AutoRAG 0.3.2__tar.gz → 0.3.3rc16__tar.gz

autorag-0.3.3rc16/.github/workflows/docker-push.yml

@@ -0,0 +1,91 @@
+name: Docker Push
+
+on:
+  push:
+    branches: [ "main" ]
+
+env:
+  DOCKER_REPO: "autoraghq/autorag"
+
+jobs:
+  check-version:
+    runs-on: ubuntu-latest
+    outputs:
+      changed: ${{ steps.version_changed.outputs.changed }}
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4.2.0
+
+    - name: Get changed files
+      id: changed-files
+      uses: tj-actions/changed-files@v45.0.3
+
+    - name: Check for VERSION file change
+      id: version_changed
+      env:
+        ALL_CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
+      run: |
+        echo "changed=false" >> $GITHUB_OUTPUT
+        if echo "${ALL_CHANGED_FILES}" | grep -q 'VERSION'; then
+          echo "changed=true" >> $GITHUB_OUTPUT
+        fi
+
+  build-and-push:
+    needs: check-version
+    if: needs.check-version.outputs.changed == 'true'
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        variant: [all] # [all, ko, dev, parsing]
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4.2.0
+
+    - name: Read VERSION file
+      run: echo "VERSION=$(cat autorag/VERSION)" >> $GITHUB_ENV
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v3.7.1
+
+    - name: Login to Docker Hub
+      uses: docker/login-action@v3.3.0
+      with:
+        username: ${{ secrets.DOCKER_HUB_USERNAME }}
+        password: ${{ secrets.DOCKER_HUB_ACCESS_TOKEN }}
+
+    - name: Extract metadata (tags, labels) for Docker
+      id: meta
+      uses: docker/metadata-action@v5.5.1
+      with:
+        images: ${{ env.DOCKER_REPO }}
+        tags: |
+          type=raw,value=${{ env.VERSION }}-${{ matrix.variant }}
+          type=raw,value=${{ matrix.variant }},enable=${{ github.ref == format('refs/heads/{0}', 'main') }}
+          type=raw,value=latest-${{ matrix.variant }},enable=${{ github.ref == format('refs/heads/{0}', 'main') }}
+
+    - name: Build and push Docker image
+      uses: docker/build-push-action@v6.9.0
+      with:
+        context: .
+        file: ./Dockerfile
+        push: true
+        # push: ${{ github.event_name != 'pull_request' && matrix.variant != 'test' }}
+        tags: ${{ env.DOCKER_REPO }}:${{ env.VERSION }}-${{matrix.variant}}
+        labels: ${{ steps.meta.outputs.labels }}
+        build-args: |
+          TARGET_STAGE=${{ matrix.variant }}
+
+    - name: Tag and push 'all' for production
+      run: |
+        docker pull ${{ env.DOCKER_REPO }}:${{ env.VERSION }}-${{matrix.variant}}
+        docker tag ${{ env.DOCKER_REPO }}:${{ env.VERSION }}-${{matrix.variant}} ${{ env.DOCKER_REPO }}:latest
+        docker push ${{ env.DOCKER_REPO }}:${{ env.VERSION }}-${{matrix.variant}}
+        docker push ${{ env.DOCKER_REPO }}:latest
+
+    # - name: Update Docker Hub description
+    #   uses: peter-evans/dockerhub-description@v3
+    #   with:
+    #     username: ${{ secrets.DOCKER_HUB_USERNAME }}
+    #     password: ${{ secrets.DOCKER_HUB_PASSWORD }}
+    #     repository: ${{ env.DOCKER_REPO }}
+    #     short-description: ${{ github.event.repository.description }}

