How to write scenes

یک راهنمای کوتاه در زمینه‌ی نوشتن صحنه‌های داستانی

Machine Learning Design Patterns: Solutions to Common Challenges in Data Preparation, Model Building, and MLOps

این رو نخوندم ولی واقعا کتاب خیلی خیلی جالبی به نظر می‌رسه

کتاب‌های هنرستان موسیقی میتونن مرجع خوب و راحتی برای یاد گرفتن اصول موسیقی باشند:

http://chap.sch.ir/category/100

دارم کامپیوترم رو خالی می‌کنم به همین خاطر کلی چیز میفرستم هی 😅

یه مستند جالب پیدا کردم به اسم Coded Bias
When MIT researcher, poet and computer scientist Joy Buolamwini uncovers racial and gender bias in AI systems sold by big tech companies, she embarks on a journey alongside pioneering women sounding the alarm about the dangers of unchecked artificial intelligence that impacts us all. Through Joy’s transformation from scientist to steadfast advocate and the stories of everyday people experiencing technical harms, Coded Bias sheds light on the threats A.I. poses to civil rights and democracy.

https://ezbebin.com/coded-bias-2020/

یه مقاله مروری که یکی دو سال پیش در زمینه‌ی یادگیری فدرال نوشتم.

یه لیست نسبتا خوبی از سوال‌هایی برای طراحی شخصیت‌های داستانی
معمولا جواب دادن به این سوالات، حتی اگه مستقیم به داستانمون ربط نداشته باشه، می‌تونه کمک خیلی زیادی به پخته شدن اون شخصیت توی داستان میشه

سریال "BoJack Horseman" رو که دیدم فهمیدم هیچی بدتر از این نیست که آدم در گذشته غرق باشه و نتونه خودش رو دوست داشته باشه
این سریال جزو بهترین سریال‌های ساخته شده هست که به خوبی افسردگی رو به تصویر کشیده و باعث میشه شما جوری با یه اسب همزاد پنداری کنید که خودتون هم باورتون نشه
دیدنش رو به همه پیشنهاد می‌کنم، چیزای زیادی ازش یاد می‌گیرید
این انیمیشن رو از سایت و ربات مووی کاتیج می‌تونید رایگان دانلود کنید یا آنلاین ببینید. #معرفی‌انیمیشن

من هم پیشنهاد می‌کنم

وقتی برای آیلتس می‌خوندم، این دو تا رو پیدا کرده بودم که گفته بود که تمام سوالات احتمالی برای بخش دوم و سوم speaking رو آورده. احتمالا جدیدش رو هم بتونید پیدا کنید اگه بگردید:
من خودم از این کانال گرفته بودمشون: https://news.1rj.ru/str/afarineshbooks

این هم یه کتاب خیلی خوب در زمینه‌ی طراحی میکروسرویس‌هاست

Data Driven: Creating a Data Culture

این یه کتابچه است که dj patil و hilary mason که از خفن‌های حوزه‌ی داده هستند نوشتند راجع به اینکه داده‌محوری چیه و چطور یه فرهنگ داده‌محور در سازمان داشته باشیم.
این نسخه‌ای که اینجا می‌فرستم نسخه‌ایه که خودم خوندم و هایلایت کردم

Hidden Technical Debt in Machine Learning Systems

یه مقاله خیلی خیلی خوب همراه با هایلایت‌های من :)

Stable diffusion

Prompt:
Game of thrones written by franz kafka

ابزاری برای کشیدن معماری‌های یادگیری عمیق:

https://alexlenail.me/NN-SVG/index.html

BREAKING: White House issues new policy that will require, by 2026, all federally-funded research results to be freely available to public without delay, ending longstanding ability of journals to paywall results for up to 1 year.
Some journal publishers had long fought this open access requirement, fearing it would harm their subnoscription business model...

- ScienceInsider.

خیلی خفنه

رشته توییت:

TwitterThreadUnrollBot:
Today, along with my collaborators at @GoogleAI, we announce DreamBooth! It allows a user to generate a subject of choice (pet, object, etc.) in myriad contexts and with text-guided semantic variations! The options are endless. (Thread 👇)
webpage: https://t.co/EDpIyalqiK
1/N https://t.co/FhHFAMtLwS

Text-to-image diffusion models are extremely powerful and allow for flexible generation of images with complex user captions. One limitation is that controlling the subject’s appearance and identity using text is very hard.
2/N https://t.co/1y7C0hnUr4

By finetuning a model (Imagen here) with few images of a subject (~3-5), a user can generate variations of the subject. E.g. by controlling the environment and context of the subject. Ever wanted to have a high-quality picture of your dog in Paris (no travel required)?
3/N https://t.co/iSb04jcU5R

Our method has some surprising capabilities inherited from large diffusion models. For example it can generate novel art renditions of a subject! Here are some renditions of a specific dog in the style of famous painters.
4/N https://t.co/Oyg2j3SK1B

We can also change semantic attributes of a subject. Re-coloring, chimeras, material changes, etc.
5/N https://t.co/sRd6356mdW

What about accessorization? Given a few images of your pet you could accessorize them with extreme flexibility. Imagination is the limit!
6/N https://t.co/ovNNGvMb1b

We can even do realistic viewpoint changes for some subjects which have a strong class prior! Here are some examples of different viewpoints for a cat. Notice the detailed fur patterns in the forehead are conserved. 🤯
7/N https://t.co/XH2Jki79s3

Finally, our method can generate new images of a subject with different expressions/emotions. Note that the original images of the subject dog here did not exhibit any of these expressions.
8/N https://t.co/Fmv6IJOaJN

One main difficulty in finetuning a diffusion model using few images is overfitting. We tackle this problem by presenting an autogenous class-specific prior preservation loss. More details in the paper.
9/N https://t.co/4bqLW1qDwi

We are able to alleviate overfitting using this approach. We show that finetuning without this loss term leads to accelerated overfitting of subject pose and appearance, or context. This decreases generation variability and incorrect scenes.
10/N https://t.co/cCpR1L2r4m

Thank you for your time. And thank you to all of my collaborators @AbermanKfir, Yuanzhen Li, @jampani_varun, @MikiRubinstein, Yael Pritch. I had an amazing time working on this with you and am looking forward to future uses of this technology and more research!
12/N

We have many other details on the method in the paper. Feel free to check it out!
arxiv: https://t.co/GelJOBDa7H
11/N

We also thank the Imagen team for lending us access to their incredible model. And we deeply thank all of the great people who helped with reviews and feedback (all acknowledged in the paper).
Again, our project website is: https://t.co/EDpIyaDzwS
13/13 (END)

Understanding Diffusion Models: A Unified Perspective

Diffusion models have shown incredible capabilities as generative models; indeed, they power the current state-of-the-art models on text-conditioned image generation such as Imagen and DALL-E 2. In this work we review, demystify, and unify the understanding of diffusion models across both variational and score-based perspectives. We first derive Variational Diffusion Models (VDM) as a special case of a Markovian Hierarchical Variational Autoencoder, where three key assumptions enable tractable computation and scalable optimization of the ELBO. We then prove that optimizing a VDM boils down to learning a neural network to predict one of three potential objectives: the original source input from any arbitrary noisification of it, the original source noise from any arbitrarily noisified input, or the score function of a noisified input at any arbitrary noise level. We then dive deeper into what it means to learn the score function, and connect the variational perspective of a diffusion model explicitly with the Score-based Generative Modeling perspective through Tweedie's Formula. Lastly, we cover how to learn a conditional distribution using diffusion models via guidance.

https://arxiv.org/abs/2208.11970