id
stringclasses 15
values | title
stringclasses 15
values | url
stringclasses 15
values | published
stringclasses 15
values | text
stringlengths 2
633
| start
float64 0
4.86k
| end
float64 2
4.89k
|
|---|---|---|---|---|---|---|
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and many different things. | 326 | 329 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | Lots of strange arrangements of all of these objects we'd like to recognize. | 329 | 333 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So cats coming in very different poses. | 333 | 336 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | By the way, the slides when I create them, they're quite dry, there's a lot of math and so on, | 336 | 339 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | so this is the only time I get to have fun. | 339 | 341 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So that's why I just pal everything with cat pictures. | 341 | 344 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So we have to be robust to all of these deformations. | 344 | 347 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | You can still recognize that there's a cat and all of these images, despite its arrangement. | 347 | 350 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | There's problems of occlusion, so sometimes we might not see the full object, | 350 | 354 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | but sometimes that's a cat behind a curtain. | 354 | 357 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | That's a cat behind a water bottle, and there's also a cat there inside a couch, | 357 | 361 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | even though you're seeing just tiny pieces of this class, basically. | 361 | 366 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | There's problems of background clutter, so things can blend into the environment. | 366 | 371 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | We have to be robust to that. | 371 | 373 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And there's also what we refer to as intraclass variation. | 373 | 376 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So cat, actually, there's a huge amount of cats, just species, | 376 | 380 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and so they can look different ways. | 380 | 382 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | I just like you to appreciate the complexity of the task when you consider any one of these, | 382 | 386 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | independently, it's difficult. | 386 | 387 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | But when you consider the full cross product of all these different things, | 387 | 390 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and the fact that all algorithms have to work across all of that, | 390 | 394 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | it's actually quite amazing that anything works at all. | 394 | 396 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | In fact, not only does it work, but it works really, really well, | 396 | 399 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | almost at human accuracy. | 399 | 400 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | We can recognize thousands of categories like this, | 400 | 402 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and we can do that in a few dozen milliseconds with the current technology, | 402 | 405 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and so that's what you'll learn about in this class. | 405 | 408 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So what does an image classifier look like, basically? | 408 | 411 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | We're taking this 3D array of pixel values we'd like to produce a class label. | 411 | 416 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And what I'd like you to notice is that there's no obvious way of actually encoding | 416 | 419 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | any of these classifiers, right? | 419 | 421 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | There's no simple algorithm, like say you're taking an algorithm class | 421 | 424 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and your early computer science curriculum, | 424 | 426 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | you're writing bubble sort, or you're writing something else to do any particular task. | 426 | 430 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | You can intuit all the possible steps, and you can enumerate them, | 430 | 432 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and list them, and play with it, and analyze it. | 432 | 434 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | But here, there's no algorithm for detecting a cat under all these variations, | 434 | 438 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | or it's extremely difficult to think about how you'd actually write that up, | 438 | 441 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | what is the sequence of operations you would do on an arbitrary image to detect a cat? | 441 | 445 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | That's not to say that people haven't tried, especially in the early days of computer vision. | 445 | 449 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | There were these explicit approaches, as I'd like to call them, | 449 | 452 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | where you think about, okay, a cat, say, we'd like to maybe look for little earpieces, | 452 | 457 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | so what we'll do is we'll detect all the edges, we'll trace out edges, | 457 | 460 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | we'll classify the different shapes of edges and their junctions, | 460 | 464 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | we'll create libraries of these, and we'll try to find their arrangements, | 464 | 467 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and if we ever see anything ear-like, then we'll detect a cat, | 467 | 470 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | or if we see any particular texture of some particular frequencies, | 470 | 473 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | we'll detect a cat. | 473 | 474 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So you can come up with some rules. | 474 | 476 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | But the problem is that once I tell you, okay, I'd like to actually recognize a boat now, | 476 | 480 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | or a person, then you have to go back to the drawing board, and you have to be like, | 480 | 483 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | okay, what makes a boat exactly was the arrangement of edges, | 483 | 486 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | right, it's completely unscalable approach to classification. | 486 | 489 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And so the approach we're adopting in this class, | 489 | 491 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and the approach that works much better, is the data-driven approach | 491 | 495 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | that we like in the framework of machine learning. | 495 | 498 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And just to point out that in these days, actually, in the early days, | 498 | 501 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | they did not have the luxury of using data, | 501 | 503 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | because at this point in time, you're taking, you know, | 503 | 505 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | grayscale images of very low resolution, | 505 | 507 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | you have five images, and you're trying to recognize things, | 507 | 509 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | it's obviously not going to work. | 509 | 510 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | But with the availability of internet, huge amount of data, | 510 | 513 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | I can search, for example, for cat on Google, | 513 | 516 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and I get lots of cats everywhere, and we know that these are cats | 516 | 519 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | based on the surrounding text in the web pages. | 519 | 522 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And so that gives us lots of data, so the way that this now looks like | 522 | 525 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | is that we have a training face, where you give me lots of training examples | 525 | 528 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | of cats, and you tell me that they're cats, | 528 | 531 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and you give me lots of examples of any type of other category you're interested in. | 531 | 534 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | I do, I go away, and I train a model. | 534 | 537 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | A model is a class, and I can then use that model | 537 | 540 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | to actually classify new test data. | 540 | 542 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So when I'm giving a new image, I can look at my training data, | 542 | 545 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and I can do something with this based on just a pattern matching, | 545 | 548 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and statistics, and so on. | 548 | 550 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So as a simple first example, we'll work with in this framework, | 550 | 554 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | consider the nearest neighbor classifier. | 554 | 556 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | The way nearest neighbor classifier works is that effectively, | 556 | 558 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | we're given this giant training set. | 558 | 560 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | What we'll do at training time is we'll just remember all of the training data. | 560 | 563 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So I have all the training data, I just put it here, and I remember it. | 563 | 566 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | Now when you give me a test image, what we'll do is we'll compare that test image | 566 | 570 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | to every single one of the images we saw in the training data, | 570 | 573 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and we'll just transfer the label over. | 573 | 575 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So I'll just look through all the images. | 575 | 578 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | We'll work with specific case, as I go through this, I'd like to be as concrete as possible. | 578 | 582 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So we'll work with a specific case of something called CFAR 10 dataset. | 582 | 585 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | The CFAR 10 dataset has 10 labels. | 585 | 588 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | These are the labels. | 588 | 589 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | There are 50,000 training images that you have access to, | 589 | 592 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | and then there's a test set of 10,000 images, | 592 | 595 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | where we're going to evaluate how well the classifier is working. | 595 | 598 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | And these images are quite tiny. | 598 | 599 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | They're just a little toy dataset of 32 by 32 little thumbnail images. | 599 | 603 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | So the way nearest neighbor classifier would work is we take all this training data | 603 | 607 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | that's given to us, 50,000 images. | 607 | 609 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | Now a test time, suppose we have these 10 different examples here. | 609 | 613 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | These are test images along the first column here. | 613 | 615 |
8inugqHkfvE | CS231n Winter 2016: Lecture 2: Data-driven approach, kNN, Linear Classification 1 | https://youtu.be/8inugqHkfvE | 2016-01-06T00:00:00.000000 | What we'll do is we'll look up nearest neighbors in the training set | 615 | 618 |