text
stringlengths
0
1.16k
torch.Size([3, 3, 448, 448])
question: ['How many water buffaloes 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([3, 3, 448, 448]) knan debug pixel values shape
tensor([1.0000e+00, 5.3350e-10, 6.6256e-11, 1.1467e-10, 9.3161e-11, 3.4510e-09,
8.4108e-09, 4.9420e-11], device='cuda:2', grad_fn=<SoftmaxBackward0>)
1 *************
['1', '3', '4', '8', '6', '12', '2', '47'] tensor([1.0000e+00, 1.0907e-07, 1.0467e-08, 6.3488e-09, 4.4218e-10, 9.8894e-10,
9.1808e-10, 1.6916e-10], device='cuda:1', grad_fn=<SoftmaxBackward0>)
2 *************
['2', '3', '4', '1', '5', '8', '7', '29'] tensor([1.0000e+00, 5.3350e-10, 6.6256e-11, 1.1467e-10, 9.3161e-11, 3.4510e-09,
8.4108e-09, 4.9420e-11], device='cuda:2', grad_fn=<SelectBackward0>)
tensor([1.0000e+00, 1.0907e-07, 1.0467e-08, 6.3488e-09, 4.4218e-10, 9.8894e-10,
9.1808e-10, 1.6916e-10], device='cuda:1', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1.0000, device='cuda:1', grad_fn=<DivBackward0>), False: tensor(1.2841e-07, device='cuda:1', grad_fn=<DivBackward0>), 'Execute Error': tensor(0., device='cuda:1', grad_fn=<DivBackward0>)}
{True: tensor(1., device='cuda:2', grad_fn=<DivBackward0>), False: tensor(1.2719e-08, device='cuda:2', grad_fn=<DivBackward0>), 'Execute Error': tensor(0., device='cuda:2', grad_fn=<DivBackward0>)}
tensor([9.9954e-01, 2.6869e-09, 5.4733e-05, 4.0443e-04, 9.6539e-11, 4.0910e-06,
3.1139e-09, 3.7929e-09], device='cuda:3', grad_fn=<SoftmaxBackward0>)
5 *************
['5', '8', '4', '6', '3', '7', '11', '9'] tensor([9.9954e-01, 2.6869e-09, 5.4733e-05, 4.0443e-04, 9.6539e-11, 4.0910e-06,
3.1139e-09, 3.7929e-09], device='cuda:3', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(0.9996, device='cuda:3', grad_fn=<DivBackward0>), False: tensor(0.0004, device='cuda:3', grad_fn=<DivBackward0>), 'Execute Error': tensor(0., device='cuda:3', grad_fn=<DivBackward0>)}
ANSWER0=VQA(image=RIGHT,question='How many cheetahs are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} <= 2')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([11, 3, 448, 448])
question: ['How many cheetahs are in the image?'], responses:['2']
[('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']]
torch.Size([11, 3, 448, 448]) knan debug pixel values shape
tensor([1.0000e+00, 3.5764e-07, 7.7787e-09, 4.8401e-08, 4.3713e-10, 2.4423e-10,
7.5968e-10, 1.3464e-10], device='cuda:3', grad_fn=<SoftmaxBackward0>)
2 *************
['2', '3', '4', '1', '5', '8', '7', '29'] tensor([1.0000e+00, 3.5764e-07, 7.7787e-09, 4.8401e-08, 4.3713e-10, 2.4423e-10,
7.5968e-10, 1.3464e-10], 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(3.6699e-07, device='cuda:3', grad_fn=<DivBackward0>), 'Execute Error': tensor(0., device='cuda:3', grad_fn=<DivBackward0>)}
[2024-10-24 10:48:12,944] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | optimizer_allgather: 1.36 | optimizer_gradients: 0.34 | optimizer_step: 0.32
[2024-10-24 10:48:12,944] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | forward_microstep: 1873.90 | backward_microstep: 14905.44 | backward_inner_microstep: 1760.84 | backward_allreduce_microstep: 13144.36 | step_microstep: 7.65
[2024-10-24 10:48:12,944] [INFO] [logging.py:96:log_dist] [Rank 0] rank=0 time (ms) | forward: 1873.90 | backward: 14905.43 | backward_inner: 1760.93 | backward_allreduce: 13144.27 | step: 7.66
100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 4838/4844 [20:06:56<01:28, 14.71s/it]Registering VQA_lavis step
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
Registering EVAL step
Registering RESULT step
ANSWER0=VQA(image=LEFT,question='How many wine glasses are lined up?')
