The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 1 new columns ({'score'}) and 3 missing columns ({'width', 'image', 'height'}).
This happened while the json dataset builder was generating data using
hf://datasets/tyfeld/underwood/text_data/gsm8k_train_socratic_en.jsonl (at revision 0fa1efb609a5bd3b060f3a178c5f3890f0d13230)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1831, in _prepare_split_single
writer.write_table(table)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py", line 644, in write_table
pa_table = table_cast(pa_table, self._schema)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
id: int64
conversations: list<item: struct<from: string, value: string>>
child 0, item: struct<from: string, value: string>
child 0, from: string
child 1, value: string
score: double
to
{'id': Value('int64'), 'image': Value('string'), 'width': Value('int64'), 'height': Value('int64'), 'conversations': List({'from': Value('string'), 'value': Value('string')})}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1451, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 994, in stream_convert_to_parquet
builder._prepare_split(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1833, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 1 new columns ({'score'}) and 3 missing columns ({'width', 'image', 'height'}).
This happened while the json dataset builder was generating data using
hf://datasets/tyfeld/underwood/text_data/gsm8k_train_socratic_en.jsonl (at revision 0fa1efb609a5bd3b060f3a178c5f3890f0d13230)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
id int64 | image string | width int64 | height int64 | conversations list |
|---|---|---|---|---|
0 | images/53951.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor question: <ref>select the favourite icon which is on the left side of the page</ref>, locate the bounding box of the element in the image that holds the key to this question."
},
{
"from": "gpt",
"value": "<ref>select the favourite icon which is on the l... |
1 | images/39509.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for solving the task: <ref>select the button got it dismiss this</ref>."
},
{
"from": "gpt",
"value": "<ref>select the button got it dismiss this</ref><box>[[237, 627, 761, 684]]</box>"
}
] |
2 | images/29984.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the instruction: <ref>click the third image</ref>, output the bounding box for the region in the screen that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>click the third image</ref><box>[[38, 310, 155, 376]]</box>"
},... |
3 | images/52042.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nShow the bounding box coordinates to click for completing the objective: <ref>click on icon left to bookmarks</ref>."
},
{
"from": "gpt",
"value": "<ref>click on icon left to bookmarks</ref><box>[[38, 464, 97, 497]]</box>"
}
] |
4 | images/62636.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box for the element in the screen to complete the task: <ref>select the first icon adjacent right side of learn japanese</ref>."
},
{
"from": "gpt",
"value": "<ref>select the first icon adjacent right side of learn japanese</ref><box>[[... |
5 | images/41543.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease identify the area of the element with bounding box that can provide an answer to this question: <ref>click on the third question mark in the page</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the third question mark in the page</ref><box>[[616, ... |
6 | images/17200.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nFor question: <ref>click the logo beside fighters</ref>, locate the bounding box of the element in the image that holds the key to this question."
},
{
"from": "gpt",
"value": "<ref>click the logo beside fighters</ref><box>[[514, 898, 735, 918]]</box>"
},
... |
7 | images/19024.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the instruction: <ref>go to first image</ref>, output the bounding box for the region in the screen that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>go to first image</ref><box>[[58, 314, 228, 410]]</box>"
},
{
... |
8 | images/50656.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the instruction: <ref>click on the image icon above restaurants text option</ref>, output the bounding box for the region in the screen that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>click on the image icon above res... |
9 | images/114.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box coordinates that is relevant to the objective: <ref>click the right side height button 6</ref>."
},
{
"from": "gpt",
"value": "<ref>click the right side height button 6</ref><box>[[500, 291, 951, 345]]</box>"
}
] |
10 | images/21137.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nSpecify the bounding box coordinates for completing the instruction: <ref>select the text below the create an account</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text below the create an account</ref><box>[[38, 131, 475, 181]]</box>"
},
{
"... |
11 | images/2415.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the region to solve the task: <ref>click on the number 1</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the number 1</ref><box>[[25, 123, 74, 183]]</box>"
},
{
"from": "human",
"value": "Where is the bounding box c... |
12 | images/7739.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click on icon on the right side of verbose logging</ref>."
