AutoRAG 0.2.17__tar.gz → 0.2.18__tar.gz

{autorag-0.2.17 → autorag-0.2.18}/AutoRAG.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: AutoRAG
-Version: 0.2.17
+Version: 0.2.18
 Summary: Automatically Evaluate RAG pipelines with your own data. Find optimal structure for new RAG product.
 Author-email: Marker-Inc <vkehfdl1@gmail.com>
 License:                                  Apache License
@@ -254,14 +254,14 @@ Requires-Dist: llama-index-core>=0.11.0
 Requires-Dist: llama-index-readers-file
 Requires-Dist: llama-index-embeddings-openai
 Requires-Dist: llama-index-embeddings-huggingface
-Requires-Dist: llama-index-llms-openai>=0.1.26
+Requires-Dist: llama-index-llms-openai>=0.2.7
 Requires-Dist: llama-index-llms-huggingface
 Requires-Dist: llama-index-llms-openai-like
 Requires-Dist: llama-index-llms-ollama
 Requires-Dist: llama-index-retrievers-bm25
 Requires-Dist: streamlit
 Requires-Dist: gradio
-Requires-Dist: langchain-core>=0.1.6
+Requires-Dist: langchain-core>=0.3.0
 Requires-Dist: langchain_unstructured
 Requires-Dist: langchain-upstage
 Requires-Dist: panel

{autorag-0.2.17 → autorag-0.2.18}/AutoRAG.egg-info/SOURCES.txt

@@ -201,10 +201,12 @@ docs/source/_static/roadmap/cycle.png
 docs/source/_static/roadmap/merger.png
 docs/source/_static/roadmap/node_line_modular.png
 docs/source/_static/roadmap/policy.png
+docs/source/api_spec/autorag.data.beta.filter.rst
 docs/source/api_spec/autorag.data.beta.generation_gt.rst
 docs/source/api_spec/autorag.data.beta.query.rst
 docs/source/api_spec/autorag.data.beta.rst
 docs/source/api_spec/autorag.data.beta.schema.rst
+docs/source/api_spec/autorag.data.chunk.rst
 docs/source/api_spec/autorag.data.corpus.rst
 docs/source/api_spec/autorag.data.parse.rst
 docs/source/api_spec/autorag.data.qacreation.rst

{autorag-0.2.17 → autorag-0.2.18}/AutoRAG.egg-info/requires.txt

@@ -32,14 +32,14 @@ llama-index-core>=0.11.0
 llama-index-readers-file
 llama-index-embeddings-openai
 llama-index-embeddings-huggingface
-llama-index-llms-openai>=0.1.26
+llama-index-llms-openai>=0.2.7
 llama-index-llms-huggingface
 llama-index-llms-openai-like
 llama-index-llms-ollama
 llama-index-retrievers-bm25
 streamlit
 gradio
-langchain-core>=0.1.6
+langchain-core>=0.3.0
 langchain_unstructured
 langchain-upstage
 panel

{autorag-0.2.17 → autorag-0.2.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: AutoRAG
-Version: 0.2.17
+Version: 0.2.18
 Summary: Automatically Evaluate RAG pipelines with your own data. Find optimal structure for new RAG product.
 Author-email: Marker-Inc <vkehfdl1@gmail.com>
 License:                                  Apache License
@@ -254,14 +254,14 @@ Requires-Dist: llama-index-core>=0.11.0
 Requires-Dist: llama-index-readers-file
 Requires-Dist: llama-index-embeddings-openai
 Requires-Dist: llama-index-embeddings-huggingface
-Requires-Dist: llama-index-llms-openai>=0.1.26
+Requires-Dist: llama-index-llms-openai>=0.2.7
 Requires-Dist: llama-index-llms-huggingface
 Requires-Dist: llama-index-llms-openai-like
 Requires-Dist: llama-index-llms-ollama
 Requires-Dist: llama-index-retrievers-bm25
 Requires-Dist: streamlit
 Requires-Dist: gradio
-Requires-Dist: langchain-core>=0.1.6
+Requires-Dist: langchain-core>=0.3.0
 Requires-Dist: langchain_unstructured
 Requires-Dist: langchain-upstage
 Requires-Dist: panel

