https://github.com/osmr/utct
Universal Training Control Tools
https://github.com/osmr/utct
Last synced: about 2 months ago
JSON representation
Universal Training Control Tools
- Host: GitHub
- URL: https://github.com/osmr/utct
- Owner: osmr
- License: mit
- Created: 2017-08-04T12:24:31.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2017-12-19T08:25:59.000Z (over 8 years ago)
- Last Synced: 2025-12-26T18:52:14.645Z (6 months ago)
- Language: Python
- Size: 70.3 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Universal Training Control Tools
A set of utilities and wrappers for uniform work with various frameworks such as Tensorflow, TFLearn, Keras, MXNet, Lasagne, and CNTK. This allows us to apply universal methods for hyper parameters tuning, overfitting detection, etc. Also, it simplifies porting/converting models from one framework to another.
## Instructions for using TFLearn-branch on Linux (GPU):
1. Install TensorFlow (https://www.tensorflow.org/install/install_linux):
```
sudo pip install tensorflow-gpu
```
2. Install TFLearn (http://tflearn.org/installation):
```
sudo pip install git+https://github.com/tflearn/tflearn.git
```
3. Install extra python packages:
```
sudo pip install --upgrade pip
sudo pip install opencv-python
sudo pip install bayesian-optimization
sudo pip install pandas
sudo pip install h5py
```
## Instructions for using TFLearn-branch on Win10 x64 (CPU):
1. Install Python3:
- Install Anaconda3 x64 for Windows (https://www.continuum.io/downloads).
- Add the path to Anaconda for ability of running the python from a console:
```
PATH=%PATH%; %USERPROFILE%\Anaconda3
```
2. Install TensorFlow (https://www.tensorflow.org/install/install_windows):
```
pip install --upgrade tensorflow
```
3. Install TFLearn (http://tflearn.org/installation):
```
pip install git+https://github.com/tflearn/tflearn.git
```
4. Install OpenCV (from http://www.lfd.uci.edu/~gohlke/pythonlibs/#opencv):
```
pip install opencv_python-3.2.0+contrib-cp36-cp36m-win_amd64.whl
```
5. Install h5py (http://www.lfd.uci.edu/~gohlke/pythonlibs/#h5py):
```
pip install h5py-2.7.0-cp36-cp36m-win_amd64.whl
```
6. Install extra python packages:
```
pip install bayesian-optimization
pip install pandas
```
## Instructions for using MXNet-branch on Linux (GPU):
1. Install extra python packages:
```
sudo pip install --upgrade pip
sudo pip install opencv-python
sudo pip install bayesian-optimization
sudo pip install pandas
sudo pip install h5py
sudo pip install graphviz
```
2. Install MXNet (http://mxnet.io/get_started/install.html):
```
sudo pip install mxnet-cu80
```
## Instructions for using MXNet-branch on Win10 x64 (CPU):
1. Install Python3:
- Install Anaconda3 x64 for Windows (https://www.continuum.io/downloads).
- Add the path to Anaconda for ability of running the python from a console:
```
PATH=%PATH%; %USERPROFILE%\Anaconda3
```
2. Install MXNet:
- Download MXNet binary packages for Windows x64 (https://github.com/yajiedesign/mxnet/releases).
- Unpack packages into `%LANGS_LIBS%/mxnet` directory.
- Setup environmental variables (from script `setupenv.cmd`, NB: This bugged script slightly breaks existing paths!):
```
set MXNET_HOME=%LANGS_LIBS%\mxnet
set PATH=%PATH%; %MXNET_HOME%\3rdparty\openblas\bin
set PATH=%PATH%; %MXNET_HOME%\3rdparty\gnuwin
set PATH=%PATH%; %MXNET_HOME%\3rdparty\vc
set PATH=%PATH%; %MXNET_HOME%\3rdparty\opencv
set PATH=%PATH%; %MXNET_HOME%\3rdparty\cudart
set PATH=%PATH%; %MXNET_HOME%\3rdparty\cudnn\bin
set PATH=%PATH%; %MXNET_HOME%\lib
```
- Install the MXNet python package by running `python setup.py install`.
- Copy content of folder `%MXNET_HOME%\python\mxnet` to `%USERPROFILE%\Anaconda3\Lib\site-packages\mxnet-0.10.1-py3.6.egg\mxnet`.
3. Install OpenCV (from http://www.lfd.uci.edu/~gohlke/pythonlibs/#opencv):
```
pip install opencv_python-3.2.0+contrib-cp36-cp36m-win_amd64.whl
```
4. Install h5py (http://www.lfd.uci.edu/~gohlke/pythonlibs/#h5py):
```
pip install h5py-2.7.0-cp36-cp36m-win_amd64.whl
```
5. Install Graphviz (http://www.graphviz.org/Download_windows.php) and add path to the program in to the environmental variables:
```
set PATH=%PATH%; C:\Program Files (x86)\Graphviz2.38\bin
```
6. Install extra python packages:
```
pip install bayesian-optimization
pip install pandas
```