from dingo.config import InputArgs
from dingo.exec import Executor
def huggingface_plaintext():
input_data = {
"input_path": "chupei/format-text",
"dataset": {
"format": "plaintext",
},
"evaluator": [
{
"evals": [
{"name": "RuleColonEnd"}
]
}
]
}
input_args = InputArgs(**input_data)
executor = Executor.exec_map["local"](input_args)
result = executor.execute()
print(result)
def huggingface_json():
input_data = {
"input_path": "chupei/format-json",
"dataset": {
"format": "json",
},
"evaluator": [
{
"fields": {"prompt": "origin_prompt", "content": "prediction"},
"evals": [
{"name": "RuleColonEnd"}
]
}
]
}
input_args = InputArgs(**input_data)
executor = Executor.exec_map["local"](input_args)
result = executor.execute()
print(result)
def huggingface_jsonl():
input_data = {
"input_path": "chupei/format-jsonl",
"dataset": {
"format": "jsonl",
},
"evaluator": [
{
"fields": {"content": "content"},
"evals": [
{"name": "RuleColonEnd"}
]
}
]
}
input_args = InputArgs(**input_data)
executor = Executor.exec_map["local"](input_args)
result = executor.execute()
print(result)
def huggingface_listjson():
input_data = {
"input_path": "chupei/format-listjson",
"dataset": {
"format": "listjson",
},
"evaluator": [
{
"fields": {"prompt": "instruction", "content": "output"},
"evals": [
{"name": "RuleColonEnd"}
]
}
]
}
input_args = InputArgs(**input_data)
executor = Executor.exec_map["local"](input_args)
result = executor.execute()
print(result)
if __name__ == '__main__':
huggingface_plaintext()
huggingface_json()
huggingface_jsonl()
huggingface_listjson()
