From 82b548fa3e0cca0b3f2e99fd78d0352b471d9653 Mon Sep 17 00:00:00 2001 From: Dushyant Behl Date: Thu, 28 Nov 2024 18:13:43 +0530 Subject: [PATCH] data folders should be together Signed-off-by: Dushyant Behl --- tests/acceleration/test_acceleration_framework.py | 6 +++--- .../predefined_data_configs/__init__.py | 4 ++-- .../predefined_data_configs/apply_custom_template.yaml | 0 .../pretokenized_json_data.yaml | 0 .../tokenize_and_apply_input_masking.yaml | 0 tests/{ => artifacts}/testdata/__init__.py | 0 tests/{ => artifacts}/testdata/empty_data.json | 0 tests/{ => artifacts}/testdata/malformatted_data.json | 0 .../testdata/trainercontroller/__init__.py | 0 .../epoch-level-eval-loss-patience.yaml | 0 .../trainercontroller/epoch-level-eval-loss.yaml | 0 .../trainercontroller/epoch-level-training-loss.yaml | 0 .../testdata/trainercontroller/exposed_metrics.yaml | 0 .../incorrect_source_event_exposed_metrics.yaml | 0 .../testdata/trainercontroller/log_controller.yaml | 0 .../testdata/trainercontroller/loss_custom_metric.yaml | 0 .../trainercontroller/loss_custom_operation.yaml | 0 .../loss_custom_operation_invalid_action.yaml | 0 .../trainercontroller/loss_invalid_metric.yaml | 0 .../trainercontroller/loss_invalid_operation.yaml | 0 .../loss_invalid_operation_action.yaml | 0 .../trainercontroller/loss_invalid_trigger.yaml | 0 .../testdata/trainercontroller/loss_on_threshold.yaml | 0 .../loss_on_threshold_with_trainer_state.yaml | 0 .../trainercontroller/loss_unavailable_metric.yaml | 0 .../trainercontroller/loss_with_invalid_type_rule.yaml | 0 .../loss_with_malicious_input_rule.yaml | 0 .../trainercontroller/loss_with_malicious_os_rule.yaml | 0 .../non-decreasing-training-loss.yaml | 0 .../testdata/trainercontroller/on-save.yaml | 0 .../trainercontroller/thresholded-training-loss.yaml | 0 .../testdata/twitter_complaints_input_output.json | 0 .../testdata/twitter_complaints_input_output.jsonl | 0 .../testdata/twitter_complaints_small.json | 0 .../testdata/twitter_complaints_small.jsonl | 0 ...complaints_tokenized_with_maykeye_tinyllama_v0.json | 0 ...omplaints_tokenized_with_maykeye_tinyllama_v0.jsonl | 0 tests/build/test_launch_script.py | 2 +- tests/data/test_data_handlers.py | 2 +- tests/data/test_data_preprocessing_utils.py | 10 +++++----- tests/test_sft_trainer.py | 2 +- .../trainercontroller/test_tuning_trainercontroller.py | 2 +- tests/utils/test_tokenizer_data_utils.py | 2 +- tuning/config/configs.py | 3 +-- 44 files changed, 16 insertions(+), 17 deletions(-) rename tests/{ => artifacts}/predefined_data_configs/__init__.py (88%) rename tests/{ => artifacts}/predefined_data_configs/apply_custom_template.yaml (100%) rename tests/{ => artifacts}/predefined_data_configs/pretokenized_json_data.yaml (100%) rename tests/{ => artifacts}/predefined_data_configs/tokenize_and_apply_input_masking.yaml (100%) rename tests/{ => artifacts}/testdata/__init__.py (100%) rename tests/{ => artifacts}/testdata/empty_data.json (100%) rename tests/{ => artifacts}/testdata/malformatted_data.json (100%) rename tests/{ => artifacts}/testdata/trainercontroller/__init__.py (100%) rename tests/{ => artifacts}/testdata/trainercontroller/epoch-level-eval-loss-patience.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/epoch-level-eval-loss.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/epoch-level-training-loss.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/exposed_metrics.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/incorrect_source_event_exposed_metrics.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/log_controller.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_custom_metric.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_custom_operation.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_custom_operation_invalid_action.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_invalid_metric.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_invalid_operation.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_invalid_operation_action.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_invalid_trigger.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_on_threshold.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_on_threshold_with_trainer_state.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_unavailable_metric.