optimize import of seqToseq_net_v2 for book

change sumCost to sum

Complete documentation generation for v2 model configuration

word2vec v2 api

move paddle.reader.batch to paddle.batch

dataset reader for wmt14

Start doing v2 documentation

example usage:

import paddle.v2 as paddle

if __name__ == '__main__':
    dict_en, dict_fr =	paddle.dataset.wmt14.build_dict()
    train = paddle.dataset.wmt14.train(dict_en, dict_fr)
    test = paddle.dataset.wmt14.test(dict_en, dict_fr)

    total_train = 0
    for i in train():
        total_train += 1

    total_test = 0
    for i in test():
        total_test += 1

    print total_train, total_test

Rename Argument::sumCost to Argument::cost since Argument should not …

remove inferencer and Inference from __All__, since infer is sufficen…

expose build_dict in imikolov dataset, fix bug that len(word_dict) is…

remove ITrainer interface since parameters for train is already getti…

change argument name of data_type.integer_value/integer_value_sequenc…

remove inferencer and Inference from __All__, since infer is sufficent, they are implementation detail.

rename inferencer.py to inference.py

cost is Argument, but argument does not have to be cost.

Use the sequence_conv_pool define inside the networks.py

Redirect code coverage output into /dev/null

* Change build process to generate v2 documentation
* Add directory for v2 api
* Correct sphinx index soft-link

v2 api for seqToseq(machine translation)

Disable test_ProtoServer on Travis CI

dataset: print download progress

fix merge conflict