Large encoder and decoder model

[deepcode123]

Hi,
I trained both encoder and decoder, I got the large model.net. Encoder model is about 400M; decoder model is 19M. Both of these are much larger than yours, are these normal or something wrong?

First, I trained encoder:
th run.lua --dataset cs --datapath /home/janice/Pictures/Cityscapes --model models/encoder.lua --save /home/janice/Documents/ENet-training/train/trained/encoder_model/ --imHeight 256 --imWidth 512 --labelHeight 32 --labelWidth 64 --cachepath /home/janice/Documents/ENet-training/train/trained/encoder_dataset_cache/ --nGPU 1 --learningRate 5e-4 --weightDecay 2e-4 --batchSize 5

Then I trained decoder:
th run.lua --dataset cs --datapath /home/janice/Pictures/Cityscapes/ --model models/decoder.lua --save /home/janice/Documents/ENet-training/train/trained/decoder_model/ --imHeight 256 --imWidth 512 --labelHeight 256 --labelWidth 512 --cachepath /home/janice/Documents/ENet-training/train/trained/decoder_dataset_cache/ --nGPU 1 --learningRate 5e-4 --weightDecay 2e-4 --batchSize 5 --CNNEncoder /home/janice/Documents/ENet-training/train/trained/encoder_model/model-best.net

Thanks!!

[Mina1368]

Hi,

Would you please say how much memory on GPU it is required to run the model?

Thank You

[deepcode123]

hi, @Mina1368
about 820MB when I run the trained model
Thanks!

[culurciello]

save model as:

torch.save('filename.t7', model:clearState())

to make model take less disk space for deployment

[deepcode123]

hi, @culurciello
I saved model in this way:

torch.save(filename, model:clearState():get(1))

or in this way:

model:clearState()
torch.save(filename, model)

both of these are the same for encoder, about 400M

I also tried "torch.save('filename.t7', model:clearState())"
but it's still the same, about 400M

Do you have any idea?
Thanks!

[FrancisGee]

When I want to trian ENet in Cityscape Dataset or Camvid Dataset, how much GPU memory should I use