autorag-0.2.18/autorag/VERSION

@@ -0,0 +1 @@
+0.2.18

{autorag-0.2.17 → autorag-0.2.18}/autorag/__init__.py

@@ -15,7 +15,7 @@ from llama_index.llms.huggingface import HuggingFaceLLM
 from llama_index.llms.ollama import Ollama
 from llama_index.llms.openai import OpenAI
 from llama_index.llms.openai_like import OpenAILike
-from langchain.embeddings import OpenAIEmbeddings
+from langchain_openai.embeddings import OpenAIEmbeddings
 from rich.logging import RichHandler
 
 version_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), "VERSION")

{autorag-0.2.17 → autorag-0.2.18}/autorag/data/__init__.py

@@ -15,7 +15,7 @@ from langchain_community.document_loaders import (
 	DirectoryLoader,
 )
 from langchain_unstructured import UnstructuredLoader
-from langchain_upstage import UpstageLayoutAnalysisLoader
+from langchain_upstage import UpstageDocumentParseLoader
 
 from llama_index.core.node_parser import (
 	TokenTextSplitter,
@@ -56,7 +56,7 @@ parse_modules = {
 	# 6. All files
 	"directory": DirectoryLoader,
 	"unstructured": UnstructuredLoader,
-	"upstagelayoutanalysis": UpstageLayoutAnalysisLoader,
+	"upstagedocumentparse": UpstageDocumentParseLoader,
 }
 
 chunk_modules = {

{autorag-0.2.17 → autorag-0.2.18}/autorag/data/chunk/base.py

@@ -101,14 +101,14 @@ def __get_chunk_instance(module_type: str, chunk_method: str, **kwargs):
 def add_file_name(
 	file_name_language: str, file_names: List[str], chunk_texts: List[str]
 ) -> List[str]:
-	if file_name_language == "english":
+	if file_name_language == "en":
 		return list(
 			map(
 				lambda x: f"file_name: {x[1]}\n contents: {x[0]}",
 				zip(chunk_texts, file_names),
 			)
 		)
-	elif file_name_language == "korean":
+	elif file_name_language == "ko":
 		return list(
 			map(
 				lambda x: f"파일 제목: {x[1]}\n 내용: {x[0]}",
@@ -117,5 +117,5 @@ def add_file_name(
 		)
 	else:
 		raise ValueError(
-			f"Unsupported file_name_language: {file_name_language}. Choose from 'english' or 'korean'."
+			f"Unsupported file_name_language: {file_name_language}. Choose from 'en' or 'ko'."
 		)

{autorag-0.2.17 → autorag-0.2.18}/autorag/data/parse/clova.py

@@ -16,7 +16,7 @@ def clova_ocr(
 	data_path_list: List[str],
 	url: Optional[str] = None,
 	api_key: Optional[str] = None,
-	batch: int = 8,
+	batch: int = 5,
 	table_detection: bool = False,
 ) -> Tuple[List[str], List[str], List[int]]:
 	"""
@@ -46,6 +46,8 @@ def clova_ocr(
 			"Please set the API key for Clova OCR in the environment variable CLOVA_API_KEY "
 			"or directly set it on the config YAML file."
 		)
+	if batch > 5:
+		raise ValueError("The batch size should be less than or equal to 5.")
 
 	image_data_lst = list(
 		map(lambda data_path: pdf_to_images(data_path), data_path_list)

{autorag-0.2.17 → autorag-0.2.18}/autorag/evaluator.py

@@ -18,7 +18,11 @@ from autorag.nodes.retrieval.base import get_bm25_pkl_name
 from autorag.nodes.retrieval.bm25 import bm25_ingest
 from autorag.nodes.retrieval.vectordb import vectordb_ingest
 from autorag.schema import Node
-from autorag.schema.node import module_type_exists, extract_values_from_nodes
+from autorag.schema.node import (
+	module_type_exists,
+	extract_values_from_nodes,
+	extract_values_from_nodes_strategy,
+)
 from autorag.utils import (
 	cast_qa_dataset,
 	cast_corpus_dataset,
@@ -82,7 +86,7 @@ class Evaluator:
 
 		validate_qa_from_corpus_dataset(self.qa_data, self.corpus_data)
 