{autorag-0.3.2 → autorag-0.3.3rc16}/AutoRAG.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: AutoRAG
-Version: 0.3.2
+Version: 0.3.3rc16
 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
@@ -235,19 +235,18 @@ Requires-Dist: rouge_score
 Requires-Dist: rich
 Requires-Dist: chromadb>=0.5.0
 Requires-Dist: click
-Requires-Dist: Flask
 Requires-Dist: torch
 Requires-Dist: sentencepiece
-Requires-Dist: guidance
 Requires-Dist: cohere>=5.8.0
 Requires-Dist: tokenlog>=0.0.2
 Requires-Dist: aiohttp
 Requires-Dist: bert_score
 Requires-Dist: sentence-transformers
 Requires-Dist: FlagEmbedding
-Requires-Dist: ragas
 Requires-Dist: llmlingua
 Requires-Dist: peft
+Requires-Dist: optimum[nncf,openvino]
+Requires-Dist: quart
 Requires-Dist: llama-index>=0.11.0
 Requires-Dist: llama-index-core>=0.11.0
 Requires-Dist: llama-index-readers-file
@@ -263,6 +262,7 @@ Requires-Dist: gradio
 Requires-Dist: langchain-core>=0.3.0
 Requires-Dist: langchain-unstructured>=0.1.5
 Requires-Dist: langchain-upstage
+Requires-Dist: langchain-community>=0.3.0
 Requires-Dist: panel
 Requires-Dist: seaborn
 Requires-Dist: ipykernel
@@ -656,14 +656,17 @@ You can run this pipeline as an API server.
 Check out the API endpoint at [here](deploy/api_endpoint.md).
 
 ```python
-from autorag.deploy import Runner
+import nest_asyncio
+from autorag.deploy import ApiRunner
 
-runner = Runner.from_trial_folder('/your/path/to/trial_dir')
+nest_asyncio.apply()
+
+runner = ApiRunner.from_trial_folder('/your/path/to/trial_dir')
 runner.run_api_server()
 ```
 
 ```bash
-autorag run_api --config_path your/path/to/pipeline.yaml --host 0.0.0.0 --port 8000
+autorag run_api --trial_dir your/path/to/trial_dir --host 0.0.0.0 --port 8000
 ```
 
 The cli command uses extracted config YAML file. If you want to know it more, check out [here](https://docs.auto-rag.com/tutorial.html#extract-pipeline-and-evaluate-test-dataset).

{autorag-0.3.2 → autorag-0.3.3rc16}/AutoRAG.egg-info/SOURCES.txt

@@ -6,6 +6,7 @@ CONTRIBUTING.md
 Dockerfile
 LICENSE
 README.md
+build_and_push.sh
 docker-compose.yml
 pyproject.toml
 requirements.txt
@@ -13,6 +14,7 @@ requirements.txt
 .github/dependabot.yml
 .github/ISSUE_TEMPLATE/bug_report.md
 .github/ISSUE_TEMPLATE/feature_request.md
+.github/workflows/docker-push.yml
 .github/workflows/publish.yml
 .github/workflows/sphinx.yml
 .github/workflows/test.yml
@@ -27,7 +29,6 @@ autorag/__init__.py
 autorag/chunker.py
 autorag/cli.py
 autorag/dashboard.py
-autorag/deploy.py
 autorag/evaluator.py
 autorag/node_line.py
 autorag/parser.py
@@ -81,12 +82,18 @@ autorag/data/qa/query/openai_gen_query.py
 autorag/data/qa/query/prompt.py
 autorag/data/utils/__init__.py
 autorag/data/utils/util.py
+autorag/deploy/__init__.py
+autorag/deploy/api.py
+autorag/deploy/base.py
+autorag/deploy/gradio.py
+autorag/deploy/swagger.yml
 autorag/evaluation/__init__.py
 autorag/evaluation/generation.py
 autorag/evaluation/retrieval.py
 autorag/evaluation/retrieval_contents.py
 autorag/evaluation/util.py
 autorag/evaluation/metric/__init__.py
+autorag/evaluation/metric/deepeval_prompt.py
 autorag/evaluation/metric/generation.py
 autorag/evaluation/metric/retrieval.py
 autorag/evaluation/metric/retrieval_contents.py
@@ -133,6 +140,7 @@ autorag/nodes/passagereranker/flag_embedding_llm.py
 autorag/nodes/passagereranker/jina.py
 autorag/nodes/passagereranker/koreranker.py
 autorag/nodes/passagereranker/monot5.py
+autorag/nodes/passagereranker/openvino.py
 autorag/nodes/passagereranker/pass_reranker.py
 autorag/nodes/passagereranker/rankgpt.py
 autorag/nodes/passagereranker/run.py
@@ -291,6 +299,7 @@ docs/source/nodes/passage_reranker/flag_embedding_reranker.md
 docs/source/nodes/passage_reranker/jina_reranker.md
 docs/source/nodes/passage_reranker/koreranker.md
 docs/source/nodes/passage_reranker/monot5.md
+docs/source/nodes/passage_reranker/openvino_reranker.md
 docs/source/nodes/passage_reranker/passage_reranker.md
 docs/source/nodes/passage_reranker/rankgpt.md
 docs/source/nodes/passage_reranker/sentence_transformer_reranker.md
@@ -419,6 +428,7 @@ tests/autorag/nodes/passagereranker/test_flag_embedding_llm.py
 tests/autorag/nodes/passagereranker/test_jina_reranker.py
 tests/autorag/nodes/passagereranker/test_koreranker.py
 tests/autorag/nodes/passagereranker/test_monot5.py
+tests/autorag/nodes/passagereranker/test_openvino_reranker.py
 tests/autorag/nodes/passagereranker/test_pass_reranker.py
 tests/autorag/nodes/passagereranker/test_passage_reranker_base.py
 tests/autorag/nodes/passagereranker/test_passage_reranker_run.py
@@ -461,6 +471,7 @@ tests/resources/simple.yaml
 tests/resources/simple_chunk.yaml
 tests/resources/simple_mock.yaml
 tests/resources/simple_parse.yaml
+tests/resources/simple_with_llm.yaml
 tests/resources/test_bm25_retrieval.pkl
 tests/resources/chunk_data/sample_parsed.parquet
 tests/resources/data_creation/raw_dir/sample1.txt

