Issues
- 2
run simple inference
#92 opened by jdso1988 - 0
如何可视化该模型?
#119 opened by sizhandailiawa - 0
About 3D Patch Linear
#117 opened by JUNJIE99 - 12
How to download "cross_pytorch_model.bin" as pretrained weights used in the project? & problem in DDP
#72 opened by AAUfoa - 0
Evaluation Procedure for Reporting Performance
#116 opened by willyfh - 0
post-pretrain or downstream tasks parameters
#115 opened by 1240446371 - 1
- 3
What do Pair, L, T stand for in the code?
#106 opened by nhw649 - 0
Problem about Multi-GPU-Train architecture
#112 opened by zsnoob - 4
about the results on MSVD
#65 opened by yangxiao2 - 6
Reproduction on LSMDC DataSet
#105 opened by JingXiaolun - 0
ValueError: Parameter config in `CLIP4Clip(config)` should be an instance of class `PretrainedConfig`. To create a model from a Google pretrained model use `model = CLIP4Clip.from_pretrained(PRETRAINED_MODEL_NAME)`
#111 opened by Xiyu-AI - 1
Problem about reproducing the model
#104 opened by sweet132 - 0
Video-to-video retrieval?
#110 opened by wzlxjtu - 0
- 0
train dataset is shuffled regardless of seed
#108 opened by goonbamm - 3
Results on MSRVTT and MSVD
#102 opened by STARK1234 - 1
running the project
#103 opened by giraffeooo - 1
Use my own videos
#100 opened by akshaymanj - 0
- 1
Large performance drop if trained with fp32.
#96 opened by klauscc - 1
loss NaN when training on MSRVTT
#93 opened by TXH-mercury - 1
train files
#98 opened by 1617226214 - 0
- 0
[PAD_TOKEN] is not used, but just adding 0
#97 opened by goonbamm - 0
Evaluation batch
#95 opened by celestialxevermore - 2
About mean_pooling on text sequence
#94 opened by celestialxevermore - 1
- 3
Some new issues about nccl : torch.distributed.elastic.multiprocessing.errors.ChildFailedError :
#78 opened by celestialxevermore - 1
About tightTrans and type embedding
#89 opened by PanZP-CUC - 5
loss becomes nan
#85 opened by fake-warrior8 - 2
- 1
train on DiDeMo
#87 opened by qjyyyy - 4
About Performance
#86 opened by celestialxevermore - 14
- 1
how long have you been training
#82 opened by flyinghpluo - 1
MSVD Weights
#83 opened by ntseng450 - 3
- 5
Data split issue
#81 opened by zhixinma - 1
Finetuning on custom dataset
#76 opened by avinashsai - 2
VIDIOC_REQBUFS: Inappropriate ioctl for device
#77 opened by wangyu0303 - 1
Log stopping problem with no response.
#79 opened by celestialxevermore - 9
subprocess.CalledProcessError
#71 opened by alvinlin1271320 - 1
Question about loss and batch size
#69 opened by JonnyS1226 - 2
How about taking ViViT as video backbone
#68 opened by aries-young - 1
- 1
question about special token "[MASK]"
#66 opened by ForawardStar - 5
torch.distributed.init_process_group(backend="nccl") error and some other errors
#73 opened by celestialxevermore - 2
Question about video input shape
#75 opened by celestialxevermore - 0