Spaces:
Running
Running
Commit
·
73f1b9b
1
Parent(s):
e854023
fix table bug
Browse files- app.py +4 -1
- src/about.py +4 -0
app.py
CHANGED
@@ -10,6 +10,8 @@ from src.about import (
|
|
10 |
EVALUATION_QUEUE_TEXT,
|
11 |
INTRODUCTION_TEXT,
|
12 |
LLM_BENCHMARKS_TEXT,
|
|
|
|
|
13 |
TITLE,
|
14 |
EVALUATION_METRIC_TEXT,
|
15 |
)
|
@@ -429,7 +431,8 @@ with demo:
|
|
429 |
gr.Markdown(EVALUATION_METRIC_TEXT, elem_classes="markdown-text")
|
430 |
with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
|
431 |
gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
|
432 |
-
|
|
|
433 |
|
434 |
with gr.Row():
|
435 |
with gr.Accordion("📙 Citation", open=False):
|
|
|
10 |
EVALUATION_QUEUE_TEXT,
|
11 |
INTRODUCTION_TEXT,
|
12 |
LLM_BENCHMARKS_TEXT,
|
13 |
+
LLM_BENCHMARKS_TEXT2,
|
14 |
+
TABLE_TEXT,
|
15 |
TITLE,
|
16 |
EVALUATION_METRIC_TEXT,
|
17 |
)
|
|
|
431 |
gr.Markdown(EVALUATION_METRIC_TEXT, elem_classes="markdown-text")
|
432 |
with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
|
433 |
gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
|
434 |
+
gr.HTML(TABLE_TEXT)
|
435 |
+
gr.Markdown(LLM_BENCHMARKS_TEXT2, elem_classes="markdown-text")
|
436 |
|
437 |
with gr.Row():
|
438 |
with gr.Accordion("📙 Citation", open=False):
|
src/about.py
CHANGED
@@ -176,7 +176,9 @@ Planning
|
|
176 |
我们对上述10个数据集的数据进行了能力维度划分,归纳出具身智能场景需要的4大能力维度空间理解,感知,预测,规划。并按照能力维度,采样出一个样本数为2042的优质子集,能力维度定义和各维度的数据量如下:
|
177 |
|
178 |
We have categorized the data of the above 10 datasets by capability dimensions, and summarized four major capability dimensions required for embodied intelligence scenarios: spatial reasoning, perception, prediction, and planning. According to the capability dimensions, a high-quality subset with 2,042 samples was sampled. The definitions of the capability dimensions and the data volume of each dimension are as follows:
|
|
|
179 |
|
|
|
180 |
<table>
|
181 |
<thead>
|
182 |
<tr>
|
@@ -263,7 +265,9 @@ We have categorized the data of the above 10 datasets by capability dimensions,
|
|
263 |
</tr>
|
264 |
</tbody>
|
265 |
</table>
|
|
|
266 |
|
|
|
267 |
## EmbodiedVerse Tool - FlagEvalMM
|
268 |
|
269 |
FlagEvalMM是一个开源评估框架,旨在全面评估多模态模型,其提供了一种标准化的方法来评估跨各种任务和指标使用多种模式(文本、图像、视频)的模型。
|
|
|
176 |
我们对上述10个数据集的数据进行了能力维度划分,归纳出具身智能场景需要的4大能力维度空间理解,感知,预测,规划。并按照能力维度,采样出一个样本数为2042的优质子集,能力维度定义和各维度的数据量如下:
|
177 |
|
178 |
We have categorized the data of the above 10 datasets by capability dimensions, and summarized four major capability dimensions required for embodied intelligence scenarios: spatial reasoning, perception, prediction, and planning. According to the capability dimensions, a high-quality subset with 2,042 samples was sampled. The definitions of the capability dimensions and the data volume of each dimension are as follows:
|
179 |
+
"""
|
180 |
|
181 |
+
TABLE_TEXT = """
|
182 |
<table>
|
183 |
<thead>
|
184 |
<tr>
|
|
|
265 |
</tr>
|
266 |
</tbody>
|
267 |
</table>
|
268 |
+
"""
|
269 |
|
270 |
+
LLM_BENCHMARKS_TEXT2 = """
|
271 |
## EmbodiedVerse Tool - FlagEvalMM
|
272 |
|
273 |
FlagEvalMM是一个开源评估框架,旨在全面评估多模态模型,其提供了一种标准化的方法来评估跨各种任务和指标使用多种模式(文本、图像、视频)的模型。
|