{autorag-0.3.2 → autorag-0.3.3rc16}/AutoRAG.egg-info/requires.txt

@@ -14,19 +14,18 @@ rouge_score
 rich
 chromadb>=0.5.0
 click
-Flask
 torch
 sentencepiece
-guidance
 cohere>=5.8.0
 tokenlog>=0.0.2
 aiohttp
 bert_score
 sentence-transformers
 FlagEmbedding
-ragas
 llmlingua
 peft
+optimum[nncf,openvino]
+quart
 llama-index>=0.11.0
 llama-index-core>=0.11.0
 llama-index-readers-file
@@ -42,6 +41,7 @@ gradio
 langchain-core>=0.3.0
 langchain-unstructured>=0.1.5
 langchain-upstage
+langchain-community>=0.3.0
 panel
 seaborn
 ipykernel

autorag-0.3.3rc16/Dockerfile

@@ -0,0 +1,61 @@
+# Base stage: Install common dependencies
+FROM python:3.10-slim AS base
+
+# Install system dependencies
+RUN apt-get update && \
+    apt-get install -y \
+    build-essential \
+    gcc \
+    libssl-dev \
+    poppler-utils \
+    tesseract-ocr \
+    tesseract-ocr-eng \
+    tesseract-ocr-kor && \
+    rm -rf /var/lib/apt/lists/*
+
+# Set work directory
+WORKDIR /usr/src/app
+
+# Install Python dependencies
+RUN pip install --upgrade pip setuptools setuptools-scm
+COPY requirements.txt /usr/src/app/requirements.txt
+RUN pip install -r requirements.txt
+
+# Copy project files
+COPY . /usr/src/app
+RUN pip install -e ./
+
+# Test stage
+FROM base AS test
+RUN pip install pytest pytest-xdist
+
+# Ko stage
+FROM base AS ko
+RUN pip install --no-cache-dir kiwipiepy>=0.18.0 konlpy
+ENTRYPOINT ["python", "-m", "autorag.cli"]
+
+# Dev stage
+FROM base AS dev
+RUN pip install --no-cache-dir ruff pre-commit
+ENTRYPOINT ["python", "-m", "autorag.cli"]
+
+# Parsing stage
+FROM base AS parsing
+RUN pip install --no-cache-dir PyMuPDF pdfminer.six pdfplumber unstructured jq "unstructured[pdf]" "PyPDF2<3.0" pdf2image
+ENTRYPOINT ["python", "-m", "autorag.cli"]
+
+# Parsing stage
+FROM base AS all
+# TODO
+ENTRYPOINT ["python", "-m", "autorag.cli"]
+
+# Production stage (includes all features)
+FROM base AS production
+RUN pip install --no-cache-dir \
+    kiwipiepy>=0.18.0 konlpy \
+    ruff pre-commit \
+    PyMuPDF pdfminer.six pdfplumber unstructured jq "unstructured[pdf]" "PyPDF2<3.0" pdf2image
+
+COPY projects /usr/src/app/projects
+
+ENTRYPOINT ["python", "-m", "autorag.cli"]

