Update README.md
Browse files
README.md
CHANGED
@@ -116,39 +116,44 @@ configs:
|
|
116 |
path: data/sg_what_relation-*
|
117 |
---
|
118 |
|
119 |
-
# Dataset Card for TaskMeAnything-v1-imageqa-
|
120 |
-
<h2 align="center"> TaskMeAnything-v1-imageqa-
|
121 |
|
122 |
<h2 align="center"> <a href="https://www.task-me-anything.org/">π Website</a> | <a href="https://arxiv.org/abs/2406.11775">π Paper</a> | <a href="https://huggingface.co/collections/jieyuz2/taskmeanything-664ebf028ab2524c0380526a">π€ Huggingface</a> | <a href="https://huggingface.co/spaces/zixianma/TaskMeAnything-UI">π» Interface</a></h2>
|
123 |
|
124 |
<h5 align="center"> If you like our project, please give us a star β on GitHub for latest update. </h2>
|
125 |
|
126 |
-
## TaskMeAnything-v1-
|
127 |
-
[TaskMeAnything-v1-imageqa-
|
128 |
-
|
129 |
|
130 |
-
|
|
|
|
|
|
|
131 |
```
|
132 |
import datasets
|
133 |
|
134 |
-
dataset_name = 'weikaih/TaskMeAnything-v1-imageqa-
|
135 |
dataset = datasets.load_dataset(dataset_name, split = TASK_GENERATOR_SPLIT)
|
136 |
```
|
137 |
-
where `TASK_GENERATOR_SPLIT` is one of the task generators, eg, `
|
138 |
|
139 |
## Evaluation Results
|
140 |
|
141 |
### Overall
|
142 |
-
|
|
|
143 |
|
144 |
### Breakdown performance on each task types
|
145 |
-
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/8gq7G9Ky228eooi9Mt4ep.png)
|
146 |
|
147 |
-
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31
|
|
|
|
|
148 |
|
149 |
-
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/
|
150 |
|
151 |
-
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/
|
152 |
|
153 |
## Out-of-Scope Use
|
154 |
This dataset should not be used for training models.
|
|
|
116 |
path: data/sg_what_relation-*
|
117 |
---
|
118 |
|
119 |
+
# Dataset Card for TaskMeAnything-v1-imageqa-2024
|
120 |
+
<h2 align="center"> TaskMeAnything-v1-imageqa-2024 benchmark dataset</h2>
|
121 |
|
122 |
<h2 align="center"> <a href="https://www.task-me-anything.org/">π Website</a> | <a href="https://arxiv.org/abs/2406.11775">π Paper</a> | <a href="https://huggingface.co/collections/jieyuz2/taskmeanything-664ebf028ab2524c0380526a">π€ Huggingface</a> | <a href="https://huggingface.co/spaces/zixianma/TaskMeAnything-UI">π» Interface</a></h2>
|
123 |
|
124 |
<h5 align="center"> If you like our project, please give us a star β on GitHub for latest update. </h2>
|
125 |
|
126 |
+
## TaskMeAnything-v1-2024
|
127 |
+
[TaskMeAnything-v1-imageqa-2024](https://huggingface.co/datasets/weikaih/TaskMeAnything-v1-imageqa-2024) is a benchmark for reflecting the current progress of MLMs by `automatically` finding tasks that SOTA MLMs struggle with using the TaskMeAnything Top-K queries.
|
128 |
+
This benchmark includes 3,279 2d questions, 7,095 3d questions, and 1,896 real image questions that the TaskMeAnything algorithm automatically approximated as challenging for over 12 popular MLMs.
|
129 |
|
130 |
+
|
131 |
+
The dataset contains 19 splits, while each splits contains 400+ questions from a specific task generator in TaskMeAnything-v1. For each row of dataset, it includes: image, question, options, answer and its corresponding task plan.
|
132 |
+
|
133 |
+
## Load TaskMeAnything-v1-2024 ImageQA Dataset
|
134 |
```
|
135 |
import datasets
|
136 |
|
137 |
+
dataset_name = 'weikaih/TaskMeAnything-v1-imageqa-2024'
|
138 |
dataset = datasets.load_dataset(dataset_name, split = TASK_GENERATOR_SPLIT)
|
139 |
```
|
140 |
+
where `TASK_GENERATOR_SPLIT` is one of the task generators, eg, `2024_2d_how_many`.
|
141 |
|
142 |
## Evaluation Results
|
143 |
|
144 |
### Overall
|
145 |
+
|
146 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/_KadJKJSHhZXXfIfePaUg.png)
|
147 |
|
148 |
### Breakdown performance on each task types
|
|
|
149 |
|
150 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/-DrQ90FuGatJE4CuHsWS9.png)
|
151 |
+
|
152 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/6D33K2tSc1OYF4_f6YJ63.png)
|
153 |
|
154 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/eKzh5ghGNVrCluVmnkZW0.png)
|
155 |
|
156 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/65cb0dcc4913057ac82a7a31/sm8dAmjxsXmJu8oeqLaeQ.png)
|
157 |
|
158 |
## Out-of-Scope Use
|
159 |
This dataset should not be used for training models.
|