Roger Peng ** 2018/06/30

值得一提的是，尽管在 4 月底 PyTorch 在 0.4 版本中已经加入了对于 Windows 的支持，但其对比 Keras 与 TensorFlow 在 Windows 上的稳定性还略有差距。

In this post I’d like to share some of my recommended books for learning data science and machine learning, both in theory and and practice.

In our research reports, we cover “the recently possible,” and what makes “the recently possible” possible. In addition to a detailed how-to guide of new machine learning capabilities, each of our reports contains a section on open source projects, commercial offerings, and vendors that help implement this new machine learning capability to realize the opportunities opened up by technological innovation. We like to keep an eye on the the technologies we research, of course. Our report on deep learning (for image classification) was published a few years ago, and we have seen noteworthy new developments.

Title image adapted from bigocheatsheet.com[/caption]

Sequence labeling tasks attempt to assign a categorical label to each member in the sequence. In natural language processing, where a sequence generally refers to a sentence, examples of sequence labeling include named entity recognition (NER), part-of-speech tagging (POS) and error detection. NER, as the name implies, tries to recognize names in a sentence and classify them into pre-defined labels such as Person and Organization. POS tagging assigns labels such as noun, verb, and adjective to each word, while error detection identifies grammatical errors in sentences. In many of these tasks, the relevant labels in the dataset are very sparse and most of the words contribute very little to the training process. But why let the data go to waste?

Learning a new skill by observing another individual, the ability to imitate, is a key part of intelligence in human and animals. Can we enable a robot to do the same, learning to manipulate a new object by simply watching a human manipulating the object just as in the video below?

At Stitch Fix, we approach nearly all problems with a humans-in-the-loop framework. In some cases this means coupling algorithmic recommendations with human oversight and feedback. In others, it means building interpretable, descriptive models that provide insight to our business partners. These insights can guide and improve company decisions by providing much-needed context for the application of human expertise.

The Data Incubator recently teamed up with MRI Network to increase its access to hiring partnerships worldwide. MRINetwork is comprised of over 1,500 search professionals who specialize in hundreds of industries, many of who came from the industries in which they now recruit. MRI recruiters combine their in depth understanding of industry; with the knowledge of who is who in almost any discipline in order to jump start the search for an enterprise’s next impact player or an entire division.

Mushrooms, rats, and smartphones: Enjoy these new, intriguing, and overlooked datasets and kernels.