{autorag-0.3.2 → autorag-0.3.3rc16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: AutoRAG
-Version: 0.3.2
+Version: 0.3.3rc16
 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
@@ -235,19 +235,18 @@ Requires-Dist: rouge_score
 Requires-Dist: rich
 Requires-Dist: chromadb>=0.5.0
 Requires-Dist: click
-Requires-Dist: Flask
 Requires-Dist: torch
 Requires-Dist: sentencepiece
-Requires-Dist: guidance
 Requires-Dist: cohere>=5.8.0
 Requires-Dist: tokenlog>=0.0.2
 Requires-Dist: aiohttp
 Requires-Dist: bert_score
 Requires-Dist: sentence-transformers
 Requires-Dist: FlagEmbedding
-Requires-Dist: ragas
 Requires-Dist: llmlingua
 Requires-Dist: peft
+Requires-Dist: optimum[nncf,openvino]
+Requires-Dist: quart
 Requires-Dist: llama-index>=0.11.0
 Requires-Dist: llama-index-core>=0.11.0
 Requires-Dist: llama-index-readers-file
@@ -263,6 +262,7 @@ Requires-Dist: gradio
 Requires-Dist: langchain-core>=0.3.0
 Requires-Dist: langchain-unstructured>=0.1.5
 Requires-Dist: langchain-upstage
+Requires-Dist: langchain-community>=0.3.0
 Requires-Dist: panel
 Requires-Dist: seaborn
 Requires-Dist: ipykernel
@@ -656,14 +656,17 @@ You can run this pipeline as an API server.
 Check out the API endpoint at [here](deploy/api_endpoint.md).
 
 ```python
-from autorag.deploy import Runner
+import nest_asyncio
+from autorag.deploy import ApiRunner
 
-runner = Runner.from_trial_folder('/your/path/to/trial_dir')
+nest_asyncio.apply()
+
+runner = ApiRunner.from_trial_folder('/your/path/to/trial_dir')
 runner.run_api_server()
 ```
 
 ```bash
-autorag run_api --config_path your/path/to/pipeline.yaml --host 0.0.0.0 --port 8000
+autorag run_api --trial_dir your/path/to/trial_dir --host 0.0.0.0 --port 8000
 ```
 
 The cli command uses extracted config YAML file. If you want to know it more, check out [here](https://docs.auto-rag.com/tutorial.html#extract-pipeline-and-evaluate-test-dataset).

{autorag-0.3.2 → autorag-0.3.3rc16}/README.md

@@ -366,14 +366,17 @@ You can run this pipeline as an API server.
 Check out the API endpoint at [here](deploy/api_endpoint.md).
 
 ```python
-from autorag.deploy import Runner
+import nest_asyncio
+from autorag.deploy import ApiRunner
 
-runner = Runner.from_trial_folder('/your/path/to/trial_dir')
+nest_asyncio.apply()
+
+runner = ApiRunner.from_trial_folder('/your/path/to/trial_dir')
 runner.run_api_server()
 ```
 
 ```bash
-autorag run_api --config_path your/path/to/pipeline.yaml --host 0.0.0.0 --port 8000
+autorag run_api --trial_dir your/path/to/trial_dir --host 0.0.0.0 --port 8000
 ```
 
 The cli command uses extracted config YAML file. If you want to know it more, check out [here](https://docs.auto-rag.com/tutorial.html#extract-pipeline-and-evaluate-test-dataset).