-		# copy dataset to project directory
+		# copy dataset to the project directory
 		if not os.path.exists(os.path.join(self.project_dir, "data")):
 			os.makedirs(os.path.join(self.project_dir, "data"))
 		qa_path_in_project = os.path.join(self.project_dir, "data", "qa.parquet")
@@ -100,7 +104,7 @@ class Evaluator:
 		trial_name = self.__get_new_trial_name()
 		self.__make_trial_dir(trial_name)
 
-		# copy yaml file to trial directory
+		# copy YAML file to the trial directory
 		shutil.copy(
 			yaml_path, os.path.join(self.project_dir, trial_name, "config.yaml")
 		)
@@ -148,13 +152,12 @@ class Evaluator:
 			bm25_tokenizer_list = list(
 				chain.from_iterable(
 					map(
-						lambda nodes: extract_values_from_nodes(
-							nodes, "bm25_tokenizer"
-						),
+						lambda nodes: self._find_bm25_tokenizer(nodes),
 						node_lines.values(),
 					)
 				)
 			)
+
 			if len(bm25_tokenizer_list) == 0:
 				bm25_tokenizer_list = ["porter_stemmer"]
 			for bm25_tokenizer in bm25_tokenizer_list:
@@ -166,6 +169,7 @@ class Evaluator:
 				# ingest because bm25 supports update new corpus data
 				bm25_ingest(bm25_dir, self.corpus_data, bm25_tokenizer=bm25_tokenizer)
 			logger.info("BM25 corpus embedding complete.")
+
 		if any(
 			list(
 				map(
@@ -178,9 +182,7 @@ class Evaluator:
 			embedding_models_list = list(
 				chain.from_iterable(
 					map(
-						lambda nodes: extract_values_from_nodes(
-							nodes, "embedding_model"
-						),
+						lambda nodes: self._find_embedding_model(nodes),
 						node_lines.values(),
 					)
 				)
@@ -517,3 +519,31 @@ class Evaluator:
 			}
 		)
 		return summary_lst
+
+	@staticmethod
+	def _find_bm25_tokenizer(nodes: List[Node]):
+		bm25_tokenizer_list = extract_values_from_nodes(nodes, "bm25_tokenizer")
+		strategy_tokenizer_list = list(
+			chain.from_iterable(
+				extract_values_from_nodes_strategy(nodes, "bm25_tokenizer")
+			)
+		)
+		return list(set(bm25_tokenizer_list + strategy_tokenizer_list))
+
+	@staticmethod
+	def _find_embedding_model(nodes: List[Node]):
+		embedding_models_list = extract_values_from_nodes(nodes, "embedding_model")
+		retrieval_module_dicts = extract_values_from_nodes_strategy(
+			nodes, "retrieval_modules"
+		)
+		for retrieval_modules in retrieval_module_dicts:
+			vectordb_modules = list(
+				filter(lambda x: x["module_type"] == "vectordb", retrieval_modules)
+			)
+			embedding_models_list.extend(
+				list(map(lambda x: x.get("embedding_model", None), vectordb_modules))
+			)
+		embedding_models_list = list(
+			filter(lambda x: x is not None, embedding_models_list)
+		)
+		return list(set(embedding_models_list))

{autorag-0.2.17 → autorag-0.2.18}/autorag/nodes/generator/openai_llm.py

@@ -12,10 +12,16 @@ from autorag.utils.util import get_event_loop, process_batch
 logger = logging.getLogger("AutoRAG")
 
 MAX_TOKEN_DICT = {  # model name : token limit
+	"o1-preview": 128_000,
+	"o1-preview-2024-09-12": 128_000,
+	"o1-mini": 128_000,
+	"o1-mini-2024-09-12": 128_000,
 	"gpt-4o-mini": 128_000,
 	"gpt-4o-mini-2024-07-18": 128_000,
 	"gpt-4o": 128_000,
+	"gpt-4o-2024-08-06": 128_000,
 	"gpt-4o-2024-05-13": 128_000,
+	"chatgpt-4o-latest": 128_000,
 	"gpt-4-turbo": 128_000,
 	"gpt-4-turbo-2024-04-09": 128_000,
 	"gpt-4-turbo-preview": 128_000,
@@ -82,7 +88,11 @@ def openai_llm(
 		kwargs.pop("n")
 		logger.warning("parameter n does not effective. It always set to 1.")
 