},
{
"from": "gpt",
"value": "<ref>click on icon on the right side of verbose logging</ref><box>[[330, 385, 9... |
13 | images/52871.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCan you provide the click location with bounding box for completing the task: <ref>click on right arrow at the bottom</ref>."
},
{
"from": "gpt",
"value": "<ref>click on right arrow at the bottom</ref><box>[[805, 824, 981, 923]]</box>"
},
{
"from": "... |
14 | images/35663.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nHighlight the click area with bounding box to complete the task: <ref>select the icon left to flight status</ref>."
},
{
"from": "gpt",
"value": "<ref>select the icon left to flight status</ref><box>[[24, 105, 97, 146]]</box>"
},
{
"from": "human",
... |
15 | images/2935.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nFor the operation: <ref>click on the icon on the right side of undisclosed</ref>, output the bounding box of the area in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>click on the icon on the right side of undisclosed</ref><box>[[... |
16 | images/22780.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nCan you determine the bounding box coordinates to complete my task: <ref>select the symbol which is to the second right of menu button</ref>."
},
{
"from": "gpt",
"value": "<ref>select the symbol which is to the second right of menu button</ref><box>[[299, 4... |
17 | images/36625.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCan you pinpoint the location with bounding box for completing the task: <ref>select the image of police scanner</ref>."
},
{
"from": "gpt",
"value": "<ref>select the image of police scanner</ref><box>[[442, 282, 559, 347]]</box>"
},
{
"from": "human... |
18 | images/50898.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for completing the objective: <ref>select more at the bottom of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select more at the bottom of the page</ref><box>[[800, 853, 999, 933]]</box>"
},
{
"fro... |
19 | images/70031.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for completing the objective: <ref>click on the option beside view</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the option beside view</ref><box>[[666, 798, 1000, 864]]</box>"
},
{
"from": "human"... |
20 | images/35361.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the region correctly responding to this task: <ref>click on the contact icon under dth beside text field</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the contact icon under dth beside text field</ref><box>[[845, 283, ... |
21 | images/16665.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates to solve the task: <ref>go to star icon</ref>."
},
{
"from": "gpt",
"value": "<ref>go to star icon</ref><box>[[926, 109, 975, 136]]</box>"
},
{
"from": "human",
"value": "Indicate the precise location with bou... |
22 | images/60410.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the task: <ref>click on menu button</ref>, output the bounding box for the region in the image that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>click on menu button</ref><box>[[883, 57, 951, 95]]</box>"
},
{
"f... |
23 | images/66371.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCan you provide the click location with bounding box for completing the task: <ref>click on the first image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the first image</ref><box>[[12, 192, 255, 327]]</box>"
},
{
"from": "human",
"value": ... |
24 | images/18451.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nCould you specify the bounding box that is most relevant to the task: <ref>click the music icon</ref>."
},
{
"from": "gpt",
"value": "<ref>click the music icon</ref><box>[[4, 805, 174, 901]]</box>"
},
{
"from": "human",
"value": "For the operatio... |
25 | images/62178.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the region correctly responding to this task: <ref>click the first circular shaped image on the left side of the web page</ref>."
},
{
"from": "gpt",
"value": "<ref>click the first circular shaped image on the left side of t... |
26 | images/7724.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box to solve the task: <ref>click on image right to 12 pm</ref>."
},
{
"from": "gpt",
"value": "<ref>click on image right to 12 pm</ref><box>[[147, 826, 220, 875]]</box>"
},
{
"from": "human",
"value": "Point o... |
27 | images/29643.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the operation: <ref>select 1 line text above acura</ref>, output the bounding box of the area in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>select 1 line text above acura</ref><box>[[312, 829, 1000, 866]]</box>"
},
{
... |
28 | images/35089.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nCould you specify the bounding box that is most relevant to the task: <ref>click on the button which is next to button</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the button which is next to button</ref><box>[[600, 868, 799, 934]]</box>"
},
{
... |
29 | images/18861.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor task: <ref>go to the third image from the top of the page</ref>, where should I click to complete the task, and provide the bounding box of the region."