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_with_invalid_type_rule.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_with_malicious_input_rule.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/loss_with_malicious_os_rule.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/non-decreasing-training-loss.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/on-save.yaml (100%) rename tests/{ => artifacts}/testdata/trainercontroller/thresholded-training-loss.yaml (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_input_output.json (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_input_output.jsonl (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_small.json (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_small.jsonl (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json (100%) rename tests/{ => artifacts}/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.jsonl (100%) diff --git a/tests/acceleration/test_acceleration_framework.py b/tests/acceleration/test_acceleration_framework.py index d6fbcadf0..d25554fe6 100644 --- a/tests/acceleration/test_acceleration_framework.py +++ b/tests/acceleration/test_acceleration_framework.py @@ -54,13 +54,13 @@ from tuning.utils.import_utils import is_fms_accelerate_available # for some reason the CI will raise an import error if we try to import -# these from tests.testdata +# these from tests.artifacts.testdata TWITTER_COMPLAINTS_JSON_FORMAT = os.path.join( - os.path.dirname(__file__), "../data/twitter_complaints_json.json" + os.path.dirname(__file__), "../artifacts/testdata/twitter_complaints_json.json" ) TWITTER_COMPLAINTS_TOKENIZED = os.path.join( os.path.dirname(__file__), - "../data/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json", + "../artifacts/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json", ) # pylint: disable=import-error diff --git a/tests/predefined_data_configs/__init__.py b/tests/artifacts/predefined_data_configs/__init__.py similarity index 88% rename from tests/predefined_data_configs/__init__.py rename to tests/artifacts/predefined_data_configs/__init__.py index 94a044dd1..ef01946a0 100644 --- a/tests/predefined_data_configs/__init__.py +++ b/tests/artifacts/predefined_data_configs/__init__.py @@ -25,6 +25,6 @@ PRETOKENIZE_JSON_DATA_YAML = os.path.join( PREDEFINED_DATA_CONFIGS, "pretokenized_json_data.yaml" ) -TOKENIZE_AND_INSTRUCTION_MASKING_YAML = os.path.join( - PREDEFINED_DATA_CONFIGS, "tokenize_and_instruction_masking.yaml" +TOKENIZE_AND_INPUT_MASKING_YAML = os.path.join( + PREDEFINED_DATA_CONFIGS, "tokenize_and_input_masking.yaml" ) diff --git a/tests/predefined_data_configs/apply_custom_template.yaml b/tests/artifacts/predefined_data_configs/apply_custom_template.yaml similarity index 100% rename from tests/predefined_data_configs/apply_custom_template.yaml rename to tests/artifacts/predefined_data_configs/apply_custom_template.yaml diff --git a/tests/predefined_data_configs/pretokenized_json_data.yaml b/tests/artifacts/predefined_data_configs/pretokenized_json_data.yaml similarity index 100% rename from tests/predefined_data_configs/pretokenized_json_data.yaml rename to tests/artifacts/predefined_data_configs/pretokenized_json_data.yaml diff --git a/tests/predefined_data_configs/tokenize_and_apply_input_masking.yaml b/tests/artifacts/predefined_data_configs/tokenize_and_apply_input_masking.yaml similarity index 100% rename from tests/predefined_data_configs/tokenize_and_apply_input_masking.yaml rename to tests/artifacts/predefined_data_configs/tokenize_and_apply_input_masking.yaml diff --git a/tests/testdata/__init__.py b/tests/artifacts/testdata/__init__.py similarity index 100% rename from tests/testdata/__init__.py rename to tests/artifacts/testdata/__init__.py diff --git a/tests/testdata/empty_data.json b/tests/artifacts/testdata/empty_data.json similarity index 100% rename from tests/testdata/empty_data.json rename to tests/artifacts/testdata/empty_data.json diff --git a/tests/testdata/malformatted_data.json b/tests/artifacts/testdata/malformatted_data.json similarity index 100% rename from tests/testdata/malformatted_data.json rename to tests/artifacts/testdata/malformatted_data.json diff --git a/tests/testdata/trainercontroller/__init__.py b/tests/artifacts/testdata/trainercontroller/__init__.py