Datasets:
TMarcus
/
InternVL

Dataset card Data Studio Files
xet
 Community
1
text
stringlengths
0
1.16k
torch.Size([7, 3, 448, 448]) knan debug pixel values shape
tensor([9.9995e-01, 2.7797e-05, 7.0056e-07, 9.9015e-10, 1.3882e-06, 1.2893e-07,
7.4820e-07, 2.0707e-05], device='cuda:2', grad_fn=<SoftmaxBackward0>)
0 *************
['0', 'circles', 'maroon', 'large', 'rooster', 'nuts', 'beige', 'bottle'] tensor([9.9995e-01, 2.7797e-05, 7.0056e-07, 9.9015e-10, 1.3882e-06, 1.2893e-07,
7.4820e-07, 2.0707e-05], device='cuda:2', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(0., device='cuda:2', grad_fn=<MulBackward0>), False: tensor(0.9999, device='cuda:2', grad_fn=<DivBackward0>), 'Execute Error': tensor(5.1498e-05, device='cuda:2', grad_fn=<DivBackward0>)}
ANSWER0=VQA(image=LEFT,question='Is there a castle with a broken tower in the image?')
ANSWER1=EVAL(expr='{ANSWER0}')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([3, 3, 448, 448])
question: ['Is there a castle with a broken tower in the image?'], responses:['yes']
[('yes', 0.1298617250866936), ('congratulations', 0.12464161604141298), ('no', 0.12445222599225532), ('honey', 0.12437056445881921), ('solid', 0.12422595371654564), ('right', 0.12419889376311324), ('candle', 0.12414264780165109), ('chocolate', 0.12410637313950891)]
[['yes', 'congratulations', 'no', 'honey', 'solid', 'right', 'candle', 'chocolate']]
torch.Size([3, 3, 448, 448]) knan debug pixel values shape
tensor([9.9998e-01, 1.6373e-07, 6.5502e-09, 4.0355e-09, 3.9420e-09, 3.3596e-07,
2.2828e-05, 3.4271e-08], device='cuda:3', grad_fn=<SoftmaxBackward0>)
1 *************
['1', '3', '4', '8', '6', '12', '2', '47'] tensor([9.9998e-01, 1.6373e-07, 6.5502e-09, 4.0355e-09, 3.9420e-09, 3.3596e-07,
2.2828e-05, 3.4271e-08], device='cuda:3', grad_fn=<SelectBackward0>)
question: ['Is the man drinking his beer?'], responses:['yes']
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(2.2828e-05, device='cuda:3', grad_fn=<DivBackward0>), False: tensor(1.0000, device='cuda:3', grad_fn=<DivBackward0>), 'Execute Error': tensor(5.9605e-08, device='cuda:3', grad_fn=<DivBackward0>)}
ANSWER0=VQA(image=LEFT,question='Are the balls in shadow?')
ANSWER1=EVAL(expr='not {ANSWER0}')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([1, 3, 448, 448])
[('yes', 0.1298617250866936), ('congratulations', 0.12464161604141298), ('no', 0.12445222599225532), ('honey', 0.12437056445881921), ('solid', 0.12422595371654564), ('right', 0.12419889376311324), ('candle', 0.12414264780165109), ('chocolate', 0.12410637313950891)]
[['yes', 'congratulations', 'no', 'honey', 'solid', 'right', 'candle', 'chocolate']]
question: ['Are the balls in shadow?'], responses:['no']
[('no', 0.1313955057270409), ('yes', 0.12592208734904367), ('no smoking', 0.12472972590078177), ('gone', 0.12376514658020793), ('man', 0.12367833016285167), ('meow', 0.1235796378467502), ('kia', 0.12347643720898455), ('no clock', 0.12345312922433942)]
[['no', 'yes', 'no smoking', 'gone', 'man', 'meow', 'kia', 'no clock']]
torch.Size([1, 3, 448, 448]) knan debug pixel values shape
torch.Size([13, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3395
tensor([1.0000e+00, 1.9288e-09, 4.8095e-11, 1.1621e-08, 3.9338e-10, 7.8066e-11,
8.5770e-11, 5.7954e-09], device='cuda:2', grad_fn=<SoftmaxBackward0>)
yes *************
['yes', 'congratulations', 'no', 'honey', 'solid', 'right', 'candle', 'chocolate'] tensor([1.0000e+00, 1.9288e-09, 4.8095e-11, 1.1621e-08, 3.9338e-10, 7.8066e-11,
8.5770e-11, 5.7954e-09], device='cuda:2', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1., device='cuda:2', grad_fn=<DivBackward0>), False: tensor(4.8095e-11, device='cuda:2', grad_fn=<DivBackward0>), 'Execute Error': tensor(-4.8095e-11, device='cuda:2', grad_fn=<DivBackward0>)}
tensor([1.0000e+00, 6.2861e-10, 3.7505e-07, 6.7740e-13, 2.3665e-12, 2.8052e-09,
1.7110e-10, 6.2746e-07], device='cuda:1', grad_fn=<SoftmaxBackward0>)
no *************
['no', 'yes', 'no smoking', 'gone', 'man', 'meow', 'kia', 'no clock'] tensor([1.0000e+00, 6.2861e-10, 3.7505e-07, 6.7740e-13, 2.3665e-12, 2.8052e-09,
