Split (1)

train · 714k rows

text stringlengths 0 1.16k
question: ['How many dogs are in the image?'], responses:['1']
[('1', 0.12829009354978346), ('3', 0.12529928082343206), ('4', 0.12464806219229535), ('8', 0.12460015878893425), ('6', 0.12451220062887247), ('12', 0.124338487048427), ('2', 0.12420459433498025), ('47', 0.12410712263327517)]
[['1', '3', '4', '8', '6', '12', '2', '47']]
torch.Size([13, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
tensor([7.9300e-01, 1.9852e-02, 5.5128e-03, 1.3048e-03, 2.2201e-03, 1.1146e-03,
1.7694e-01, 5.7163e-05], device='cuda:2', grad_fn=<SoftmaxBackward0>)
1 *************
['1', '3', '4', '8', '6', '12', '2', '47'] tensor([7.9300e-01, 1.9852e-02, 5.5128e-03, 1.3048e-03, 2.2201e-03, 1.1146e-03,
1.7694e-01, 5.7163e-05], device='cuda:2', grad_fn=<SelectBackward0>)
最后的概率分布为: {True: tensor(0.1769, device='cuda:2', grad_fn=<DivBackward0>), False: tensor(0.8231, device='cuda:2', grad_fn=<DivBackward0>), 'Execute Error': tensor(-1.1921e-07, device='cuda:2', grad_fn=<DivBackward0>)}
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3396
tensor([9.7463e-01, 3.4067e-03, 1.2531e-03, 4.7473e-04, 7.3634e-04, 4.7046e-04,
1.9001e-02, 2.7210e-05], device='cuda:0', grad_fn=<SoftmaxBackward0>)
1 *************
['1', '3', '4', '8', '6', '12', '2', '47'] tensor([9.7463e-01, 3.4067e-03, 1.2531e-03, 4.7473e-04, 7.3634e-04, 4.7046e-04,
1.9001e-02, 2.7210e-05], device='cuda:0', grad_fn=<SelectBackward0>)
最后的概率分布为: {True: tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>), False: tensor(0.9810, device='cuda:0', grad_fn=<DivBackward0>), 'Execute Error': tensor(0., device='cuda:0', grad_fn=<DivBackward0>)}
[2024-10-23 14:51:31,996] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) \| optimizer_allgather: 1.49 \| optimizer_gradients: 0.26 \| optimizer_step: 0.32
[2024-10-23 14:51:31,997] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) \| forward_microstep: 9041.75 \| backward_microstep: 8751.50 \| backward_inner_microstep: 8745.93 \| backward_allreduce_microstep: 5.48 \| step_microstep: 7.57
[2024-10-23 14:51:31,997] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) \| forward: 9041.76 \| backward: 8751.50 \| backward_inner: 8745.96 \| backward_allreduce: 5.44 \| step: 7.58

1%\| \| 40/4844 [10:15<22:29:55, 16.86s/it]Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
Registering VQA_lavis step
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
ANSWER0=VQA(image=LEFT,question='How many dogs are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} > 10')
FINAL_ANSWER=RESULT(var=ANSWER1)
Registering EVAL step
Registering RESULT step
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
torch.Size([1, 3, 448, 448])
ANSWER0=VQA(image=RIGHT,question='How many round plates are visible in the image?')
ANSWER1=EVAL(expr='{ANSWER0} >= 2')
FINAL_ANSWER=RESULT(var=ANSWER1)
ANSWER0=VQA(image=LEFT,question='Is the dog's head laying down?')
ANSWER1=EVAL(expr='{ANSWER0}')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([3, 3, 448, 448])
ANSWER0=VQA(image=LEFT,question='How many antelopes are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} >= 3')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([3, 3, 448, 448])
torch.Size([13, 3, 448, 448])
question: ['How many dogs are in the image?'], responses:['15']
[('15', 0.12850265658859292), ('14', 0.12554598114685298), ('13', 0.12491622450863256), ('16', 0.12450938797787274), ('29', 0.12444750181633149), ('35', 0.12413627702798803), ('22', 0.12400388658176363), ('21', 0.12393808435196574)]
[['15', '14', '13', '16', '29', '35', '22', '21']]
torch.Size([1, 3, 448, 448]) knan debug pixel values shape
question: ['How many round plates are visible in the image?'], responses:['2']
question: ['How many antelopes are in the image?'], responses:['1']
[('2', 0.12961991198727602), ('3', 0.12561270547489775), ('4', 0.12556127085987287), ('1', 0.1254920833223361), ('5', 0.12407835939022728), ('8', 0.124024076973589), ('7', 0.12288810153923228), ('29', 0.12272349045256851)]
[['2', '3', '4', '1', '5', '8', '7', '29']]
[('1', 0.12829009354978346), ('3', 0.12529928082343206), ('4', 0.12464806219229535), ('8', 0.12460015878893425), ('6', 0.12451220062887247), ('12', 0.124338487048427), ('2', 0.12420459433498025), ('47', 0.12410712263327517)]
[['1', '3', '4', '8', '6', '12', '2', '47']]
torch.Size([3, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
torch.Size([3, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
tensor([0.2603, 0.1686, 0.2139, 0.1740, 0.0323, 0.0221, 0.0681, 0.0608],
device='cuda:1', grad_fn=<SoftmaxBackward0>)
15 *************
['15', '14', '13', '16', '29', '35', '22', '21'] tensor([0.2603, 0.1686, 0.2139, 0.1740, 0.0323, 0.0221, 0.0681, 0.0608],
device='cuda:1', grad_fn=<SelectBackward0>)
最后的概率分布为: {True: tensor(1.0000, device='cuda:1', grad_fn=<DivBackward0>), False: tensor(0., device='cuda:1', grad_fn=<MulBackward0>), 'Execute Error': tensor(5.9605e-08, device='cuda:1', grad_fn=<DivBackward0>)}
ANSWER0=VQA(image=LEFT,question='How many ferrets are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} == 2')
FINAL_ANSWER=RESULT(var=ANSWER1)
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
torch.Size([13, 3, 448, 448])
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
dynamic ViT batch size: 3, images per sample: 3.0, dynamic token length: 838
tensor([0.4378, 0.2194, 0.1716, 0.0760, 0.0532, 0.0229, 0.0184, 0.0006],
device='cuda:0', grad_fn=<SoftmaxBackward0>)
2 *************
['2', '3', '4', '1', '5', '8', '7', '29'] tensor([0.4378, 0.2194, 0.1716, 0.0760, 0.0532, 0.0229, 0.0184, 0.0006],
device='cuda:0', grad_fn=<SelectBackward0>)
tensor([9.2107e-01, 1.4887e-02, 5.3082e-03, 2.0766e-03, 2.7539e-03, 1.8240e-03,
5.1965e-02, 1.1047e-04], device='cuda:3', grad_fn=<SoftmaxBackward0>)
1 *************
['1', '3', '4', '8', '6', '12', '2', '47'] tensor([9.2107e-01, 1.4887e-02, 5.3082e-03, 2.0766e-03, 2.7539e-03, 1.8240e-03,
5.1965e-02, 1.1047e-04], device='cuda:3', grad_fn=<SelectBackward0>)
question: ['Is the dog'], responses:['yes']
最后的概率分布为: {True: tensor(0.9240, device='cuda:0', grad_fn=<DivBackward0>), False: tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>), 'Execute Error': tensor(5.9605e-08, device='cuda:0', grad_fn=<DivBackward0>)}
ANSWER0=VQA(image=LEFT,question='How many beetles are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} == 2')

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.