similarity index 100% rename from tests/testdata/trainercontroller/__init__.py rename to tests/artifacts/testdata/trainercontroller/__init__.py diff --git a/tests/testdata/trainercontroller/epoch-level-eval-loss-patience.yaml b/tests/artifacts/testdata/trainercontroller/epoch-level-eval-loss-patience.yaml similarity index 100% rename from tests/testdata/trainercontroller/epoch-level-eval-loss-patience.yaml rename to tests/artifacts/testdata/trainercontroller/epoch-level-eval-loss-patience.yaml diff --git a/tests/testdata/trainercontroller/epoch-level-eval-loss.yaml b/tests/artifacts/testdata/trainercontroller/epoch-level-eval-loss.yaml similarity index 100% rename from tests/testdata/trainercontroller/epoch-level-eval-loss.yaml rename to tests/artifacts/testdata/trainercontroller/epoch-level-eval-loss.yaml diff --git a/tests/testdata/trainercontroller/epoch-level-training-loss.yaml b/tests/artifacts/testdata/trainercontroller/epoch-level-training-loss.yaml similarity index 100% rename from tests/testdata/trainercontroller/epoch-level-training-loss.yaml rename to tests/artifacts/testdata/trainercontroller/epoch-level-training-loss.yaml diff --git a/tests/testdata/trainercontroller/exposed_metrics.yaml b/tests/artifacts/testdata/trainercontroller/exposed_metrics.yaml similarity index 100% rename from tests/testdata/trainercontroller/exposed_metrics.yaml rename to tests/artifacts/testdata/trainercontroller/exposed_metrics.yaml diff --git a/tests/testdata/trainercontroller/incorrect_source_event_exposed_metrics.yaml b/tests/artifacts/testdata/trainercontroller/incorrect_source_event_exposed_metrics.yaml similarity index 100% rename from tests/testdata/trainercontroller/incorrect_source_event_exposed_metrics.yaml rename to tests/artifacts/testdata/trainercontroller/incorrect_source_event_exposed_metrics.yaml diff --git a/tests/testdata/trainercontroller/log_controller.yaml b/tests/artifacts/testdata/trainercontroller/log_controller.yaml similarity index 100% rename from tests/testdata/trainercontroller/log_controller.yaml rename to tests/artifacts/testdata/trainercontroller/log_controller.yaml diff --git a/tests/testdata/trainercontroller/loss_custom_metric.yaml b/tests/artifacts/testdata/trainercontroller/loss_custom_metric.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_custom_metric.yaml rename to tests/artifacts/testdata/trainercontroller/loss_custom_metric.yaml diff --git a/tests/testdata/trainercontroller/loss_custom_operation.yaml b/tests/artifacts/testdata/trainercontroller/loss_custom_operation.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_custom_operation.yaml rename to tests/artifacts/testdata/trainercontroller/loss_custom_operation.yaml diff --git a/tests/testdata/trainercontroller/loss_custom_operation_invalid_action.yaml b/tests/artifacts/testdata/trainercontroller/loss_custom_operation_invalid_action.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_custom_operation_invalid_action.yaml rename to tests/artifacts/testdata/trainercontroller/loss_custom_operation_invalid_action.yaml diff --git a/tests/testdata/trainercontroller/loss_invalid_metric.yaml b/tests/artifacts/testdata/trainercontroller/loss_invalid_metric.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_invalid_metric.yaml rename to tests/artifacts/testdata/trainercontroller/loss_invalid_metric.yaml diff --git a/tests/testdata/trainercontroller/loss_invalid_operation.yaml b/tests/artifacts/testdata/trainercontroller/loss_invalid_operation.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_invalid_operation.yaml rename to tests/artifacts/testdata/trainercontroller/loss_invalid_operation.yaml diff --git a/tests/testdata/trainercontroller/loss_invalid_operation_action.yaml b/tests/artifacts/testdata/trainercontroller/loss_invalid_operation_action.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_invalid_operation_action.yaml rename to tests/artifacts/testdata/trainercontroller/loss_invalid_operation_action.yaml diff --git a/tests/testdata/trainercontroller/loss_invalid_trigger.yaml b/tests/artifacts/testdata/trainercontroller/loss_invalid_trigger.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_invalid_trigger.yaml rename to tests/artifacts/testdata/trainercontroller/loss_invalid_trigger.yaml diff --git a/tests/testdata/trainercontroller/loss_on_threshold.yaml b/tests/artifacts/testdata/trainercontroller/loss_on_threshold.