1.7110e-10, 6.2746e-07], device='cuda:1', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(6.2861e-10, device='cuda:1', grad_fn=<UnbindBackward0>), False: tensor(1.0000, device='cuda:1', grad_fn=<UnbindBackward0>), 'Execute Error': tensor(9.5367e-07, device='cuda:1', grad_fn=<SubBackward0>)}
ANSWER0=VQA(image=RIGHT,question='How many trucks are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} >= 2')
FINAL_ANSWER=RESULT(var=ANSWER1)
tensor([1.0000e+00, 2.1941e-09, 1.0341e-07, 1.8983e-11, 4.9702e-11, 3.4961e-09,
2.6197e-10, 2.8559e-07], device='cuda:3', grad_fn=<SoftmaxBackward0>)
no *************
['no', 'yes', 'no smoking', 'gone', 'man', 'meow', 'kia', 'no clock'] tensor([1.0000e+00, 2.1941e-09, 1.0341e-07, 1.8983e-11, 4.9702e-11, 3.4961e-09,
2.6197e-10, 2.8559e-07], device='cuda:3', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1.0000, device='cuda:3', grad_fn=<DivBackward0>), False: tensor(2.1941e-09, device='cuda:3', grad_fn=<DivBackward0>), 'Execute Error': tensor(3.5763e-07, device='cuda:3', grad_fn=<DivBackward0>)}
torch.Size([13, 3, 448, 448])
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3398
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3395
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: 3395
question: ['How many trucks are in the image?'], responses:['4']
[('4', 0.12804651361935848), ('5', 0.12521071898947128), ('3', 0.12515925906184908), ('8', 0.12489091845155219), ('6', 0.1245383468146311), ('1', 0.12441141527606933), ('2', 0.12403713327181662), ('11', 0.12370569451525179)]
[['4', '5', '3', '8', '6', '1', '2', '11']]
dynamic ViT batch size: 13, images per sample: 13.0, dynamic token length: 3395
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
tensor([1.0000e+00, 6.7965e-09, 7.8115e-07, 3.7276e-09, 6.2591e-12, 4.9519e-12,
7.4622e-11, 1.8290e-09], device='cuda:0', grad_fn=<SoftmaxBackward0>)
yes *************
['yes', 'congratulations', 'no', 'honey', 'solid', 'right', 'candle', 'chocolate'] tensor([1.0000e+00, 6.7965e-09, 7.8115e-07, 3.7276e-09, 6.2591e-12, 4.9519e-12,
7.4622e-11, 1.8290e-09], device='cuda:0', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1.0000, device='cuda:0', grad_fn=<DivBackward0>), False: tensor(7.8115e-07, device='cuda:0', grad_fn=<DivBackward0>), 'Execute Error': tensor(5.3317e-08, device='cuda:0', grad_fn=<DivBackward0>)}
tensor([9.9892e-01, 1.0324e-03, 5.1399e-05, 2.4590e-10, 5.1476e-08, 7.0480e-09,
1.3191e-09, 5.2135e-09], device='cuda:1', grad_fn=<SoftmaxBackward0>)
4 *************
['4', '5', '3', '8', '6', '1', '2', '11'] tensor([9.9892e-01, 1.0324e-03, 5.1399e-05, 2.4590e-10, 5.1476e-08, 7.0480e-09,
1.3191e-09, 5.2135e-09], device='cuda:1', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1.0000, device='cuda:1', grad_fn=<DivBackward0>), False: tensor(7.0480e-09, device='cuda:1', grad_fn=<DivBackward0>), 'Execute Error': tensor(5.9605e-08, device='cuda:1', grad_fn=<DivBackward0>)}
[2024-10-24 10:48:26,695] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | optimizer_allgather: 1.38 | optimizer_gradients: 0.30 | optimizer_step: 0.32
[2024-10-24 10:48:26,695] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | forward_microstep: 5166.32 | backward_microstep: 8563.14 | backward_inner_microstep: 4943.39 | backward_allreduce_microstep: 3619.68 | step_microstep: 8.01
[2024-10-24 10:48:26,695] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | forward: 5166.32 | backward: 8563.13 | backward_inner: 4943.41 | backward_allreduce: 3619.67 | step: 8.02
100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 4839/4844 [20:07:10<01:12, 14.42s/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
Registering EVAL step
Registering RESULT step
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
ANSWER0=VQA(image=RIGHT,question='How many basil leaves are on the pizza?')
ANSWER1=EVAL(expr='{ANSWER0} >= 1')
FINAL_ANSWER=RESULT(var=ANSWER1)
ANSWER0=VQA(image=RIGHT,question='Is the animal in the image lying down?')
ANSWER1=EVAL(expr='{ANSWER0}')