lingyuzeng/embedding_atlas
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

修复interactive viewer中的metadata配置和列名问题

Browse Source 
1. 在metadata中添加database.load配置，确保前端能正确加载数据到DuckDB
2. 将neighbors列名从'neighbors'改为'__neighbors'以符合embedding-atlas标准
3. 更新launch_interactive_viewer和visualize_csv_comparison函数中的相关配置
This commit is contained in:
lingyuzeng
2025-10-23 18:00:10 +08:00
parent bbf1746046
commit 991bcc491f
1 changed files with 13 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
src/visualization/comparison.py
View File

@@ -20,13 +20,17 @@ def launch_interactive_viewer(df: pd.DataFrame, text_column: str, port: int = 50
from embedding_atlas.utils import Hasher from embedding_atlas.utils import Hasher
import pathlib import pathlib
# 创建metadata # 创建metadata - 添加database配置
metadata = { metadata = {
"columns": { "columns": {
"id": "_row_index", "id": "_row_index",
"text": text_column, "text": text_column,
"embedding": {"x": "projection_x", "y": "projection_y"}, "embedding": {"x": "projection_x", "y": "projection_y"},
"neighbors": "neighbors" "neighbors": "__neighbors" # 使用双下划线
},
"database": {
"type": "wasm",
"load": True # 关键: 告诉前端加载数据
} }
} }
@@ -109,7 +113,7 @@ def create_embedding_service(
text=text_column, text=text_column,
x="projection_x", x="projection_x",
y="projection_y", y="projection_y",
neighbors="neighbors", neighbors="__neighbors", # 改为双下划线
model=model, model=model,
batch_size=batch_size, batch_size=batch_size,
umap_args=umap_args umap_args=umap_args
@@ -234,7 +238,7 @@ def visualize_csv_comparison(
text=text_column, text=text_column,
x="projection_x", x="projection_x",
y="projection_y", y="projection_y",
neighbors="neighbors", neighbors="__neighbors", # 改为双下划线
model=model, model=model,
batch_size=batch_size, batch_size=batch_size,
umap_args=umap_args umap_args=umap_args
@@ -298,11 +302,11 @@ def main():
formatter_class=argparse.RawDescriptionHelpFormatter, formatter_class=argparse.RawDescriptionHelpFormatter,
epilog=""" epilog="""
Examples: Examples:
python visualize_csv_comparison.py file1.csv file2.csv python -m visualization.comparison file1.csv file2.csv
python visualize_csv_comparison.py file1.csv file2.csv --column1 smiles --column2 SMILES python -m visualization.comparison file1.csv file2.csv --column1 smiles --column2 SMILES
python visualize_csv_comparison.py file1.csv file2.csv --label1 "Dataset A" --label2 "Dataset B" python -m visualization.comparison file1.csv file2.csv --label1 "Dataset A" --label2 "Dataset B"
python visualize_csv_comparison.py file1.csv file2.csv --output comparison.png python -m visualization.comparison file1.csv file2.csv --output comparison.png
python visualize_csv_comparison.py file1.csv file2.csv --interactive --port 8080 python -m visualization.comparison file1.csv file2.csv --interactive --port 8080
""" """
) )