ANSWER1=EVAL(expr='{ANSWER0} == 3')
FINAL_ANSWER=RESULT(var=ANSWER1)
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
Registering VQA_lavis step
Registering EVAL step
Registering RESULT step
ANSWER0=VQA(image=RIGHT,question='Is the animal in the image on a plain white background?')
FINAL_ANSWER=RESULT(var=ANSWER0)
torch.Size([3, 3, 448, 448])
ANSWER0=VQA(image=RIGHT,question='How many flute illustrations are in the image?')
ANSWER1=EVAL(expr='{ANSWER0} == 2')
FINAL_ANSWER=RESULT(var=ANSWER1)
ANSWER0=VQA(image=RIGHT,question='Does the sleepwear feature a Disney Princess theme on the front?')
ANSWER1=RESULT(var=ANSWER0)
torch.Size([1, 3, 448, 448])
torch.Size([5, 3, 448, 448])
torch.Size([7, 3, 448, 448])
question: ['Does the sleepwear feature a Disney Princess theme on the front?'], 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([1, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 329
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 332
question: ['How many wine glasses are lined up?'], responses:['0']
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 329
[('0', 0.13077743594303964), ('circles', 0.12449813349255197), ('maroon', 0.12428926693968681), ('large', 0.1242263466991631), ('rooster', 0.12409315512763705), ('nuts', 0.12408018414184876), ('beige', 0.1240288472550799), ('bottle', 0.12400663040099273)]
[['0', 'circles', 'maroon', 'large', 'rooster', 'nuts', 'beige', 'bottle']]
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 330
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 329
torch.Size([3, 3, 448, 448]) knan debug pixel values shape
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 329
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 330
question: ['How many flute illustrations are in the image?'], responses:['1']
dynamic ViT batch size: 1, images per sample: 1.0, dynamic token length: 330
[('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']]
tensor([1.0000e+00, 1.3339e-08, 2.4064e-10, 1.1475e-07, 9.4271e-10, 6.6406e-10,
3.0997e-11, 6.7480e-08], device='cuda:0', grad_fn=<SoftmaxBackward0>)
yes *************
['yes', 'congratulations', 'no', 'honey', 'solid', 'right', 'candle', 'chocolate'] tensor([1.0000e+00, 1.3339e-08, 2.4064e-10, 1.1475e-07, 9.4271e-10, 6.6406e-10,
3.0997e-11, 6.7480e-08], device='cuda:0', grad_fn=<SelectBackward0>)
ζœ€εŽηš„ζ¦‚ηŽ‡εˆ†εΈƒδΈΊ: {True: tensor(1.0000, device='cuda:0', grad_fn=<UnbindBackward0>), False: tensor(2.4064e-10, device='cuda:0', grad_fn=<UnbindBackward0>), 'Execute Error': tensor(2.3818e-07, device='cuda:0', grad_fn=<SubBackward0>)}
ANSWER0=VQA(image=LEFT,question='Is the man drinking his beer?')
ANSWER1=EVAL(expr='{ANSWER0}')
FINAL_ANSWER=RESULT(var=ANSWER1)
torch.Size([5, 3, 448, 448]) knan debug pixel values shape
question: ['Is the animal in the image on a plain white background?'], responses:['no']
torch.Size([13, 3, 448, 448])
[('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']]