word2vec, sentence2vec, machine reading comprehension, dialog system, text classification, pretrained language model (i.e., XLNet, BERT, ELMo, GPT), sequence labeling, information retrieval, information extraction (i.e., entity, relation and event extraction), knowledge graph, text generation, network embedding
Natural Language Processing projects, which includes concepts and scripts about:
gensim
, fastText
and tensorflow
implementations. See Chinese notes, 中文解读
doc2vec
, word2vec averaging
and Smooth Inverse Frequency
implementationstensorflow LSTM
(See Chinese notes 1, 中文解读 1 and Chinese notes 2, 中文解读 2)fastText
implementationHMM Viterbi
implementations. See Chinese notes, 中文解读
tensorflow
implementation. See Chinese notes, 中文解读
Parallelization [1]
Long-range dependency [1]
N * [Convolution + skip-connection]
Position [1]
CNNs
Solutions
Semantic features extraction [2]
Data
S
to increase target task/domain T
S
has a zero/one/few instances, we call it zero-shot, one-shot, few-shot learning, respectivelyModel
Learning (change loss definition)
conda install tensorflow-gpu