-	tokenizer = tiktoken.encoding_for_model(llm)
+	# TODO: fix this after updating tiktoken for the o1 model. It is not yet supported yet.
+	if llm.startswith("o1"):
+		tokenizer = tiktoken.get_encoding("o200k_base")
+	else:
+		tokenizer = tiktoken.encoding_for_model(llm)
 	if truncate:
 		max_token_size = MAX_TOKEN_DICT.get(llm) - 7  # because of chat token usage
 		if max_token_size is None:
@@ -99,7 +109,15 @@ def openai_llm(
 
 	client = AsyncOpenAI(api_key=api_key)
 	loop = get_event_loop()
-	tasks = [get_result(prompt, client, llm, tokenizer, **kwargs) for prompt in prompts]
+	if llm.startswith("o1"):
+		tasks = [
+			get_result_o1(prompt, client, llm, tokenizer, **kwargs)
+			for prompt in prompts
+		]
+	else:
+		tasks = [
+			get_result(prompt, client, llm, tokenizer, **kwargs) for prompt in prompts
+		]
 	result = loop.run_until_complete(process_batch(tasks, batch))
 	answer_result = list(map(lambda x: x[0], result))
 	token_result = list(map(lambda x: x[1], result))
@@ -132,6 +150,31 @@ async def get_result(
 	return answer, tokens, logprobs
 
 
+async def get_result_o1(
+	prompt: str, client: AsyncOpenAI, model: str, tokenizer: Encoding, **kwargs
+):
+	assert model.startswith("o1"), "This function only supports o1 model."
+	# The default temperature for the o1 model is 1. 1 is only supported.
+	# See https://platform.openai.com/docs/guides/reasoning about beta limitation of o1 models.
+	kwargs["temperature"] = 1
+	kwargs["top_p"] = 1
+	kwargs["presence_penalty"] = 0
+	kwargs["frequency_penalty"] = 0
+	response = await client.chat.completions.create(
+		model=model,
+		messages=[
+			{"role": "user", "content": prompt},
+		],
+		logprobs=False,
+		n=1,
+		**kwargs,
+	)
+	answer = response.choices[0].message.content
+	tokens = tokenizer.encode(answer, allowed_special="all")
+	pseudo_log_probs = [0.5] * len(tokens)
+	return answer, tokens, pseudo_log_probs
+
+
 def truncate_by_token(prompt: str, tokenizer: Encoding, max_token_size: int):
 	tokens = tokenizer.encode(prompt, allowed_special="all")
 	return tokenizer.decode(tokens[:max_token_size])

{autorag-0.2.17 → autorag-0.2.18}/autorag/schema/node.py

@@ -2,13 +2,13 @@ import itertools
 import logging
 from copy import deepcopy
 from dataclasses import dataclass, field
-from typing import Dict, List, Callable, Tuple
+from typing import Dict, List, Callable, Tuple, Any
 
 import pandas as pd
 
 from autorag.schema.module import Module
 from autorag.support import get_support_nodes
-from autorag.utils.util import make_combinations, explode
+from autorag.utils.util import make_combinations, explode, find_key_values
 
 logger = logging.getLogger("AutoRAG")
 
@@ -101,6 +101,23 @@ def extract_values_from_nodes(nodes: List[Node], key: str) -> List[str]:
 	return list(set(list(itertools.chain.from_iterable(values))))
 
 
+def extract_values_from_nodes_strategy(nodes: List[Node], key: str) -> List[Any]:
+	"""
+	This function extract values from nodes' strategy.
+
+	:param nodes: The nodes you want to extract values from.
+	:param key: The key string that you want to extract.
+	:return: The list of extracted values.
+	    It removes duplicated elements automatically.
+	"""
+	values = []
+	for node in nodes:
+		value_list = find_key_values(node.strategy, key)
+		if value_list:
+			values.extend(value_list)
+	return values
+
+
 def module_type_exists(nodes: List[Node], module_type: str) -> bool:
 	"""
 	This function check if the module type exists in the nodes.