},
{
"from": "gpt",
"value": "<ref>go to the third image from the top of the page</ref><box>[[3, 285,... |
30 | images/21959.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates that can solve this task: <ref>click on the second blog</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the second blog</ref><box>[[19, 292, 980, 443]]</box>"
},
{
"from": "human",
"value": "Highlight the ... |
31 | images/47615.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for solving the task: <ref>click on the back arrow in top left corner</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the back arrow in top left corner</ref><box>[[34, 62, 116, 84]]</box>"
},
{
"from... |
32 | images/65160.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhat is the bounding box coordinates that is most relevant to this task: <ref>select icon of under armour</ref>."
},
{
"from": "gpt",
"value": "<ref>select icon of under armour</ref><box>[[38, 579, 174, 655]]</box>"
},
{
"from": "human",
"value":... |
33 | images/61302.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nHighlight the area with bounding box that I can click to execute the task: <ref>click on search icon next below 3 vertical dots at the top right corner of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>click on search icon next below 3 vertical dots at ... |
34 | images/16994.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nShow me the bounding box coordinates that can provide an answer to the operation: <ref>select the checkbox just right to flickr</ref>."
},
{
"from": "gpt",
"value": "<ref>select the checkbox just right to flickr</ref><box>[[839, 99, 917, 143]]</box>"
}
] |
35 | images/12880.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nTask: <ref>select the icon which is before cash</ref>, output the bounding box of the area in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>select the icon which is before cash</ref><box>[[38, 311, 97, 344]]</box>"
}
] |
36 | images/38022.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box for the element in the screen to complete the task: <ref>select the text gallery below downloads</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text gallery below downloads</ref><box>[[10, 360, 495, 601]]</box>"
},
{
... |
37 | images/33229.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click on the search icon</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the search icon</ref><box>[[863, 38, 1000, 104]]</box>"
},
{
"from": "human",
"... |
38 | images/25689.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCould you specify the bounding box that is most relevant to the task: <ref>click on the image which is just above the close button</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the image which is just above the close button</ref><box>[[248, 463, 751, 6... |
39 | images/34280.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the task: <ref>select the image below review reference which consists of three options</ref>, output the bounding box for the region in the image that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>select the image below ... |
40 | images/65566.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor question: <ref>click on the dragon image</ref>, locate the bounding box of the element in the image that holds the key to this question."
},
{
"from": "gpt",
"value": "<ref>click on the dragon image</ref><box>[[4, 400, 165, 490]]</box>"
},
{
"fro... |
41 | images/69224.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease identify the location with bounding box that can complete the instruction: <ref>select the text field which is below mail on the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text field which is below mail on the page</ref><box>[[38, 526,... |
42 | images/51055.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box that can finalize the instruction: <ref>select filipino on the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select filipino on the page</ref><box>[[38, 349, 961, 393]]</box>"
},
{
"from": "human",
"value... |
43 | images/42255.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the instruction: <ref>select the image above fan</ref>, output the bounding box for the region in the screen that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>select the image above fan</ref><box>[[472, 875, 526, 906]]<... |
44 | images/18134.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCan you determine the bounding box coordinates to complete my task: <ref>select the last four alphabets in the second row of the key board</ref>."
},
{
"from": "gpt",
"value": "<ref>select the last four alphabets in the second row of the key board</ref><bo... |
45 | images/67053.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhat is the bounding box coordinates that is most relevant to this task: <ref>click on the first image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the first image</ref><box>[[829, 157, 975, 291]]</box>"
},
{
"from": "human",
"value": "Spe... |
46 | images/9448.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for solving the task: <ref>select the next button which is on the bottom right of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select the next button which is on the bottom right of the page</ref><box>[[7... |
47 | images/688.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box coordinates that is relevant to the objective: <ref>select the text box which says take a shortcut</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text box which says take a shortcut</ref><box>[[302, 629, 697, 694]]</box>"
}... |
48 | images/70400.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box to solve the task: <ref>select the check box that is to the right of hide warning bar</ref>."
},
{
"from": "gpt",
"value": "<ref>select the check box that is to the right of hide warning bar</ref><box>[[839, 562, 917, ... |
49 | images/3800.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nPlease identify the area of the element with bounding box that can provide an answer to this question: <ref>select the text box which says password</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text box which says password</ref><box>[[37, 318, 962, 3... |
50 | images/64302.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhat is the bounding box coordinates that is most relevant to this task: <ref>click on icon beside share</ref>."
