search

Justin-Tan / generative-compression

TensorFlow Implementation of Generative Adversarial Networks for Extreme Learned Image Compression
MIT License
512 stars 106 forks source link

Unable to train the network #14

Closed AlverGant closed 6 years ago

AlverGant commented 6 years ago

2018-06-26 14:21:25.706113: W tensorflow/core/framework/op_kernel.cc:1318] OP_REQUIRES failed at whole_file_read_ops.cc:114 : Not found: data/leftImg8bit/train/tubingen/tubingen_000082_000019_leftImg8bit.png; No such file or directory Traceback (most recent call last): File "/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py", line 1322, in _do_call return fn(*args) File "/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py", line 1307, in _run_fn options, feed_dict, fetch_list, target_list, run_metadata) File "/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py", line 1409, in _call_tf_sessionrun run_metadata) tensorflow.python.framework.errors_impl.NotFoundError: data/leftImg8bit/train/tubingen/tubingen_000082_000019_leftImg8bit.png; No such file or directory [[Node: ReadFile = ReadFile]] [[Node: IteratorGetNext = IteratorGetNextoutput_shapes=[[?,?,?,3]], output_types=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"]]

Maybe there are training files missing?

Command issued was ~/generative-compression$ python3 train.py -opt momentum --name my_network

Installation script used was: on ubuntu 14.04 server 64 bits

update OS

sudo apt update && sudo apt upgrade

Prerequisites

sudo apt-get install python3-pip python3-dev pkg-config \ libfreetype6-dev gfortran libopenblas-dev python3-tables sudo pip3 install -U setuptools sudo pip3 install matplotlib pip3 install numpy pandas sudo pip3 install --upgrade https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.8.0-cp34-cp34m-linux_x86_64.whl

Compile OpenBLAS

cd "$HOME" && git clone https://github.com/xianyi/OpenBLAS.git cd OpenBLAS git pull make FC=gfortran -j $(($(nproc) + 1)) sudo make PREFIX=/usr/local install

sudo pip3 install seaborn

git clone https://github.com/Justin-Tan/generative-compression.git cd generative-compression python3 train.py -opt momentum --name my_network python3 compress.py -r /path/to/model/checkpoint -i /path/to/image -o path/to/output/image

Justin-Tan commented 6 years ago

You need to download the respective datasets before starting training, see the Data/Setup section in the readme. Under the data directory there are sample .h5 files for the cityscapes dataset and a script for resizing the raw cityscapes leftimg8bit images.

zhiqiang-zhu commented 6 years ago

Hello, are you address the problem? And which datasets you download ?

Justin-Tan commented 6 years ago

Read the data/setup section under the readme, it is explained in depth there.