autorag-0.3.3rc16/autorag/VERSION

@@ -0,0 +1 @@
+0.3.3-rc16

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/cli.py

@@ -7,10 +7,11 @@ from pathlib import Path
 from typing import Optional
 
 import click
+import nest_asyncio
 
 from autorag import dashboard
-from autorag.deploy import Runner
 from autorag.deploy import extract_best_config as original_extract_best_config
+from autorag.deploy.api import ApiRunner
 from autorag.evaluator import Evaluator
 from autorag.validator import Validator
 
@@ -48,15 +49,27 @@ def evaluate(config, qa_data_path, corpus_data_path, project_dir):
 
 
 @click.command()
-@click.option("--config_path", type=str, help="Path to extracted config yaml file.")
+@click.option(
+	"--config_path", type=str, help="Path to extracted config yaml file.", default=None
+)
 @click.option("--host", type=str, default="0.0.0.0", help="Host address")
 @click.option("--port", type=int, default=8000, help="Port number")
+@click.option(
+	"--trial_dir",
+	type=click.Path(file_okay=False, dir_okay=True, exists=True),
+	default=None,
+	help="Path to trial directory.",
+)
 @click.option(
 	"--project_dir", help="Path to project directory.", type=str, default=None
 )
-def run_api(config_path, host, port, project_dir):
-	runner = Runner.from_yaml(config_path, project_dir=project_dir)
+def run_api(config_path, host, port, trial_dir, project_dir):
+	if trial_dir is None:
+		runner = ApiRunner.from_yaml(config_path, project_dir=project_dir)
+	else:
+		runner = ApiRunner.from_trial_folder(trial_dir)
 	logger.info(f"Running API server at {host}:{port}...")
+	nest_asyncio.apply()
 	runner.run_api_server(host, port)
 
 

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/legacy/qacreation/base.py

@@ -166,7 +166,7 @@ def make_qa_with_existing_qa(
 		content_size = len(corpus_df)
 
 	logger.info("Loading local embedding model...")
-	embeddings = autorag.embedding_models[embedding_model]
+	embeddings = autorag.embedding_models[embedding_model]()
 
 	# Vector DB creation
 	if collection is None:

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/legacy/qacreation/ragas.py

@@ -5,8 +5,6 @@ import pandas as pd
 from langchain_core.embeddings import Embeddings
 from langchain_core.language_models import BaseChatModel
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
-from ragas.testset import TestsetGenerator
-from ragas.testset.evolutions import simple, reasoning, multi_context
 
 from autorag.data.utils.util import corpus_df_to_langchain_documents
 from autorag.utils import cast_qa_dataset
@@ -37,6 +35,9 @@ def generate_qa_ragas(
 	    You can input 'with_debugging_logs', 'is_async', 'raise_exceptions', and 'run_config'.
 	:return: QA dataset dataframe.
 	"""
+	from ragas.testset import TestsetGenerator
+	from ragas.testset.evolutions import simple, reasoning, multi_context
+
 	if generator_llm is None:
 		generator_llm = ChatOpenAI(model="gpt-3.5-turbo-16k")
 	if critic_llm is None:

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/legacy/qacreation/simple.py

@@ -3,12 +3,10 @@ import pathlib
 import uuid
 from typing import Callable
 
-import guidance
 import pandas as pd
-from guidance import models, gen
 
 
-def generate_qa_row(llm: models.Model, corpus_data_row):
+def generate_qa_row(llm, corpus_data_row):
 	"""
 	this sample code to generate rag dataset using OpenAI chat model
 
@@ -16,6 +14,9 @@ def generate_qa_row(llm: models.Model, corpus_data_row):
 	:param corpus_data_row: need "contents" column
 	:return: should to be dict which has "query", "generation_gt" columns at least.
 	"""
+	from guidance import gen
+	import guidance
+
 	temp_llm = llm
 	with guidance.user():
 		temp_llm += f"""
@@ -48,7 +49,7 @@ def generate_qa_row(llm: models.Model, corpus_data_row):
 
 
 def generate_simple_qa_dataset(
-	llm: models.Model,
+	llm,
 	corpus_data: pd.DataFrame,
 	output_filepath: str,
 	generate_row_function: Callable,

