git clone git@github.com:emorynlp/doc-classify.git
pip install -r requirements.txt
# install tensorflow depending on the server type (version should be 0.10.0rc0)
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0rc0-cp27-none-linux_x86_64.whl
pip install --upgrade $TF_BINARY_URL
- Python 2.7
- requirements
- boto==2.42.0
- bz2file==0.98
- funcsigs==1.0.2
- gensim==0.13.2
- mock==2.0.0
- numpy==1.11.2
- pbr==1.10.0
- protobuf==3.0.0b2
- requests==2.11.1
- scipy==0.18.1
- six==1.10.0
- smart-open==1.3.5
- tensorflow==0.10.0rc0
- Input file format - note that the text should be tokenized
| line number |
sentiment |
tokenized tweet |
| 1 |
object |
I may be the only one… |
| 2 |
negative |
If Scotland woke… |
cd cnntweets
python train_model.py -v ./w2v-50.bin -t ./trn -d ./dev -l ./lex_config.txt -m ./mymodel
- Input file format - note that the text should be tokenized
| tokenized tweet |
| I may be the only one… |
| If Scotland woke… |
cd cnntweets
python decode.py -m ./mymodel -v ./w2v-50.bin -l ./lex_config.txt -i ./input