Open summer1030 opened 5 years ago
Hi @summer1030 I got ValueError: not enough values to unpack (expected 3, got 2) while running train.py. Details are:
Traceback (most recent call last):
File "train.py", line 133, in
Msr vtt dataset have 10000 videos and 20 captions for each video but in this implementation only a video-caption pair for each example in train phase is considered. Therefore in total <= 10000 example for train. someone has seen the same thing???? has anyone changed the code?
Our result is even worse...Bleu_4 is only 18%.
Hi @summer1030 I got ValueError:
Traceback (most recent call last):
File "prepro_vocab.py", line 85, in
when i run:python train.py --gpu 0 --epochs 3001 --batch_size 300 --checkpoint_path data/save --feats_dir data/feats/resnet152 --model S2VTAttModel --dim_vid 4096
it will error:
RuntimeError: size mismatch, m1: [12000 x 2048], m2: [4096 x 512] at C:/w/1/s/tmp_conda_3.7_055457/conda/conda-bld/pytorch_1565416617654/work/atenc\THC/generic/THCTensorMathBlas.cu:
273
How do you solve this problem?
I cannot get good performance on msr-vtt dataset. So i want to know the Bleu_4 you got on this dataset (vgg16 features without c3d features). I have tried different learning rate, dim_hidden, optimizer.... however, not i only get 31.13% . Thank you .