bump TF version

README.md

@@ -53,7 +53,7 @@ The components are designed to be independent. You can use Model or DataFlow in
 ## Dependencies:
 
 + Python 2 or 3
-+ TensorFlow >= 0.11
++ TensorFlow >= 0.12
 + Python bindings for OpenCV
 + other requirements:
 ```

examples/ConvolutionalPoseMachines/README.md

 # Convolutional Pose Machines
 
 A script to load and run pre-trained CPM model released by Shih-En. The original code in caffe is [here](https://github.com/shihenw/convolutional-pose-machines-release).
+Reference paper: [Convolutional Pose Machines](https://arxiv.org/abs/1602.00134), Shih-En et al., CVPR16.
 
 ## Usage:
 
@@ -13,7 +14,7 @@ wget https://github.com/shihenw/convolutional-pose-machines-release/raw/master/m
 python -m tensorpack.utils.loadcaffe pose_deploy_resize.prototxt pose_iter_320000.caffemodel CPM-original.npy
 ```
 
-Or you can download the converted model from [model zoo](https://drive.google.com/open?id=0B9IPQTvr2BBkRU8zM2w2ZGh3eU0)
+Or you can download the converted model from [model zoo](https://drive.google.com/open?id=0B9IPQTvr2BBkRU8zM2w2ZGh3eU0).
 
 Run it on an image, and produce `output.jpg`:
 ```
@@ -24,4 +25,4 @@ person has to be in the center of the image (and not too small).
 
 ![demo](demo.jpg)
 
-For a real-time CPM application in tensorpack, check out [Stereo Pose Machines](https://github.com/ppwwyyxx/Stereo-Pose-Machines).
+For a __real-time__ CPM application in tensorpack, check out [Stereo Pose Machines](https://github.com/ppwwyyxx/Stereo-Pose-Machines).

examples/DoReFa-Net/README.md

@@ -14,7 +14,7 @@ It's provided in the format of numpy dictionary, so it should be very easy to po
 
 To use the script. You'll need:
 
-+ TensorFlow >= 0.11
++ TensorFlow >= 0.12
 
 + OpenCV bindings for Python
 

examples/ResNet/README.md

@@ -2,7 +2,7 @@
 ## imagenet-resnet.py
 
 __Training__ code of pre-activation ResNet on ImageNet. It follows the setup in
-[fb.resnet.torch](https://github.com/facebook/fb.resnet.torch) and gets similar performance (with much fewer lines of code).
+[fb.resnet.torch](https://github.com/facebook/fb.resnet.torch) (except for the weight decay) and gets similar performance (with much fewer lines of code).
 Models can be [downloaded here](https://goo.gl/6XjK9V).
 
 | Model              | Top 5 Error | Top 1 Error |