},
{
"from": "gpt",
"value": "<ref>click on icon beside share</ref><box>[[883, 38, 1000, 104]]</box>"
},
{
"from": "human",
"value": ... |
51 | images/38749.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates to solve the task: <ref>click on the add icon</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the add icon</ref><box>[[649, 32, 766, 104]]</box>"
},
{
"from": "human",
"value": "Can you pinpoint the locati... |
52 | images/46218.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the element in the image that directly relates to solving this question: <ref>icon before ideas text</ref>."
},
{
"from": "gpt",
"value": "<ref>icon before ideas text</ref><box>[[774, 882, 808, 904]]</box>"
},
{
"from": "hu... |
53 | images/45600.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nCan you pinpoint the location with bounding box for completing the task: <ref>click the arrow left to suggested retail</ref>."
},
{
"from": "gpt",
"value": "<ref>click the arrow left to suggested retail</ref><box>[[16, 376, 42, 403]]</box>"
},
{
"fro... |
54 | images/36006.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box that can finalize the instruction: <ref>select the image on the top right corner</ref>."
},
{
"from": "gpt",
"value": "<ref>select the image on the top right corner</ref><box>[[883, 32, 1000, 101]]</box>"
},
{
... |
55 | images/57253.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the region to solve the task: <ref>go to the first icon from bottom</ref>."
},
{
"from": "gpt",
"value": "<ref>go to the first icon from bottom</ref><box>[[38, 866, 161, 931]]</box>"
},
{
"from": "human",
"value": "Plea... |
56 | images/21553.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>select the icon which is left side of the mcdelivery</ref>."
},
{
"from": "gpt",
"value": "<ref>select the icon which is left side of the mcdelivery</ref><box>[[68, 43,... |
57 | images/9416.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nFor the operation: <ref>below the password option</ref>, output the bounding box of the region in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>below the password option</ref><box>[[20, 694, 98, 738]]</box>"
},
{
"from": "... |
58 | images/60295.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nSpecify the location with bounding box to finish the objective: <ref>click on second image from bottom</ref>."
},
{
"from": "gpt",
"value": "<ref>click on second image from bottom</ref><box>[[449, 552, 812, 757]]</box>"
},
{
"from": "human",
"val... |
59 | images/52924.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click on the chat icon to the top right corner</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the chat icon to the top right corner</ref><box>[[917, 43, 980, 77]]<... |
60 | images/19963.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nShow me the bounding box coordinates that can provide an answer to the operation: <ref>click on xkill</ref>."
},
{
"from": "gpt",
"value": "<ref>click on xkill</ref><box>[[19, 811, 137, 841]]</box>"
},
{
"from": "human",
"value": "Identify the bo... |
61 | images/9543.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nCan you determine the bounding box coordinates to complete my task: <ref>select option privacy which is on the right corner of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select option privacy which is on the right corner of the page</ref><box>[[829,... |
62 | images/48926.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click on the email option in the required field</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the email option in the required field</ref><box>[[12, 437, 987, 524... |
63 | images/29257.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the region to solve the task: <ref>select the rotation icon</ref>."
},
{
"from": "gpt",
"value": "<ref>select the rotation icon</ref><box>[[472, 858, 550, 902]]</box>"
},
{
"from": "human",
"value": "Show me the boundin... |
64 | images/200.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the region to solve the task: <ref>select the more options icon</ref>."
},
{
"from": "gpt",
"value": "<ref>select the more options icon</ref><box>[[728, 32, 1000, 98]]</box>"
},
{
"from": "human",
"value": "Output the b... |
65 | images/47622.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease identify the location with bounding box that can complete the instruction: <ref>click on champaign il </ref>."
},
{
"from": "gpt",
"value": "<ref>click on champaign il </ref><box>[[34, 133, 357, 167]]</box>"
},
{
"from": "human",
"value": ... |
66 | images/2768.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the operation: <ref>select the k button</ref>, output the bounding box of the region in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>select the k button</ref><box>[[431, 128, 567, 192]]</box>"
}
] |
67 | images/53607.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nInstruction: <ref>click on the online option</ref>, output the bounding box of the region in the screen that can complete the task."