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_on_threshold.yaml rename to tests/artifacts/testdata/trainercontroller/loss_on_threshold.yaml diff --git a/tests/testdata/trainercontroller/loss_on_threshold_with_trainer_state.yaml b/tests/artifacts/testdata/trainercontroller/loss_on_threshold_with_trainer_state.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_on_threshold_with_trainer_state.yaml rename to tests/artifacts/testdata/trainercontroller/loss_on_threshold_with_trainer_state.yaml diff --git a/tests/testdata/trainercontroller/loss_unavailable_metric.yaml b/tests/artifacts/testdata/trainercontroller/loss_unavailable_metric.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_unavailable_metric.yaml rename to tests/artifacts/testdata/trainercontroller/loss_unavailable_metric.yaml diff --git a/tests/testdata/trainercontroller/loss_with_invalid_type_rule.yaml b/tests/artifacts/testdata/trainercontroller/loss_with_invalid_type_rule.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_with_invalid_type_rule.yaml rename to tests/artifacts/testdata/trainercontroller/loss_with_invalid_type_rule.yaml diff --git a/tests/testdata/trainercontroller/loss_with_malicious_input_rule.yaml b/tests/artifacts/testdata/trainercontroller/loss_with_malicious_input_rule.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_with_malicious_input_rule.yaml rename to tests/artifacts/testdata/trainercontroller/loss_with_malicious_input_rule.yaml diff --git a/tests/testdata/trainercontroller/loss_with_malicious_os_rule.yaml b/tests/artifacts/testdata/trainercontroller/loss_with_malicious_os_rule.yaml similarity index 100% rename from tests/testdata/trainercontroller/loss_with_malicious_os_rule.yaml rename to tests/artifacts/testdata/trainercontroller/loss_with_malicious_os_rule.yaml diff --git a/tests/testdata/trainercontroller/non-decreasing-training-loss.yaml b/tests/artifacts/testdata/trainercontroller/non-decreasing-training-loss.yaml similarity index 100% rename from tests/testdata/trainercontroller/non-decreasing-training-loss.yaml rename to tests/artifacts/testdata/trainercontroller/non-decreasing-training-loss.yaml diff --git a/tests/testdata/trainercontroller/on-save.yaml b/tests/artifacts/testdata/trainercontroller/on-save.yaml similarity index 100% rename from tests/testdata/trainercontroller/on-save.yaml rename to tests/artifacts/testdata/trainercontroller/on-save.yaml diff --git a/tests/testdata/trainercontroller/thresholded-training-loss.yaml b/tests/artifacts/testdata/trainercontroller/thresholded-training-loss.yaml similarity index 100% rename from tests/testdata/trainercontroller/thresholded-training-loss.yaml rename to tests/artifacts/testdata/trainercontroller/thresholded-training-loss.yaml diff --git a/tests/testdata/twitter_complaints_input_output.json b/tests/artifacts/testdata/twitter_complaints_input_output.json similarity index 100% rename from tests/testdata/twitter_complaints_input_output.json rename to tests/artifacts/testdata/twitter_complaints_input_output.json diff --git a/tests/testdata/twitter_complaints_input_output.jsonl b/tests/artifacts/testdata/twitter_complaints_input_output.jsonl similarity index 100% rename from tests/testdata/twitter_complaints_input_output.jsonl rename to tests/artifacts/testdata/twitter_complaints_input_output.jsonl diff --git a/tests/testdata/twitter_complaints_small.json b/tests/artifacts/testdata/twitter_complaints_small.json similarity index 100% rename from tests/testdata/twitter_complaints_small.json rename to tests/artifacts/testdata/twitter_complaints_small.json diff --git a/tests/testdata/twitter_complaints_small.jsonl b/tests/artifacts/testdata/twitter_complaints_small.jsonl similarity index 100% rename from tests/testdata/twitter_complaints_small.jsonl rename to tests/artifacts/testdata/twitter_complaints_small.jsonl diff --git a/tests/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json b/tests/artifacts/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json similarity index 100% rename from tests/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json rename to tests/artifacts/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.json diff --git