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/parse/clova.py

@@ -99,8 +99,10 @@ async def clova_ocr_pure(
 			raise RuntimeError(
 				f"Invalid response from Clova API: {resp_json['detail']}"
 			)
-
-		table_html = json_to_html_table(resp_json["images"][0]["tables"][0]["cells"])
+		if "tables" in resp_json["images"][0].keys():
+			table_html = json_to_html_table(
+				resp_json["images"][0]["tables"][0]["cells"]
+			)
 		page_text = extract_text_from_fields(resp_json["images"][0]["fields"])
 
 		if table_html:

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/parse/langchain_parse.py

@@ -1,4 +1,5 @@
 import multiprocessing as mp
+from itertools import chain
 from typing import List, Tuple
 
 from autorag.data import parse_modules
@@ -20,6 +21,7 @@ def langchain_parse(
 	if parse_method in ["directory", "unstructured"]:
 		results = parse_all_files(data_path_list, parse_method, **kwargs)
 		texts, path = results[0], results[1]
+		pages = [-1] * len(texts)
 
 	else:
 		num_workers = mp.cpu_count()
@@ -30,14 +32,14 @@ def langchain_parse(
 				[(data_path, parse_method, kwargs) for data_path in data_path_list],
 			)
 
-		texts, path = zip(*results)
-		texts, path = list(texts), list(path)
-	pages = [-1] * len(texts)
+		texts, path, pages = (list(chain.from_iterable(item)) for item in zip(*results))
 
 	return texts, path, pages
 
 
-def langchain_parse_pure(data_path: str, parse_method: str, kwargs) -> Tuple[str, str]:
+def langchain_parse_pure(
+	data_path: str, parse_method: str, kwargs
+) -> Tuple[List[str], List[str], List[int]]:
 	"""
 	Parses a single file using the specified parse method.
 
@@ -54,12 +56,18 @@ def langchain_parse_pure(data_path: str, parse_method: str, kwargs) -> Tuple[str
 
 	# Load the text from the file
 	documents = parse_instance.load()
-	text = documents[0].page_content
+
+	texts = list(map(lambda x: x.page_content, documents))
+	path = [data_path] * len(texts)
+	if parse_method in ["pymupdf", "pdfplumber", "pypdf", "pypdfium2"]:
+		pages = list(range(1, len(documents) + 1))
+	else:
+		pages = [-1] * len(texts)
 
 	# Clean up the parse instance
 	del parse_instance
 
-	return text, data_path
+	return texts, path, pages
 
 
 def parse_all_files(

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/parse/llamaparse.py

@@ -1,4 +1,5 @@
 from typing import List, Tuple
+from itertools import chain
 
 from llama_parse import LlamaParse
 
@@ -30,15 +31,18 @@ def llama_parse(
 
 	del parse_instance
 
-	texts, path = zip(*results)
-	pages = [-1] * len(texts)
+	texts, path, pages = (list(chain.from_iterable(item)) for item in zip(*results))
 
-	return list(texts), list(path), pages
+	return texts, path, pages
 
 
-async def llama_parse_pure(data_path: str, parse_instance) -> Tuple[str, str]:
+async def llama_parse_pure(
+	data_path: str, parse_instance
+) -> Tuple[List[str], List[str], List[int]]:
 	documents = await parse_instance.aload_data(data_path)
 
-	text = documents[0].text
+	texts = list(map(lambda x: x.text, documents))
+	path = [data_path] * len(texts)
+	pages = list(range(1, len(documents) + 1))
 
-	return text, data_path
+	return texts, path, pages

{autorag-0.3.2 → autorag-0.3.3rc16}/autorag/data/qa/schema.py

@@ -79,7 +79,7 @@ class Corpus:
 		"""
 		if not save_path.endswith(".parquet"):
 			raise ValueError("save_path must be ended with .parquet")
-		save_df = self.data[["doc_id", "contents", "metadata"]].reset_index(drop=True)
+		save_df = self.data.reset_index(drop=True)
 		save_df.to_parquet(save_path)
 
 	def batch_apply(

autorag-0.3.3rc16/autorag/deploy/__init__.py

@@ -0,0 +1,9 @@
+from .base import (
+	extract_node_line_names,
+	extract_node_strategy,
+	summary_df_to_yaml,
+	extract_best_config,
+	Runner,
+)
+from .api import ApiRunner
+from .gradio import GradioRunner