{autorag-0.2.17 → autorag-0.2.18}/autorag/utils/util.py

@@ -574,3 +574,27 @@ def get_event_loop() -> AbstractEventLoop:
 		loop = asyncio.new_event_loop()
 		asyncio.set_event_loop(loop)
 	return loop
+
+
+def find_key_values(data, target_key: str) -> List[Any]:
+	"""
+	Recursively find all values for a specific key in a nested dictionary or list.
+
+	:param data: The dictionary or list to search.
+	:param target_key: The key to search for.
+	:return: A list of values associated with the target key.
+	"""
+	values = []
+
+	if isinstance(data, dict):
+		for key, value in data.items():
+			if key == target_key:
+				values.append(value)
+			if isinstance(value, (dict, list)):
+				values.extend(find_key_values(value, target_key))
+	elif isinstance(data, list):
+		for item in data:
+			if isinstance(item, (dict, list)):
+				values.extend(find_key_values(item, target_key))
+
+	return values

autorag-0.2.18/docs/source/api_spec/autorag.data.beta.filter.rst

@@ -0,0 +1,29 @@
+autorag.data.beta.filter package
+================================
+
+Submodules
+----------
+
+autorag.data.beta.filter.dontknow module
+----------------------------------------
+
+.. automodule:: autorag.data.beta.filter.dontknow
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+autorag.data.beta.filter.prompt module
+--------------------------------------
+
+.. automodule:: autorag.data.beta.filter.prompt
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Module contents
+---------------
+
+.. automodule:: autorag.data.beta.filter
+   :members:
+   :undoc-members:
+   :show-inheritance:

{autorag-0.2.17 → autorag-0.2.18}/docs/source/api_spec/autorag.data.beta.rst

@@ -14,6 +14,14 @@ Subpackages
 Submodules
 ----------
 
+autorag.data.beta.extract\_evidence module
+------------------------------------------
+
+.. automodule:: autorag.data.beta.extract_evidence
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
 autorag.data.beta.sample module
 -------------------------------
 

autorag-0.2.18/docs/source/api_spec/autorag.data.chunk.rst

@@ -0,0 +1,45 @@
+autorag.data.chunk package
+==========================
+
+Submodules
+----------
+
+autorag.data.chunk.base module
+------------------------------
+
+.. automodule:: autorag.data.chunk.base
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+autorag.data.chunk.langchain\_chunk module
+------------------------------------------
+
+.. automodule:: autorag.data.chunk.langchain_chunk
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+autorag.data.chunk.llama\_index\_chunk module
+---------------------------------------------
+
+.. automodule:: autorag.data.chunk.llama_index_chunk
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+autorag.data.chunk.run module
+-----------------------------
+
+.. automodule:: autorag.data.chunk.run
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Module contents
+---------------
+
+.. automodule:: autorag.data.chunk
+   :members:
+   :undoc-members:
+   :show-inheritance:

{autorag-0.2.17 → autorag-0.2.18}/docs/source/api_spec/autorag.schema.rst

@@ -4,6 +4,14 @@ autorag.schema package
 Submodules
 ----------
 
+autorag.schema.metricinput module
+---------------------------------
+
+.. automodule:: autorag.schema.metricinput
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
 autorag.schema.module module
 ----------------------------
 

{autorag-0.2.17 → autorag-0.2.18}/docs/source/data_creation/beta/chunk/chunk.md

@@ -21,7 +21,7 @@ chunker.start_chunking("your/path/to/chunk_config.yaml")
 ## Features
 
 ### 1. Add File Name
-You need to set one of 'English' and 'Korean'
+You need to set one of 'en'(=English) and 'ko'(=Korean) for the `add_file_name` parameter.
 The 'add_file_name' feature is to add a file_name to chunked_contents.
 This is used to prevent hallucination by retrieving contents from the wrong document.
 Default form of English is `"file_name: {file_name}\n contents: {content}"`

