eval_split
AlaaHassanAli opened this issue · comments
@LuoweiZhou when i train with this command (th train_new.lua -finetune_cnn_after 10000 -max_iters 60000 -cnn_weight_decay 0.001 -cnn_learning_rate 0.00001 -learning_rate_decay_every 10000 -learning_rate_decay_start 10000) , this error appers
DataLoader loading json file: coco/cocotalk.json
vocab size is 8102
DataLoader loading h5 file: coco/cocotalk.h5
read 82783 images of size 3x256x256
max sequence length in data is 16
assigned 82783 images to split train
[libprotobuf WARNING google/protobuf/io/coded_stream.cc:537] Reading dangerously large protocol message. If the message turns out to be l
arger than 1073741824 bytes, parsing will be halted for security reasons. To increase the limit (or to disable these warnings), see Coded
InputStream::SetTotalBytesLimit() in google/protobuf/io/coded_stream.h.
[libprotobuf WARNING google/protobuf/io/coded_stream.cc:78] The total number of bytes read was 553432081
Successfully loaded model/VGG_ILSVRC_16_layers.caffemodel
conv1_1: 64 3 3 3
conv1_2: 64 64 3 3
conv2_1: 128 64 3 3
conv2_2: 128 128 3 3
conv3_1: 256 128 3 3
conv3_2: 256 256 3 3
conv3_3: 256 256 3 3
conv4_1: 512 256 3 3
conv4_2: 512 512 3 3
conv4_3: 512 512 3 3
conv5_1: 512 512 3 3
conv5_2: 512 512 3 3
conv5_3: 512 512 3 3
fc6: 1 1 25088 4096
fc7: 1 1 4096 4096
fc8: 1 1 4096 1000
converting first layer conv filters from BGR to RGB...
total number of parameters in LM: 11455399
total number of parameters in CNN: 136358208
constructing clones inside the LanguageModel
iter 0: 9.027307
/root/torch/install/bin/luajit: ./misc/DataLoader.lua:84: split val not found.
stack traceback:
[C]: in function 'assert'
./misc/DataLoader.lua:84: in function 'getBatch'
train_new.lua:196: in function 'eval_split'
train_new.lua:321: in main chunk
[C]: in function 'dofile'
/root/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00405d50
any help ?? (note : in preprocessing i put num of val & test images =0)
The # of val and test should be put to 5000, respectively.
thanks it works ^_^