Recently, a friend introduced me to HuggingTweets: A product demo built by Boris Dayma of HuggingFace that allows you to train a language model based on your favorite Twitter account and to generate new tweets based on a given prompt. As a person who will take any test and explore any technology that promises to give me some insight into my personality or myself (from the whole gamut of personality tests available online and in my alma mater’s career development center to the most questionable of astrological insights), I obviously chose to create a model based on my own tweets.
Quickly, everything becomes
my cat asleep in my pillowcase
and having nothing left to do
but watch the stillness of the room
in the doldrum white-noise hum
of several rooftop HVAC units
so tireless in their endless fight
against the climate. My hair lifts
with the reach of waiting and
my knuckles still burning from
the spaces made amid (paper-
thin) walls too weak to carry
so many knickknacks,
so much free time,
all the false starts.
But when the past dissipates again becomes a muted thing and the ring on my finger reminds how none will change except —…
I was initially attracted to data science because of its incredible potential to solve a plurality of complex problems, and I think the same is true of many people working as data scientists today. Given almost any problem, a dataset can be assembled through which insight can be gleaned and a solution developed. Quite appropriately, therefore, data scientists are a wildly diverse bunch that is perhaps best characterized by an insatiable curiosity and a passion for problem-solving — and I think that is the field’s greatest strength.
Despite the diversity of my peers, however, as I reenter the job market…
Henry was mystified by my new pages. Was it more than brocade, he asked, more than beautiful language?… Then he said, as everybody else has said, “Well, you should give a clue, you should lead up to it; we are thrown into the strangeness unexpectedly. This must be read a hundred times.”
“Who is going to read it a hundred times?” I said sadly. But then I thought of Ulysses and the studies which accompany it… Here, I faced my lifelong problem. I wanted to go on in that abstract, intense way, but could anyone bear it? Hugo understood it…
Like the northeastern girl, there are thousands of girls scattered throughout the tenement slums, vacancies in beds in a room, behind the shop counters working to the point of exhaustion. They don’t realize how easily substitutable they are and that they could just as soon drop off the face of the earth. Few protest and as far as I know they never complain since they don’t know to whom. Does this whom exist?
The Hour of the Star by Clarice Lispector (translated by Benjamin Moser)
I have a strange, voyeuristic fascination with reading the responses to tweets and other forms…
Note: Although this article is written as a five-part series, it is not necessary to read each part in order, nor is it necessary to read all parts, as each part has been written as a standalone piece.
In Part III of this series (linked below), inspired by a quote from Deep Learning with PyTorch (further below), to build toward propositions for (1) verified humans on social media platforms and (2) transparency regarding the use of language models to generate text, I discussed the possibility of a machine forming a thesis by exploring both mechanical desires and algorithmically generated worldviews.
…
Note: Although this article is written as a five-part series, it is not necessary to read each part in order, nor is it necessary to read all parts, as each part has been written as a standalone piece.
In Part II of this series (linked below), inspired by a quote from a guidebook called Deep Learning with PyTorch (further below), I explored the changing nature of communication with human language to build toward propositions for (1) verified humans on social media platforms and (2) transparency regarding the use of language models to generate text.
In this article, to answer the…
Note: Although this article is written as a five-part series, it is not necessary to read each part in order, nor is it necessary to read all parts, as each part has been written as a standalone piece.
In Part I of this series (linked below), inspired by a quote from a guidebook called Deep Learning with PyTorch (further below), I embarked on a journey toward propositions for (1) verified humans on social media platforms and (2) transparency regarding the use of language models to generate text by discussing the unimportance of the semantic classification of the actions and capabilities…
Note: Although this article is written as a five-part series, it is not necessary to read each part in order, nor is it necessary to read all parts, as each part has been written as a standalone piece.
As I read, I often get lost in the text, my thoughts diverging from the author’s words toward something related yet different. This process seems almost like a conversation between the author and me, with a person’s words provoking my response. At times, I enjoy this aspect of reading, but more often, it is quite distracting.
Still, if I find the thoughts…
An interesting concept from literary theory states that if a reader wants to make sense of a text, then he will find an interpretation of that text that is consistent with his own world view, or perhaps more precisely, with his view of the world he supposes the text to concern. Oftentimes, to fulfill such a desire requires the reader to fill gaps in his own knowledge, as well as gaps in the logic or rhetoric of the writer by reading between the lines. …
MSc Analytics ’16 @ Georgia Tech | BSc ChemEng ’15 @ Drexel U | @danielleboccell