a/tests/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.jsonl b/tests/artifacts/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.jsonl similarity index 100% rename from tests/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.jsonl rename to tests/artifacts/testdata/twitter_complaints_tokenized_with_maykeye_tinyllama_v0.jsonl diff --git a/tests/build/test_launch_script.py b/tests/build/test_launch_script.py index d5dd31123..e331a5e9b 100644 --- a/tests/build/test_launch_script.py +++ b/tests/build/test_launch_script.py @@ -26,7 +26,7 @@ # First Party from build.accelerate_launch import main from build.utils import serialize_args, get_highest_checkpoint -from tests.testdata import TWITTER_COMPLAINTS_DATA_JSONL +from tests.artifacts.testdata import TWITTER_COMPLAINTS_DATA_JSONL from tuning.utils.error_logging import ( USER_ERROR_EXIT_CODE, INTERNAL_ERROR_EXIT_CODE, diff --git a/tests/data/test_data_handlers.py b/tests/data/test_data_handlers.py index 1b26e1242..403881676 100644 --- a/tests/data/test_data_handlers.py +++ b/tests/data/test_data_handlers.py @@ -21,7 +21,7 @@ import pytest # First Party -from tests.testdata import MODEL_NAME, TWITTER_COMPLAINTS_DATA_JSONL +from tests.artifacts.testdata import MODEL_NAME, TWITTER_COMPLAINTS_DATA_JSONL # Local from tuning.data.data_handlers import apply_custom_data_formatting_template diff --git a/tests/data/test_data_preprocessing_utils.py b/tests/data/test_data_preprocessing_utils.py index c603f7995..44181d215 100644 --- a/tests/data/test_data_preprocessing_utils.py +++ b/tests/data/test_data_preprocessing_utils.py @@ -25,12 +25,12 @@ import yaml # First Party -from tests.predefined_data_configs import ( +from tests.artifacts.predefined_data_configs import ( APPLY_CUSTOM_TEMPLATE_YAML, PRETOKENIZE_JSON_DATA_YAML, - TOKENIZE_AND_INSTRUCTION_MASKING_YAML, + TOKENIZE_AND_INPUT_MASKING_YAML, ) -from tests.testdata import ( +from tests.artifacts.testdata import ( MODEL_NAME, TWITTER_COMPLAINTS_DATA_INPUT_OUTPUT_JSON, TWITTER_COMPLAINTS_DATA_INPUT_OUTPUT_JSONL, @@ -408,11 +408,11 @@ def test_validate_args_pretokenized(data_args, packing): (PRETOKENIZE_JSON_DATA_YAML, TWITTER_COMPLAINTS_TOKENIZED_JSON), (PRETOKENIZE_JSON_DATA_YAML, TWITTER_COMPLAINTS_TOKENIZED_JSONL), ( - TOKENIZE_AND_INSTRUCTION_MASKING_YAML, + TOKENIZE_AND_INPUT_MASKING_YAML, TWITTER_COMPLAINTS_DATA_INPUT_OUTPUT_JSON, ), ( - TOKENIZE_AND_INSTRUCTION_MASKING_YAML, + TOKENIZE_AND_INPUT_MASKING_YAML, TWITTER_COMPLAINTS_DATA_INPUT_OUTPUT_JSONL, ), ], diff --git a/tests/test_sft_trainer.py b/tests/test_sft_trainer.py index a004cc7a9..69ccbf4fa 100644 --- a/tests/test_sft_trainer.py +++ b/tests/test_sft_trainer.py @@ -31,7 +31,7 @@ # First Party from build.utils import serialize_args from scripts.run_inference import TunedCausalLM -from tests.testdata import ( +from tests.artifacts.testdata import ( EMPTY_DATA, MALFORMATTED_DATA, MODEL_NAME, diff --git a/tests/trainercontroller/test_tuning_trainercontroller.py b/tests/trainercontroller/test_tuning_trainercontroller.py index bdf5aadd3..2326e8e8c 100644 --- a/tests/trainercontroller/test_tuning_trainercontroller.py +++ b/tests/trainercontroller/test_tuning_trainercontroller.py @@ -30,7 +30,7 @@ from tests.trainercontroller.custom_operation_invalid_action import ( CustomOperationInvalidAction, ) -import tests.testdata.trainercontroller as td +import tests.artifacts.testdata.trainercontroller as td # Local import tuning.config.configs as config diff --git a/tests/utils/test_tokenizer_data_utils.py b/tests/utils/test_tokenizer_data_utils.py index 2a0040eb0..e24c90099 100644 --- a/tests/utils/test_tokenizer_data_utils.py +++ b/tests/utils/test_tokenizer_data_utils.py @@ -3,7 +3,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer # First Party -from tests.testdata import MODEL_NAME +from tests.artifacts.testdata import MODEL_NAME # Local # First party diff --git a/tuning/config/configs.py b/tuning/config/configs.py index 2520fdf94..222bf4424 100644 --- a/tuning/config/configs.py +++ b/tuning/config/configs.py @@ -99,8 +99,7 @@ class DataArguments: default=None, metadata={ "help": "data config file which specifies the data preprocessing logic to apply.\ - Supports both JSON and YAML based config files.\ - for examples see examples/predefined_data_configs" + Supports both JSON and YAML based config files." }, )