Atze00 MoViNet-pytorch issues

Atze00 / MoViNet-pytorch

MoViNets PyTorch implementation: Mobile Video Networks for Efficient Video Recognition;

MIT License

258 stars 50 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Training on the custom dataset

#41 WAL-l opened 8 months ago
1
Training

#40 giovi1 opened 1 year ago
0
Kinetics400/600

#39 LOTEAT opened 1 year ago
0
The parameters that trained on Charades.

#38 JewelChen2019 opened 1 year ago
0
weight

#37 1111554 opened 1 year ago
0
got wrong results during test

#36 poincarelee opened 2 years ago
0
need to process HMDB51 dataset?

#35 poincarelee closed 2 years ago
0
How can we access the stream buffer?

#34 bf2harven opened 2 years ago
1
fix: missing ! in pip install

#33 WurmD opened 2 years ago
0
There seems no implementation of positional_encoding

#32 zhouruirui1998 closed 2 years ago
4
。

#31 zhouruirui1998 closed 2 years ago
0
F.ToFloatTensorInZeroOne not exist

#30 haowei2020 closed 2 years ago
2
Very low validation accuracy with pretrained models!

#29 ghazalehtrb closed 2 years ago
1
Update models.py

#28 patrick620 opened 2 years ago
0
Kinetics 400 models

#27 lovelyczli opened 2 years ago
1
Modifying for binary classification

#26 ekuluke opened 2 years ago
2
Validation Loss did not decrease in the HMDB51 notebook?

#25 neonampv opened 2 years ago
9
Test model based on 'evaluate_stream' is ok, but do inference frame by frame is very different?

#24 erwangccc closed 3 years ago
2
Tips for Implementing a3 ,a4,a5 movinets streaming version

#23 papasanimohansrinivas opened 3 years ago
3
Neural network arch displayed by Netron is wrong

#22 erwangccc closed 3 years ago
7
why don't you use 'T.Normalize' when you train HMDB51?

#21 erwangccc closed 3 years ago
1
Using MoViNet in a dataset with variable-length videos

#20 alexlopezcifuentes closed 3 years ago
1
how to train customer data

#19 wojiaohumaocheng closed 3 years ago
0
Error in running movinet-pytorch to onnx converted model

#18 papasanimohansrinivas closed 3 years ago
4
About frames number of dataset

#17 erwangccc closed 3 years ago
4
Update README.md

#16 dragonpod closed 3 years ago
0
Model not compatible with TorchScript conversion (via torch.jit.script)

#15 brunoabbate opened 3 years ago
4
MoViNet-A0 block 2 should have 40 out channels

#14 SMHendryx closed 3 years ago
2
Accuracy drop when using causalConv

#13 yuridekim closed 3 years ago
2
Atze00 patch 1

#12 Atze00 closed 3 years ago
0
accuracy drop

#11 zongzhesun closed 3 years ago
1
multi-clip sampling in dataloader

#10 zongzhesun closed 3 years ago
0
When pretrained stream version will be available?

#9 Atze00 closed 3 years ago
5
stream inference

#8 riyaj8888 closed 3 years ago
5
transform for test video in movinet_tutorial.ipynb

#7 Billchen1993 closed 3 years ago
1
accuracy

#6 YongdongTan closed 3 years ago
1
Does this implementation supports Streaming Buffer and Causal Conv?

#5 IgorSondors closed 3 years ago
1
pretrain dataset?

#4 pacifinapacific closed 3 years ago
2
Training code

#3 Atze00 closed 3 years ago
5
MoViNets are now open source

#2 Hyperparticle closed 3 years ago
7
a6 implementation

#1 seandatasci closed 3 years ago
1