Florentine Stories
Some entries in my journals during my travel to Florence. They sparked my thoughts when I saw them, so I decided to keep them down in my journal.
Dante Alighieri was banished from Florence in 1302. In 2008, the council of Florence revoked and apologized for his banishment.
Midway upon the journey of our life I found myself within a forest dark, For the straightforward pathway had been lost. (Inferno Canto 1, Dante Alighieri)
In 1481, the Dominican friar Girolamo Savonarola started preachi ...
苏东坡传
有些书读来是为消遣,有些书读来是为励志,又有些书读来是为益智,读完林语堂的苏东坡传,似乎很难一言蔽之其为何用。这本书讲的是苏东坡的生平,从少年得志,到从政造福一方,又到落魄被贬,最后客死他乡。读者在许多细节中了解了这个千年前古人的一生。为此,林语堂参考了大量苏东坡的书信,札记,诗词,从而把苏东坡带到了读者面前。读者也由此认识了一个高度复杂的人 - 他有李白的才气,杜甫的仁义,辛弃疾的豪情 - 但是似乎很难找到一个形容词来评价他。
林语堂这样评价他,“苏东坡是个秉性难改的乐天派,是悲天悯人的道德家,是黎民百姓的好朋友,是散文作家,是新派的画家,是伟大的书法家,是酿酒的实验者,是工程师,是假道学的反对派,是瑜伽术的修炼者,是佛教徒,是士大夫,是皇帝的秘书,是个饮酒成瘾者,是个心肠慈悲的法官,是政治上的坚持己见者,是月下的漫步者,是诗人,是生性诙谐爱开玩笑的人。”
可是到头来,若是在中国人面前提到苏东坡,总是会引起人亲切敬佩的微笑,这也许最能概括苏东坡的一切了。我们对于苏东坡的印象并不是一个仰之弥高的伟人,更像是一个我们愿意交好的朋友。
如果苏东坡的人生是一帆风顺的,他或许不会成为我们记忆 ...
The Fall by Albert Camus
A very intense read and very different than other writings of Camus. It tells the story of Clamence, an attorney in Amsterdam who begrudges about everything, including his very existence. In many ways this stream-of-consciousness monologue is in a similar vein with The Notes from the Underground. But while Dostoevsky’a underground man presents himself as a hermit who has cut all ties to the human world, Camus’ Clamence takes pride in his ability to navigate through the human affairs. Yet he is n ...
Klara and the Sun
Finished Klara and the Sun by Kazuo Ishiguro. The story is a bit dry at the beginning, filled with internal monologues and conversations that seem irrelevant. I almost gave up on reading it. But slowly all the details piece together into a compelling story that speaks to the intricacy of human relations in an alienated world. Here are a couple thoughts I had while reading.
The first question I kept thinking over is who (or what) Klara really is. The more I try to piece her image together through ...
Finger-Printing User Through Smart Phone Sensor Data
TL;DRThis is a data project in which I investigated the possibility of using smart phone sensor data for finger-printing user (identifying the device). It turned out that sensor data can be easily used to finger-print users with high accuracy. This is a huge concern for digital privacy and security.
To read the whole notebook, visit https://www.kaggle.com/jiahuichen23/math-104-final-project.
Fortress Besieged (围城)
I finished Fortress Besieged (围城), a Chinese novel by Qian Zhongshu. The novel tells the story of Fang Hongjian, a Chinese student who traveled abroad to study in the 1930s, got a fake degree, and returned to China after using up all the money. His yearning for romantic love drove him to pursue different women throughout the novel. Most failed, and he ended up marrying Sun Roujia, a colleague he met while teaching in a university, through a series of coincidences and dramas. The marriage deteri ...
Illustrated BERT Fine-Tuning Process
Intro - BERT and FinetuningIf there is ever an NLP buzzword for the past few years, it should be BERT. You can see it in most blogs, products, and even research in the field of Natural Language Processing(NLP). BERT (Bi-Directional Encoder Representations from Transformers) is a pre-trained NLP model developed by Google back in 2016. The idea behinds BERT follows the general philosophy of NLP work in recent decades - generating embeddings that contain rich relations between words in human langu ...
北京和泉州
北京 2020.01
泉州 - 2019.08
How I Created Chinese Word Embeddings with Word2vec
IntroWord2Vec is a common tool in Natural Language Processing (NLP) for generating word embeddings from text. Word embeddings are essentially vectors in a higher dimensional space - by translating words into these vectors, we can quantitatively represent the words within their context. The hope is to group similar words together just by looking at their context. In this blog post, I will walk through how I used the Word2Vec model to generate word embeddings using Gensim’s Word2Vec model. I will ...
7784 KM of Trans-Siberian Railway | 7784公里西伯利亚铁路
在这半个月里,我们由东向西,跨越欧亚大陆,从北京到莫斯科。蒙古草原上的大雨,俄罗斯边境的漫天繁星,西伯利亚小镇上的弥撒,圣彼得堡中的文学印记,这些都成为了我这个夏天不可磨灭的记忆。借此机会分享,慰情聊胜无。
Mongolia & Siberia
Lake Baikal
St Pertersburg
The Golden Ring
Moscow