Quick cheatsheet

Install tflearn if you haven't downloaded[1]:

pip install --user tflearn

Install selectivesearch if you haven't downloaded[2]:

source ~/tensorflow/bin/activate
pip install --user cython
pip install --user selectivesearch

Install sklearn if you haven't downloaded[3]:

pip install -U scikit-learn

How to run:

The work trains with the 17 flowers dataset. Download the project from here and use the 17flowers.tar.gz file's data as the training data. (should extract the file)

git clone https://github.com/edwardbi/DeepLearningModels.git

cd DeepLearningModels/RCNN/
wget http://www.robots.ox.ac.uk/~vgg/data/flowers/17/17flowers.tgz
tar zxvf 17flowers.tgz
nohup python train_alexnet.py &>tarin_log &
nohup python fine_tune_RCNN.py &>finetune_log &

Choose any image as a testing input image and rename it as testimg7.jpg

nohup python RCNN_output.py &>ouput_log &

After this step you can see the predict result in ouput_log file.

To visualize network and performance, you could check the result of alex-net on tensorboard by following steps:

tensorboard --logdir='output'

And use your browser to check your result.

example output:

I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcublas.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcudnn.so.5.0.5 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcufft.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcurand.so.8.0 locally
Starting TensorBoard 23 on port 6006

IP: 140.114.89.60:6006