Switch RAFT dataset source from ought/raft to regisss/raft for compatibility with datasets>=4.0.0 (#2310)

Author: Grzegorz Pluto-Prondzinski

examples/language-modeling/README.md

@@ -530,7 +530,7 @@ The format of the text files (with extensions .text or .txt) is expected to be
 
 To run prompt tuning finetuning, you can use `run_prompt_tuning_clm.py`.
 Here are single-card command examples for Llama2-7B:
-- single-card finetuning of meta-llama/Llama-2-7b-hf with dataset "ought/raft" and config "twitter_complaints":
+- single-card finetuning of meta-llama/Llama-2-7b-hf with dataset "regisss/raft" and config "default":
 ```bash
 PT_HPU_LAZY_MODE=1 python3 run_prompt_tuning_clm.py \
     --model_name_or_path meta-llama/Llama-2-7b-hf \

examples/language-modeling/run_prompt_tuning_clm.py

@@ -172,10 +172,10 @@ class DataTrainingArguments:
     """
 
     dataset_name: Optional[str] = field(
-        default="ought/raft", metadata={"help": "The name of the dataset to use (via the datasets library)."}
+        default="regisss/raft", metadata={"help": "The name of the dataset to use (via the datasets library)."}
     )
     dataset_config_name: Optional[str] = field(
-        default="twitter_complaints",
+        default="default",
         metadata={"help": "The configuration name of the dataset to use (via the datasets library)."},
     )
     max_eval_samples: Optional[int] = field(
@@ -250,11 +250,11 @@ def main():
         streaming=data_args.streaming,
         trust_remote_code=model_args.trust_remote_code,
     )
-    if data_args.dataset_name == "ought/raft" and data_args.dataset_config_name == "twitter_complaints":
-        text_column = "Tweet text"
+    if data_args.dataset_name == "regisss/raft" and data_args.dataset_config_name == "default":
+        text_column = "Abstract Note"
         label_column = "text_label"
     else:
-        raise ValueError("preprocess is only for ought/raft twitter_complaints now")
+        raise ValueError("preprocess is only for regisss/raft default now")
     classes = [k.replace("_", " ") for k in dataset["train"].features["Label"].names]
     dataset = dataset.map(
         lambda x: {"text_label": [classes[label] for label in x["Label"]]},

tests/test_examples.py

@@ -1085,21 +1085,21 @@ class MultiCardCausalLanguageModelingPromptTuningExampleTester(
     ExampleTesterBase, metaclass=ExampleTestMeta, example_name="run_prompt_tuning_clm", multi_card=True
 ):
     TASK_NAME = "prompt-tuning"
-    DATASET_NAME = "ought/raft"
+    DATASET_NAME = "regisss/raft"
 
 
 class MultiCardCausalLanguageModelingPrefixTuningExampleTester(
     ExampleTesterBase, metaclass=ExampleTestMeta, example_name="run_prompt_tuning_clm", multi_card=True
 ):
     TASK_NAME = "prefix-tuning"
-    DATASET_NAME = "ought/raft"
+    DATASET_NAME = "regisss/raft"
 
 
 class MultiCardCausalLanguageModelingPTuningExampleTester(
     ExampleTesterBase, metaclass=ExampleTestMeta, example_name="run_prompt_tuning_clm", multi_card=True
 ):
     TASK_NAME = "p-tuning"
-    DATASET_NAME = "ought/raft"
+    DATASET_NAME = "regisss/raft"
 
 
 class MultiCardMultiTastPromptPeftExampleTester(