desktoop Goto Github PK

audio-emotion-recognition

Directly from voice, recognise speaker emotion, intensity, & sentiment in speaker utterances.

bert-predict

simplify the prediction process for a finetuned bert model

elmo-keras

Re-implementation of ELMo on Keras

gold-miner

🥇掘金翻译计划，可能是世界最大最好的英译中技术社区，最懂读者和译者的翻译平台：

hanjiayi.github.io

welcome~

kashgari

Kashgari is a Production-ready NLP Transfer learning framework for text-labeling and text-classification, includes Word2Vec, BERT, and GPT2 Language Embedding.

keras-bert

Implementation of BERT that could load official pre-trained models for feature extraction and prediction

nlp-chinese-sentences-punctuating

Make punctuations for Chinese Sentences with the normal RNN and LSTM.

nlpcc2013_cross-lingual_sentiment_classification

oneflow-model-compression

sentiment-analysis

In this work, we aim to predict the intensity of user emotion (anger, fear, sadness and joy) contained in a sentence. This problem is more challenging than the task of classification into sentiment categories as the degree of emotion extraction needs deeper level of granularity. We have approached the solution with various techniques such as tfidf, lexicon based approaches, Word embeddings and LSTM networks. In this process, we found a few interesting observations regarding the behavior of different approaches. Word embedding approaches perform better with the inclusion of sentiment lexicon knowledge and also the results obtained from the experiments show the inefficiency of tfidf approach for regression tasks. The observation that the emotion learning is not independent, has motivated us to develop a new architecture for leveraging the power of shared learning. The shared- LSTM architecture developed in this work eliminates the problem of insufficient data to train a particular emotion category by exploring mutual and interactive learning among the emotion categories. Shared- LSTM outperforms the other approaches significantly and also can be extended to various other tasks where there is a need to explore the dependencies among different categories.

speech_recognition

Speech recognition module for Python, supporting several engines and APIs, online and offline.

synchronized-batchnorm-pytorch

Synchronized Batch Normalization implementation in PyTorch.

trumptradewarsentiment

measure the intensity of comments about trade war

waimaiopinionminer

细粒度情感分析repository1：Wai Mai Opinion Miner，细粒度情感分析系统GUI demo。

zaidalyafeai.github.io

Implementation of web friendly ML models using TensorFlow.js