": "./checkpoints/llama2_pretrain_ckpts/runs/Dec27_04-28-35_instance-047hzlt0-4",
+        "logging_first_step": False,
+        "logging_steps": 1,
+        "lr_end": 1e-07,
+        "max_evaluate_steps": -1,
+        "max_grad_norm": 1.0,
+        "min_learning_rate": 3e-06,
+        "no_cuda": False,
+        "num_cycles": 0.5,
+        "num_train_epochs": 3.0,
+        "output_dir": "./checkpoints/llama2_pretrain_ckpts",
+    }
+
+    def test_parse_cmd_lines(self):
+        cmd_line_args = [ArgparserTest.script_name]
+        for key, value in ArgparserTest.args_dict.items():
+            cmd_line_args.extend([f"--{key}", str(value)])
+        with patch("sys.argv", cmd_line_args):
+            model_args = vars(parse_args()[0])
+        for key, value in ArgparserTest.args_dict.items():
+            self.assertEqual(model_args.get(key), value)
+
+    def test_parse_json_file(self):
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as tmpfile:
+            create_json_from_dict(ArgparserTest.args_dict, tmpfile.name)
+            tmpfile_path = tmpfile.name
+        with patch("sys.argv", [ArgparserTest.script_name, tmpfile_path]):
+            model_args = vars(parse_args()[0])
+        for key, value in ArgparserTest.args_dict.items():
+            self.assertEqual(model_args.get(key), value)
+        os.remove(tmpfile_path)
+
+    def test_parse_json_file_and_cmd_lines(self):
+        half_size = len(ArgparserTest.args_dict) // 2
+        json_part = {k: ArgparserTest.args_dict[k] for k in list(ArgparserTest.args_dict)[:half_size]}
+        cmd_line_part = {k: ArgparserTest.args_dict[k] for k in list(ArgparserTest.args_dict)[half_size:]}
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as tmpfile:
+            create_json_from_dict(json_part, tmpfile.name)
+            tmpfile_path = tmpfile.name
+        cmd_line_args = [ArgparserTest.script_name, tmpfile_path]
+        for key, value in cmd_line_part.items():
+            cmd_line_args.extend([f"--{key}", str(value)])
+        with patch("sys.argv", cmd_line_args):
+            model_args = vars(parse_args()[0])
+        for key, value in ArgparserTest.args_dict.items():
+            self.assertEqual(model_args.get(key), value)
+        os.remove(tmpfile_path)
+
+    def test_parse_json_file_and_cmd_lines_with_conflict(self):
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as tmpfile:
+            json.dump(ArgparserTest.args_dict, tmpfile)
+            tmpfile_path = tmpfile.name
+        cmd_line_args = [
+            ArgparserTest.script_name,
+            tmpfile_path,
+            "--min_learning_rate",
+            "2e-5",
+            "--max_steps",
+            "3000",
+            "--log_on_each_node",
+            "False",
+        ]
+        with patch("sys.argv", cmd_line_args):
+            model_args = vars(parse_args()[0])
+        self.assertEqual(model_args.get("min_learning_rate"), 2e-5)
+        self.assertEqual(model_args.get("max_steps"), 3000)
+        self.assertEqual(model_args.get("log_on_each_node"), False)
+        for key, value in ArgparserTest.args_dict.items():
+            if key not in ["min_learning_rate", "max_steps", "log_on_each_node"]:
+                self.assertEqual(model_args.get(key), value)
+        os.remove(tmpfile_path)