摘要：
根据训练好的Transformer模型，得到注意力矩阵，并对注意力进行可视化 首先安装：tensorflow 1.13.1 + tensor2tensor 1.13.1 可视化，请在Jupyter notebook中运行。该代码根据tensor2tensor/tensor2tensor/visual 阅读全文
摘要：
这里给出论文的SupContrast: Supervised Contrastive Learning的损失函数Tensorflow版本，代码改自：https://github.com/HobbitLong/SupContrast 损失文件losses.py """ Author: Yonglong 阅读全文
摘要：
安装sklearn后：import出现以下错误： ImportError: /lib64/libstdc++.so.6: version `CXXABI_1.3.9' not found (required by anaconda3/lib/python3.6/site-packages/scipy 阅读全文
摘要：
深度学习在自动问答系统中的应用 https://www.slideshare.net/hustwj/ss-47489370 AmazonQA: A Review-Based Question Answering Task https://arxiv.org/pdf/1908.04364.pdf 阅读全文
摘要：
https://www.k4all.org/wp-content/uploads/2017/09/WPOC2017_paper_6.pdf A Framework for Automated Fact-Checking for Real-Time Validation of Emerging Cla 阅读全文
摘要：
Light GBM demand-forecasting https://www.kaggle.com/ashishpatel26/light-gbm-demand-forecasting Hourly Time Series Forecasting using XGBoost https://ww 阅读全文
摘要：
写入Tfrecord print("convert data into tfrecord:train\n") out_file_train = "fudan_mtl/data/ace2005/bn_nw.train.tfrecord" writer = tf.python_io.TFRecordWr 阅读全文
摘要：
On the convergence analysis of gravitational search algorithm Convergence analysis and performance of an improved gravitational search algorithm 阅读全文
摘要：
import osos.environ['CUDA_VISIBLE_DEVICES'] = "0"from mlp_clf import MLPClassifierimport numpy as npimport tensorflow as tffrom sklearn.datasets impor 阅读全文
摘要：
https://pp.pcbaby.com.cn/group/339966.html 阅读全文