I've been trying to teach myself ML and AI for a while now, and though only tangentally relevant to the article perhaps others can take a few tips from my experience. First, I didn't really have a breakthrough until I ditched the 'recommended' textbooks and video courses and just started picking out a topic and learning by doing. Anything I don't understand or know when trying to implement it I just google/youtube/wikipedia it and keep messing with it until I know how it works at a conceptual level. Thats where resources like this article really come in handy.
For the heavier math parts, I just write a short summary of the formula and what its various parameters do and try to make a mental note of it. I certainly do not try to use the formulas to solve complex mathematical problems or write an implementation in python. I chalk this task up to `someday when I have the time`.
Finally, I'll get a dataset and try to solve various problems using the new skill I just learned using R/python & co.
Thats it.
This method has ridiculously accelerated the speed at which I've been able to acquire ML/AI skills that I also know how to apply in the real world. Before I felt like I was moving at a snails pace.
This method might not work well for everyone but its at least an interesting alternative to most of the recommendations of doing A,B,C online courses and reading X,Y,Z books.
Your approach resonates with me. I will recommend Practical Deep Learning For Coders[1]. This course is taught by Jeremy Howard who won kaggle competition for 2 consecutive years. His motto is to "Make Deep Learning Uncool again"
I personally found it very hands on, it jumps into practical application right of the bat which helps keeping the motivation steady. Having said that, its not easy or dumbed down in any sense.
I know NLTK isn't used for learning by itself, but it's such an essential player in the realm of NLP+Python that I think it should definitely be included (and that the list would be incomplete without it). The best link is probably http://www.nltk.org/book/ which is an excellent place to get started.
You should probably also include the gensim package https://radimrehurek.com/gensim/, since it has the most popular python word2vec implementation. One of the links I clicked on mentions it and walks you through using it, but I think it would make sense to point people to it directly in case they don't have time for a tutorial.
I work in NLP and I would really disagree that NLTK is still an essential player. Some academic courses still teach in it, but I think it's around five years since it has been the best option. I would definitely recommend gensim, but also SpaCy, which is very fast and has a better documented and more extensive API than NLTK. I have nothing to do with SpaCy personally (except being an enthusiastic user) but I recommend this argument: https://explosion.ai/blog/dead-code-should-be-buried
Thanks, I tried NLTK many times and ended up dropping it because it became too frustrating to deal with various Parser with no explanation of why there are so many and how to pick between them.
I skimmed though Spacy briefly and it looks great.
The v2 models are much smaller (15mb), because neural networks. The parsing, NER and tagging are mostly okay with the 50mb model. There are only word vectors for the top 5k words though, which can be a problem.
The v2 English models are more accurate, and can assign vectors to any word, including unknown words using the context and the word shape. Overall it's much better -- but it's still in alpha. The docs are already better, though.
These were written to demonstrate modern techniques with readable code, after seeing way too many indecipherable tangles of models. PyTorch plays a big part in that readability.
I'm actually recently started ML using a book, jupyter and Python. But i have a hard term translating the formula's to code and graphs. I know/understand what the formulas do, but the code to write seems hard. Perhaps because i don't know Python. Any advice on this?
Ps. Only 1 day since i installed Jupyter to recreate graphs and implement formulas
Ps2. Thought about implementing c# as a kernel in Jupter, but i think it's better to continue with Python.
Nice collection.. What I am looking for in addition to these are an explanation of different deep network topologis and how to construct new solutions utilizing ever more complex structures.
I'm working on using Tensorflow to improve Chinese OCR. So far I collected 1500 fonts. Now I'm exporting all the glyphs, and I ran out of disk space yesterday. Then I'll upload the data set of all 75,000 characters here on HN where people can play with it.
Before you say that OCR is a solved problem because of Tesseract, please read this:
Neither machine learning nor NLP nor Python will give you sentience. However...
Machine learning - no one will tell you the answer b/c they're working on it (whatever "it" is) right now. But here's a good tip from Mr. Andrew Ng:
"...almost anything a typical human can do with less than one second of mental thought, we can probably [do] now or in the near future automate using AI... Take a security guard looking at a video feed and saying, “Are there people in this? Are they doing something suspicious?” That task is actually a lot of one-second judgment thoughts strung together, so I think a lot of it can be automated."
I've been trying to teach myself ML and AI for a while now, and though only tangentally relevant to the article perhaps others can take a few tips from my experience. First, I didn't really have a breakthrough until I ditched the 'recommended' textbooks and video courses and just started picking out a topic and learning by doing. Anything I don't understand or know when trying to implement it I just google/youtube/wikipedia it and keep messing with it until I know how it works at a conceptual level. Thats where resources like this article really come in handy.
For the heavier math parts, I just write a short summary of the formula and what its various parameters do and try to make a mental note of it. I certainly do not try to use the formulas to solve complex mathematical problems or write an implementation in python. I chalk this task up to `someday when I have the time`.
Finally, I'll get a dataset and try to solve various problems using the new skill I just learned using R/python & co.
Thats it.
This method has ridiculously accelerated the speed at which I've been able to acquire ML/AI skills that I also know how to apply in the real world. Before I felt like I was moving at a snails pace.
This method might not work well for everyone but its at least an interesting alternative to most of the recommendations of doing A,B,C online courses and reading X,Y,Z books.
reply