Common mistakes when carrying out machine learning and data science
If we perform an EDA on the whole dataset (including the test dataset) and use that info to come up with some feature engineering than we could also suffer from data leakage.
There are many ways of imputing the values, mean, or median. It is up to you how to do it but make sure to calculate the imputation statistics only on the training data to avoid data leakage of your test set.
I would suggest also more generally:
Support Vector Classifier
Great overview of internals.
Visualizing Machine Learning Thresholds to Make Better Business Decisions
there is no need to spend years obtaining a PhD in order to practise deep learning. Creating software that learns can be taught as a craft, not as a high intellectual pursuit to be undertaken only in an ivory tower
Under SSPL, users can still change and redistribute the software, but it explicitly states that if a person or a company wants to publicly offer MongoDB as a service, they must open source that service -- meaning they must make the code available to everyone for free. Otherwise, they must obtain a commercial license.
Fastai is the first deep learning library to provide a single consistent interface to all the most commonly used deep learning applications for vision, text, tabular data, time series, and collaborative filtering. This is important for practitioners, because it means if you’ve learnt to create practical computer vision models with fastai, then you can use the same approach to create natural language processing (NLP) models, or any of the other types of model we support
What gives researchers hope is the striking resemblance the olfactory system’s structure bears to other regions of the brain across many species, particularly the hippocampus, which is implicated in memory and navigation, and the cerebellum, which is responsible for motor control. Olfaction is an ancient system dating back to chemosensation in bacteria, and is used in some form by all organisms to explore their environments.
“It seems to be closer to the evolutionary origin point of all the things we’d call cortex in general,” Marblestone said. Olfaction might provide a common denominator for learning. “The system gives us a really conserved architecture, one that’s used for a variety of things across a variety of organisms,” said Ashok Litwin-Kumar, a neuroscientist at Columbia. “There must be something fundamental there that’s good for learning.”
TODOS devem olhar para fora.
It’s important to note that a chapter lead is not a manager of the team, but is a manager of a domain that spans across teams
your main task as a chapter lead is to make sure your chapter members are happy. Sounds shallow? Easy? Think one more time!! This is the hard and most crucial part. It is your responsibility to help your chapter members with their development, both hard skills as well as the soft ones. Help them exploit their strengths and handle their weaknesses. Make sure they are properly equipped to perform their job the best they can. A happy chapter member is the best compliment a chapter lead can get!
With this model, changing teams does not mean changing your manager, and dissolving a team doesn’t leave a manager looking for a new role
One can ‘t exist without the other. No productive squad without skilled resourcing from relevant chapters. No point in having chapters when there’s no need for the expertise.
I've seen a talk about all the cool things Salesforce does with TransmogrifAI. Highly recommend checking it out
the examples contradict the notion that you have to either select an explainable model or a more complex – and therefore more accurate – one. The explanations given using an accurate and complex Gradient Boosting Machine model were as clear as a linear model could get and made total intuitive sense. This also highlights the strengths of SHAP in providing accurate and consistent explanations.
Traditional feature importance algorithms will tell us which features are most important across the entire population, but this one-size-fits-all approach doesn’t always apply to each individual customer. A factor that is an important driver for one customer may be a non-factor for another.
If the first thing we do is poke around in our data, our decision will be, at best, something I like to call data-inspired.
Someone transferred $99 million in litecoin — and it only cost them $0.40 in fees
If a machine knew how to define a contract automatically, and put it on some system that other systems could solve, there is a potential for now machine learning systems solving problems autonomously at a certain level
"Electrification enables autonomy, and autonomy accelerates adoption of electrification".
The new React license will take effect later this week with the launch of React 16
If companies want to get value from their data, they need to focus on accelerating human understanding of data, scaling the number of modeling questions they can ask of that data in a short amount of time, and assessing their implications.
ARCore is launching on the Pixel and Galaxy S8 (running 7.0 Nougat and above) to start
"A post-smartphone world is all about the shift to Augmented Reality with glasses that 50+% of people in the US wear everyday"
manufacturing two genuinely different versions of a product costs a lot more. The challenge is to predict the willingness to pay of customers while making them feel as if they have benefited from value – or better features. “If you have one product and the price is too high, people don’t buy it. But if it’s too low, you don’t exploit some customers’ willingness to pay,” he says.
You should assume that whatever specific libraries and software you learn today will be obsolete in a year or two. Just think about the number of changes of libraries and technology stacks that occur all the time in the world of web programming — and yet this is a much more mature and slow-growing area than deep learning. So we strongly believe that the focus in learning needs to be on understanding the underlying techniques and how to apply them in practice, and how to quickly build expertise in new tools and techniques as they are released.