gudongfeng / 3d-densenet Goto Github PK

Hi,
Is there any published paper about your code?

Can you share the pre-trained model on UCF-101 to test and reproduce the results?

Dear Dongfeng,
I get an error as : IndexError: index 5 is out of bounds for axis 1 with size 5 when I run the program. I would be most obliged if you can help me to solve this error. Thank you

def _add_block(self, inputs, growth_rate, layers_per_block):
    for layer in range(layers_per_block):
        with tf.variable_scope("layer_%d" % layer):
            return self._add_internal_layer(inputs, growth_rate)

This return in the loop might make only one layer in each block. you know, when the 'layer_0' was created, the 'return' will be used to stop this function.
I use tensorboard to demonstrate this guess

Hi~ when i run"sudo python run_dense_net_3d.py --train --test -ds /home/scu508/llworkspace/UCF101",and the output :I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcublas.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcudnn.so.5 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcufft.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcurand.so.8.0 locally
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:885] Found device 0 with properties:
name: TITAN X (Pascal)
major: 6 minor: 1 memoryClockRate (GHz) 1.531
pciBusID 0000:06:00.0
Total memory: 11.90GiB
Free memory: 3.07GiB
I tensorflow/core/common_runtime/gpu/gpu_device.cc:906] DMA: 0
I tensorflow/core/common_runtime/gpu/gpu_device.cc:916] 0: Y
I tensorflow/core/common_runtime/gpu/gpu_device.cc:975] Creating TensorFlow device (/gpu:0) -> (device: 0, name: TITAN X (Pascal), pci bus id: 0000:06:00.0)
scu508@scu508:~/llworkspace/3d-DenseNet-master$
i do not know how to save it?

When I use 'convert_images_to_list.sh' to create list files, it says 'jot : not found command' .
How to solve it ?

Hi,
Thank you for your interesting code. Would you please upload your model architecture? (i.e., visualize your model's architecture). This help me to re-implement your models via other deep learning frameworks and develop your model.

python run_dense_net_3d.py --train -ds /home/user/jiangsudaxuematch/daima/UCF101
when I run this instruction , wrongs occur . wrong list in log.txt
Params:
weight_decay: 0.0001
bc_mode: False
should_save_logs: True
keep_prob: 1.0
gpu_id: 0
reduction: 1.0
dataset: /home/user/jiangsudaxuematch/daima/UCF101
model_type: DenseNet3D
depth: 20
train: True
should_save_model: True
test: False
renew_logs: False
total_blocks: 3
nesterov_momentum: 0.9
growth_rate: 12
Train params:
batch_size: 10
n_epochs: 70
queue_size: 300
reduce_lr_epoch_2: 55
sequence_length: 16
validation_set: True
num_classes: 5
crop_size: (150, 100)
reduce_lr_epoch_1: 30
initial_learning_rate: 0.1
validation_split: None
normalization: std
Start thread: train data preparation ...
Start thread: validation data preparation ...
Initialize the model..
Build DenseNet3D model with 3 blocks, 5 composite layers each.
Reduction at transition layers: 1.0

Traceback (most recent call last):
File "run_dense_net_3d.py", line 157, in
model = DenseNet3D(data_provider=data_provider, **model_params)
File "/home/user/jiangsudaxuematch/daima/models/dense_net_3d.py", line 88, in init
self._initialize_session()
File "/home/user/jiangsudaxuematch/daima/models/dense_net_3d.py", line 107, in _initialize_session
self.saver = tf.train.Saver(tf.global_variables(), max_to_keep=0)
File "/home/user/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1051, in init
self.build()
File "/home/user/.local/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1072, in build
raise ValueError("No variables to save")
ValueError: No variables to save.

How can I solve this problem?

python run_dense_net_3d.py --train -ds /home/venkat/3d-DenseNet/UCF-101Resized

/home/venkat/anaconda2/lib/python2.7/site-packages/h5py/init.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type.
from ._conv import register_converters as _register_converters

Can you please provide the scripts for data preprocessing on the MERL dataset.

Thanks

Please i need to know what is the difference between the light and the general version?
To use lite or general i have to download it from the branch?

Hello, i am a students from China, i found your code is suitable to my research, but you haven't submit your pretrained model, by the way, my computer don't have good GPU to train it, so i wonder if you could update your pretrained models? i will be great appreciate. thank you.

How long does it take training on UCF101?