alexcnwy

29 ms·

121.
by alexcnwy 7y ago
Very cool, thanks for the link!
122.
by alexcnwy 7y ago
I'd love to hear more about what kind of data augmentation you're doing. A friend of mine recently got a GAN to work for timeseries which is really interesting. I've done a lot of work in the space and would love to chat - ju
123.
by alexcnwy 7y ago
Very interesting. Your nursing home project sounds like you could definitely add loads of value - best of luck for it!
124.
by alexcnwy 7y ago
Fascinating story, thanks for sharing! "we realized that there will always been a better resourced or more dedicated team willing to fight you for your alpha" - that's part of what makes financial markets such a fun and inter
125.
by alexcnwy 7y ago
This is just an example of a common problem e.g. invoices, transcripts, purchase orders, etc.
126.
by alexcnwy 7y ago
It’s surprising to hear you were getting good results with Python OCR libraries 10 years ago because there aren’t any reliable Python OCR libraries even today! ‘:) Tesseract is very fickle and doesn’t work well in poor lighting conditions (
127.
by alexcnwy 7y ago
This is a decent OCR / structured data extraction literature review but is absolutely not "building an ID card reader from scratch with deep learning". It's also very hand-wavey on the details of how to actually use grap
128.
by alexcnwy 7y ago
I’m saying you don’t just change the risk allocation by raising money, you change the risks themselves in a way that may be making your risk as a founder greater, not smaller. You don’t carry “all” the risk but your risk still goes up.
129.
by alexcnwy 7y ago
Selling equity for capital introduces new risks though - many businesses are pushed to grow too quickly and take actions in the pursuit of billion dollar exits because they had raised money which causes them to fail. Sure, the capital inj
130.
by alexcnwy 7y ago
A lot of people probably don’t know that Fastmail was started by Jeremy Howard, one of the creators of https://www.fast.ai which is by far the best way to learn deep learning IMHO :)
131.
by alexcnwy 7y ago
Faking someone's voice isn't as difficult as you'd think: https://github.com/andabi/deep-voice-conversion https://www.theverge.com/2019/9/5/20851248/deepfakes-ai-fake.
132.
by alexcnwy 7y ago
That’s so epic! Can you please give a little bit more detail on your process?
133.
by alexcnwy 7y ago
I have no doubt that deepfakes will be impossible to tell apart from real videos (at least to a human) in 3-5 years. In 10 years I think we’ll be able to generate entire movies programmatically.
134.
by alexcnwy 7y ago
This is so cool! From the page: "Your audio doesn’t get uploaded to servers to be analyzed, so your vocals stay private." From their ProductHunt launch: "We made an AI-powered singing challenge to find out. This experiment
135.
Can you sing like Freddie Mercury?(freddiemeter.withyoutube.com)
3 points by alexcnwy 7y ago|1 comments
136.
by alexcnwy 7y ago
Very cool, thanks for the detailed reply - your JS tutorial looks great :) Yeah automatic re-execution sounds like a weird design choice. Think I'll stick with Jupyter notebooks for now...
137.
by alexcnwy 7y ago
Yeah I remember seeing that one a few years ago - it's also super cool :) Especially crazy that these work as well and quickly on mobile!
138.
by alexcnwy 7y ago
Mike Bostock is a beast - this d3.js example from 2013 still blows my mind: https://archive.nytimes.com/www.nytimes.com/interactive/2012... I haven’t gotten properly stuck into observable - in what ways is it bett
139.
by alexcnwy 7y ago
There's a lot of room between a laptop with 16GB of RAM and a "Big Data cluster" - in my experience the easiest solution is to spin up a VM in GCP and crank the RAM way up. It blew my mind when I realized I could add / r
140.
by alexcnwy 7y ago
In my experience it's a bit of both. Most of a data scientist's time is definitely spent getting data from various places, possibly enriching it, cleaning it, and converting it into a dataframe that can then be modeled (and surpri
141.
by alexcnwy 7y ago
The ability to re-size GCP VMs totally blew me away when I first discovered it. Just power off the machine, drag the RAM slider way up and turn it back on. SOOO much easier than re-creating the VM on AWS. I also way prefer to just crank up
142.
by alexcnwy 7y ago
I totally get your skepticism but the bottom line is that throwing computing power at a problem generally leads to a much better solution even if only because you can do a much more extensive grid-search. This doesn’t have to be using some
143.
by alexcnwy 7y ago
The big difference is that historically you could run pretty much all analysis on a single computer (if not in Excel - I’ve seen and built many actuarial models in Excel myself) but consumer internet scale data (both in terms of the number
144.
by alexcnwy 7y ago
The impacts in improved translation aren’t marginal - they improve people’s lives in real ways. Advances in deep learning for NLP have also significantly improved search (Google and others) which has a nontrivial impact on people’s lives. M
145.
by alexcnwy 7y ago
A lot of the breakthroughs have been in image classification, object detection and image segmentation using deep convolutional models way way outperforming prior approaches and beating human level benchmarks (on constrained problems e.g. cl
146.
by alexcnwy 7y ago
I've built several object detection systems and you definitely want to fine-tune your model to get good results, even if the object class you want to track is available in a pre-trained model (often it isn't). For example, pre-tra
147.
by alexcnwy 7y ago
The flicker would be a challenge but you can use something like DeOldify [1] or super resolution [2] on the frame sequence to improve the colour and resolution... although you may get some weird artifacts between frames using image models r
148.
by alexcnwy 7y ago
GANs + "A ton of python magic" [1] :) [1] https://twitter.com/xsteenbrugge/status/1188798045158293505
149.
by alexcnwy 7y ago
This is so incredible and well executed. It's crazy to see how quickly GANs are moving (e.g. check out this tweet [1] by Ian Goodfellow on 4.5 years of GAN progress) ... excited to see what they can do a few years from now! [1] https:
150.
by alexcnwy 7y ago
Often occlusions don’t matter in practice for dense object tracking e.g. counting people entering a bus or measuring gym equipment utilization you actually don’t care if the person IDs switch. Centroid tracking works well in practice at 2 F

More ›