update readme

.travis.yml

@@ -31,6 +31,9 @@ matrix:
 install:
     - echo $TRAVIS_PYTHON_VERSION
     - pip install -U flake8 opencv-python
+    # Here we use opencv-python to overcome some travis probelms.
+    # Users are not recommended to use opencv-python because it brings issues working on GPU
+    # together with tensorflow.
     - if [ -e requirements.txt ]; then pip install -r requirements.txt;else pip install -r requirements.pip;fi
     - ./tests/install-tensorflow.sh
 

README.md

 # tensorpack
-[![Build Status](https://travis-ci.org/ppwwyyxx/tensorpack.svg?branch=master)](https://travis-ci.org/ppwwyyxx/tensorpack)
 Neural Network Toolbox on TensorFlow
 
+[![Build Status](https://travis-ci.org/ppwwyyxx/tensorpack.svg?branch=master)](https://travis-ci.org/ppwwyyxx/tensorpack)
+
 Docs & tutorials should be ready within a month. See some [examples](examples) to learn about the framework:
 
 ### Vision:
@@ -21,6 +22,7 @@ Docs & tutorials should be ready within a month. See some [examples](examples) t
 ### Speech / NLP:
 + [LSTM-CTC for speech recognition](examples/TIMIT)
 + [char-rnn for fun](examples/char-rnn)
++ [LSTM language model on PennTreebank](examples/PennTreebank)
 
 The examples are not only for demonstration of the framework -- you can train them and reproduce the results in papers.
 

examples/PennTreebank/README.md

@@ -4,7 +4,7 @@
 This example is mainly to demonstrate:
 
 1. How to train an RNN with persistent state between iterations.
-2. How to use a TF reader pipeline instead of a DataFlow for training & inference.
+2. How to use a TF reader pipeline instead of a DataFlow, for both training & inference.
 
 It trains an language model on PTB dataset, basically an equivalent of the PTB example
 in [tensorflow/models](https://github.com/tensorflow/models/tree/master/tutorials/rnn/ptb).

examples/README.md

@@ -25,11 +25,12 @@ Training examples with __reproducible__ and meaningful performance.
 ## Speech / NLP:
 + [LSTM-CTC for speech recognition](TIMIT)
 + [char-rnn for fun](char-rnn)
++ [LSTM language model on PennTreebank](PennTreebank)
 
 
 Note to contributors:
 
-We have a high bar for examples. It needs to satisfy one of the following:
+Example needs to satisfy one of the following:
 + Reproduce performance of a published or well-known paper.
 + Get state-of-the-art performance on some task.
 + Illustrate a new way of using the library that are currently not covered.

tests/install-tensorflow.sh

+#!/bin/bash -e
+
 if [ $TF_TYPE == "release" ]; then
   if [[ $TRAVIS_PYTHON_VERSION == 2* ]]; then
 		TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-${TF_VERSION}-cp27-none-linux_x86_64.whl