{autorag-0.2.17 → autorag-0.2.18}/requirements.txt

@@ -37,7 +37,7 @@ llama-index-readers-file
 llama-index-embeddings-openai
 llama-index-embeddings-huggingface
 # LLMs
-llama-index-llms-openai>=0.1.26
+llama-index-llms-openai>=0.2.7
 llama-index-llms-huggingface
 llama-index-llms-openai-like
 llama-index-llms-ollama
@@ -49,7 +49,7 @@ streamlit
 gradio
 
 ### Langchain ###
-langchain-core>=0.1.6
+langchain-core>=0.3.0
 langchain_unstructured
 langchain-upstage
 

{autorag-0.2.17 → autorag-0.2.18}/tests/autorag/data/chunk/test_langchain_chunk.py

@@ -47,7 +47,7 @@ def test_langchain_chunk_file_name_ko(chunk_instance):
 	doc_id, contents, path, start_end_idx, metadata = langchain_chunk_original(
 		base_texts,
 		chunk_instance,
-		file_name_language="korean",
+		file_name_language="ko",
 		metadata_list=base_metadata,
 	)
 	check_chunk_result(doc_id, contents, path, start_end_idx, metadata)
@@ -59,7 +59,7 @@ def test_langchain_chunk_file_name_ko(chunk_instance):
 
 def test_langchain_chunk_file_name_ko_node(chunk_instance):
 	result_df = langchain_chunk(
-		parsed_result, chunk_method="character", add_file_name="korean"
+		parsed_result, chunk_method="character", add_file_name="ko"
 	)
 	check_chunk_result_node(result_df)
 	assert len(result_df["doc_id"].tolist()) == 9
@@ -70,7 +70,7 @@ def test_langchain_chunk_file_name_eng(chunk_instance):
 	doc_id, contents, path, start_end_idx, metadata = langchain_chunk_original(
 		base_texts,
 		chunk_instance,
-		file_name_language="english",
+		file_name_language="en",
 		metadata_list=base_metadata,
 	)
 	check_chunk_result(doc_id, contents, path, start_end_idx, metadata)
@@ -82,7 +82,7 @@ def test_langchain_chunk_file_name_eng(chunk_instance):
 
 def test_langchain_chunk_file_name_eng_node():
 	result_df = langchain_chunk(
-		parsed_result, chunk_method="character", add_file_name="english"
+		parsed_result, chunk_method="character", add_file_name="en"
 	)
 	check_chunk_result_node(result_df)
 	assert len(result_df["doc_id"].tolist()) == 9

{autorag-0.2.17 → autorag-0.2.18}/tests/autorag/data/chunk/test_llama_index_chunk.py

@@ -47,7 +47,7 @@ def test_llama_index_chunk_file_name_ko(chunk_instance):
 	doc_id, contents, path, start_end_idx, metadata = llama_index_chunk_original(
 		base_texts,
 		chunk_instance,
-		file_name_language="korean",
+		file_name_language="ko",
 		metadata_list=base_metadata,
 	)
 	check_chunk_result(doc_id, contents, path, start_end_idx, metadata)
@@ -59,7 +59,7 @@ def test_llama_index_chunk_file_name_ko(chunk_instance):
 
 def test_llama_index_chunk_file_name_ko_node():
 	result_df = llama_index_chunk(
-		parsed_result, chunk_method="token", add_file_name="korean"
+		parsed_result, chunk_method="token", add_file_name="ko"
 	)
 	check_chunk_result_node(result_df)
 	assert len(result_df["doc_id"].tolist()) == 9
@@ -70,7 +70,7 @@ def test_llama_index_chunk_file_name_eng(chunk_instance):
 	doc_id, contents, path, start_end_idx, metadata = llama_index_chunk_original(
 		base_texts,
 		chunk_instance,
-		file_name_language="english",
+		file_name_language="en",
 		metadata_list=base_metadata,
 	)
 	check_chunk_result(doc_id, contents, path, start_end_idx, metadata)
@@ -82,7 +82,7 @@ def test_llama_index_chunk_file_name_eng(chunk_instance):
 
 def test_llama_index_chunk_file_name_eng_node():
 	result_df = llama_index_chunk(
-		parsed_result, chunk_method="token", add_file_name="english"
+		parsed_result, chunk_method="token", add_file_name="en"
 	)
 	check_chunk_result_node(result_df)
 	assert len(result_df["doc_id"].tolist()) == 9