},
{
"from": "gpt",
"value": "<ref>click on the online option</ref><box>[[262, 122, 500, 170]]</box>"
},
{
"from": "h... |
68 | images/13087.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhat is the bounding box coordinates that is most relevant to this task: <ref>click on more button in second image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on more button in second image</ref><box>[[917, 755, 975, 788]]</box>"
}
] |
69 | images/14778.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nShow me the bounding box coordinates that can provide an answer to the operation: <ref>click on second image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on second image</ref><box>[[24, 306, 364, 497]]</box>"
},
{
"from": "human",
"value": "T... |
70 | images/60222.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates to solve the task: <ref>click on more icon options</ref>."
},
{
"from": "gpt",
"value": "<ref>click on more icon options</ref><box>[[902, 38, 1000, 104]]</box>"
},
{
"from": "human",
"value": "For task: <ref>c... |
71 | images/47840.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates to solve the task: <ref>the third check box of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>the third check box of the page</ref><box>[[38, 397, 87, 424]]</box>"
},
{
"from": "human",
"value": "Output t... |
72 | images/10805.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box coordinates that is relevant to the objective: <ref>click on heart symbol which is on right side of super smash bros</ref>."
},
{
"from": "gpt",
"value": "<ref>click on heart symbol which is on right side of super smash bros</ref><b... |
73 | images/71864.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box for the element in the screen to complete the task: <ref>enter left arrow</ref>."
},
{
"from": "gpt",
"value": "<ref>enter left arrow</ref><box>[[55, 51, 116, 85]]</box>"
},
{
"from": "human",
"value": "Output the boundi... |
74 | images/41125.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the region to solve the task: <ref>click on message icon at top right</ref>."
},
{
"from": "gpt",
"value": "<ref>click on message icon at top right</ref><box>[[865, 60, 975, 105]]</box>"
},
{
"from": "human",
"value": "... |
75 | images/59972.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the task: <ref>click on the second image from left side</ref>, output the bounding box for the region in the image that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>click on the second image from left side</ref><box>[[6... |
76 | images/68301.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for solving the task: <ref>click on cancel button</ref>."
},
{
"from": "gpt",
"value": "<ref>click on cancel button</ref><box>[[718, 43, 805, 93]]</box>"
},
{
"from": "human",
"value": "Specify the b... |
77 | images/42309.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nProvide the bounding box coordinates to solve the task: <ref>select the date 9 of january</ref>."
},
{
"from": "gpt",
"value": "<ref>select the date 9 of january</ref><box>[[27, 608, 157, 659]]</box>"
},
{
"from": "human",
"value": "For the instr... |
78 | images/68796.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nCould you specify the bounding box that is most relevant to the task: <ref>click on camera icon</ref>."
},
{
"from": "gpt",
"value": "<ref>click on camera icon</ref><box>[[422, 763, 577, 844]]</box>"
},
{
"from": "human",
"value": "Could you spec... |
79 | images/61215.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nSpecify the bounding box coordinates for completing the instruction: <ref>click on icon left to more apps</ref>."
},
{
"from": "gpt",
"value": "<ref>click on icon left to more apps</ref><box>[[24, 558, 97, 585]]</box>"
},
{
"from": "human",
"valu... |
80 | images/25578.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click the button on the top right corner of the web page</ref>."
},
{
"from": "gpt",
"value": "<ref>click the button on the top right corner of the web page</ref><box>[... |
81 | images/1518.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box that can finalize the instruction: <ref>click on the icon to the right side of 000</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the icon to the right side of 000</ref><box>[[932, 290, 990, 322]]</box>"
},
{
... |
82 | images/15860.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the operation: <ref>click on the icon left to search icon</ref>, output the bounding box of the area in the image that can complete the task."
},
{
"from": "gpt",
"value": "<ref>click on the icon left to search icon</ref><box>[[71, 879, 129, 912]]</box>"... |
83 | images/11425.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>click on the more options icon which is on the top right corner</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the more options icon which is on the top right corn... |
84 | images/27607.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor the task: <ref>click on skip option</ref>, output the bounding box for the region in the image that is most relevant to answering the question."
},
{
"from": "gpt",
"value": "<ref>click on skip option</ref><box>[[453, 679, 546, 767]]</box>"
},
{
... |
85 | images/46575.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nPlease identify the location with bounding box that can complete the instruction: <ref>select more</ref>."
