This is an automated email from the ASF dual-hosted git repository.
jin pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/incubator-hugegraph-ai.git
The following commit(s) were added to refs/heads/main by this push:
new fa2672a refactor(llm): support mathjax in rag query block V1 (#157)
fa2672a is described below
commit fa2672a1376f2ddca5417e093750ccee9fec715c
Author: SoJGooo <[email protected]>
AuthorDate: Thu Jan 16 17:12:07 2025 +0800
refactor(llm): support mathjax in rag query block V1 (#157)
* Update gremlin_example_index_query.py
---------
Co-authored-by: imbajin <[email protected]>
---
.../src/hugegraph_llm/config/prompt_config.py | 1 +
.../src/hugegraph_llm/demo/rag_demo/rag_block.py | 18 ++++++++++++------
.../index_op/gremlin_example_index_query.py | 7 ++++++-
.../src/hugegraph_llm/resources/demo/css.py | 20 ++++++++++++++++++++
4 files changed, 39 insertions(+), 7 deletions(-)
diff --git a/hugegraph-llm/src/hugegraph_llm/config/prompt_config.py
b/hugegraph-llm/src/hugegraph_llm/config/prompt_config.py
index fee85d2..51d84a1 100644
--- a/hugegraph-llm/src/hugegraph_llm/config/prompt_config.py
+++ b/hugegraph-llm/src/hugegraph_llm/config/prompt_config.py
@@ -26,6 +26,7 @@ Your task is to provide a precise and accurate answer based
on the given context
Given the context information and without using fictive knowledge,
answer the following query in a concise and professional manner.
+Please write your answer using Markdown with MathJax syntax, where inline math
is wrapped with `$...$`
Context information is below.
---------------------
diff --git a/hugegraph-llm/src/hugegraph_llm/demo/rag_demo/rag_block.py
b/hugegraph-llm/src/hugegraph_llm/demo/rag_demo/rag_block.py
index 477660b..ba2ab7e 100644
--- a/hugegraph-llm/src/hugegraph_llm/demo/rag_demo/rag_block.py
+++ b/hugegraph-llm/src/hugegraph_llm/demo/rag_demo/rag_block.py
@@ -112,15 +112,21 @@ def rag_answer(
def create_rag_block():
- # pylint: disable=R0915 (too-many-statements)
+ # pylint: disable=R0915 (too-many-statements),C0301
gr.Markdown("""## 1. HugeGraph RAG Query""")
with gr.Row():
with gr.Column(scale=2):
- inp = gr.Textbox(value=prompt.default_question, label="Question",
show_copy_button=True, lines=2)
- raw_out = gr.Textbox(label="Basic LLM Answer",
show_copy_button=True)
- vector_only_out = gr.Textbox(label="Vector-only Answer",
show_copy_button=True)
- graph_only_out = gr.Textbox(label="Graph-only Answer",
show_copy_button=True)
- graph_vector_out = gr.Textbox(label="Graph-Vector Answer",
show_copy_button=True)
+ inp = gr.Textbox(value=prompt.default_question, label="Question",
show_copy_button=True, lines=3)
+
+ # TODO: Only support inline formula now. Should support block
formula
+ gr.Markdown("Basic LLM Answer", elem_classes="output-box-label")
+ raw_out = gr.Markdown(elem_classes="output-box",
show_copy_button=True, latex_delimiters=[{"left":"$", "right":"$",
"display":False}])
+ gr.Markdown("Vector-only Answer", elem_classes="output-box-label")
+ vector_only_out = gr.Markdown(elem_classes="output-box",
show_copy_button=True, latex_delimiters=[{"left":"$", "right":"$",
"display":False}])
+ gr.Markdown("Graph-only Answer", elem_classes="output-box-label")
+ graph_only_out = gr.Markdown(elem_classes="output-box",
show_copy_button=True, latex_delimiters=[{"left":"$", "right":"$",
"display":False}])
+ gr.Markdown("Graph-Vector Answer", elem_classes="output-box-label")
+ graph_vector_out = gr.Markdown(elem_classes="output-box",
show_copy_button=True, latex_delimiters=[{"left":"$", "right":"$",
"display":False}])
answer_prompt_input = gr.Textbox(
value=prompt.answer_prompt, label="Query Prompt",
show_copy_button=True, lines=7
diff --git
a/hugegraph-llm/src/hugegraph_llm/operators/index_op/gremlin_example_index_query.py
b/hugegraph-llm/src/hugegraph_llm/operators/index_op/gremlin_example_index_query.py
index 4029995..31d9c2b 100644
---
a/hugegraph-llm/src/hugegraph_llm/operators/index_op/gremlin_example_index_query.py
+++
b/hugegraph-llm/src/hugegraph_llm/operators/index_op/gremlin_example_index_query.py
@@ -56,7 +56,12 @@ class GremlinExampleIndexQuery:
def _build_default_example_index(self):
properties = pd.read_csv(os.path.join(resource_path, "demo",
"text2gremlin.csv")).to_dict(orient="records")
- embeddings = [self.embedding.get_text_embedding(row["query"]) for row
in tqdm(properties)]
+ from concurrent.futures import ThreadPoolExecutor
+ # TODO: use asyncio for IO tasks
+ with ThreadPoolExecutor() as executor:
+ embeddings = list(
+ tqdm(executor.map(self.embedding.get_text_embedding,
[row["query"] for row in properties]),
+ total=len(properties)))
vector_index = VectorIndex(len(embeddings[0]))
vector_index.add(embeddings, properties)
vector_index.to_index_file(self.index_dir)
diff --git a/hugegraph-llm/src/hugegraph_llm/resources/demo/css.py
b/hugegraph-llm/src/hugegraph_llm/resources/demo/css.py
index 117a068..0f2d48d 100644
--- a/hugegraph-llm/src/hugegraph_llm/resources/demo/css.py
+++ b/hugegraph-llm/src/hugegraph_llm/resources/demo/css.py
@@ -38,4 +38,24 @@ footer {
white-space: -o-pre-wrap; /* Opera 7 */
word-wrap: break-word; /* Internet Explorer 5.5+ */
}
+
+.output-box {
+ border: 1px solid #ccc;
+ border-radius: 5px;
+ padding: 8px;
+ box-sizing: border-box;
+ min-height: 50px;
+ font-family: Arial, sans-serif;
+ line-height: 1.5;
+ margin-top: -5px;
+ width: 99.5%;
+ max-height: 300px;
+ overflow-y: auto;
+}
+
+.output-box-label {
+ font-size: 14px;
+ font-weight: bold;
+ margin-bottom: -5px;
+}
"""
