通过 Django框架开发网页应用，将纽约市中央公园包含35个属性的 3023条的松鼠观测记录导入导出 数据库， 并且能够在网页上对记录的浏览、增加、删 除和更新; 利用了 MVT 开发框架，并且运用表单 Form以及 GET POST请求 等编写 Html网页脚本展现查询结果;
SHAP（SHapley Additive exPlanations）以一种统一的方法来解释任何机器学习模型的输出
Jupyter Notebook used for writing the article "Black-Box models are actually more explainable than a Logistic Regression" published in Towards Data Science: https://towardsdatascience.com/black-box-models-are-actually-more-explainable-than-a-logistic-regression-f263c22795d
2018/2019/校招/春招/秋招/算法/机器学习(Machine Learning)/深度学习(Deep Learning)/自然语言处理(NLP)/C/C++/Python/面试笔记
小白的Python入门教程实战篇：网站+iOS App源码→ http://t.cn/R2PDyWN 赞助→ http://t.cn/R5bhVpf
Classify MNIST and USPS digit images using ML classifiers
FACE RECOGNITION ---------------- The Yale Face Database contains 165 grayscale images in GIF format of 15 individuals. There are 11 images per subject, one per different facial expression or configuration: center-light, w/glasses, happy, left-light, w/no glasses, normal, right-light, sad, sleepy, surprised, and wink. Your tasks are the following: 1. I have divided image into small blocks and extracted local binary patterns (LBP) from each block. Concatenated all LBP histograms to make a feature vector of an image. 2. Another feature vector is created out of gray levels of integral image. 3. Finally gray levels of image have been used as the last feature vector. 4. After Concatenating all feature vectors. I have Taken four images of each person for testing and the rest as training examples. 5. Using PCA to classify image for one-verses all classification scheme, i have shown results for few images that are selected randomly and reported the accuracy for all testing images using individual feature sets (gray level, integral, and LBP separately) and also for concatenated feature sets.
a simple faces recongnition demo on ORL data sets,using LeNet-5 as CNN structure.