},
{
"from": "gpt",
"value": "<ref>select more</ref><box>[[792, 846, 1000, 934]]</box>"
},
{
"from": "human",
"value": "Task: <ref>button w... |
86 | images/64468.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box for the element in the screen to complete the task: <ref>select to rememb on top of the page</ref>."
},
{
"from": "gpt",
"value": "<ref>select to rememb on top of the page</ref><box>[[250, 32, 500, 95]]</box>"
}
] |
87 | images/445.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nHighlight the area with bounding box that I can click to execute the task: <ref>go to the text next to the or buy ear spy pro</ref>."
},
{
"from": "gpt",
"value": "<ref>go to the text next to the or buy ear spy pro</ref><box>[[757, 648, 975, 689]]</box>"
}... |
88 | images/50959.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nOutput the bounding box of the element in the image that directly relates to solving this question: <ref>select the first image on the web page</ref>."
},
{
"from": "gpt",
"value": "<ref>select the first image on the web page</ref><box>[[666, 434, 1000, 684]... |
89 | images/22033.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nSpecify the bounding box coordinates for completing the instruction: <ref>click on the menu bar on the bottom right image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the menu bar on the bottom right image</ref><box>[[932, 676, 990, 709]]</box>"
},
... |
90 | images/68069.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease identify the area of the element with bounding box that can provide an answer to this question: <ref>click on the text which is just beside the arrow mark</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the text which is just beside the arrow mark... |
91 | images/32770.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the region correctly responding to this task: <ref>select the first icon from the top right</ref>."
},
{
"from": "gpt",
"value": "<ref>select the first icon from the top right</ref><box>[[902, 36, 975, 94]]</box>"
},
{
... |
92 | images/27903.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIdentify the bounding box coordinates that is relevant to the objective: <ref>select the symbol which is right to network</ref>."
},
{
"from": "gpt",
"value": "<ref>select the symbol which is right to network</ref><box>[[562, 107, 980, 165]]</box>"
},
{
... |
93 | images/53551.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the precise location with bounding box that can solve the problem: <ref>select the start symbol</ref>."
},
{
"from": "gpt",
"value": "<ref>select the start symbol</ref><box>[[591, 32, 727, 98]]</box>"
},
{
"from": "human",
"value": "For ... |
94 | images/15106.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nPlease output the bounding box of the element correctly responding to this question: <ref>select the text below resend code</ref>."
},
{
"from": "gpt",
"value": "<ref>select the text below resend code</ref><box>[[257, 562, 743, 617]]</box>"
},
{
"fro... |
95 | images/5267.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nWhat is the bounding box coordinates that is most relevant to this task: <ref>click the icon at the left side of groceries</ref>."
},
{
"from": "gpt",
"value": "<ref>click the icon at the left side of groceries</ref><box>[[722, 122, 790, 161]]</box>"
}
] |
96 | images/2445.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box to solve the task: <ref>click on calender icon at top left</ref>."
},
{
"from": "gpt",
"value": "<ref>click on calender icon at top left</ref><box>[[38, 109, 79, 169]]</box>"
},
{
"from": "human",
"value": ... |
97 | images/57600.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nIndicate the location with bounding box to solve the task: <ref>click on first image</ref>."
},
{
"from": "gpt",
"value": "<ref>click on first image</ref><box>[[20, 109, 981, 366]]</box>"
},
{
"from": "human",
"value": "For the task: <ref>click o... |
98 | images/38450.jpg | 540 | 960 | [
{
"from": "human",
"value": "<image>\nWhere is the bounding box coordinates to click for solving the task: <ref>click on the profile which is on the top left side</ref>."
},
{
"from": "gpt",
"value": "<ref>click on the profile which is on the top left side</ref><box>[[29, 48, 107, 92]]</box>"
... |
99 | images/76.jpg | 1,080 | 1,920 | [
{
"from": "human",
"value": "<image>\nFor task: <ref>select the icon on the top right of the web page beside clock icon</ref>, where should I click to complete the task, and provide the bounding box of the region."
},
{
"from": "gpt",
"value": "<ref>select the icon on the top right of the web pa... |
End of preview.