Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
0
00:00:01,040 --> 00:00:01,970
[Autogenerated] and this discussion of
1
00:00:01,970 --> 00:00:04,400
cross validation brings us to the very end
2
00:00:04,400 --> 00:00:06,620
of this model. Very understood. The rule
3
00:00:06,620 --> 00:00:08,519
of features in machine learning. We
4
00:00:08,519 --> 00:00:10,470
started off the discussion off the
5
00:00:10,470 --> 00:00:13,080
importance of data in machine learning. If
6
00:00:13,080 --> 00:00:15,179
you feed garbage into your model, you're
7
00:00:15,179 --> 00:00:17,500
going to get garbage out. Your model is
8
00:00:17,500 --> 00:00:19,850
only as good as your data. In this
9
00:00:19,850 --> 00:00:22,239
context, we discuss how machine learning
10
00:00:22,239 --> 00:00:24,989
algorithms can learn from data, and he
11
00:00:24,989 --> 00:00:27,649
started what features and labels mean.
12
00:00:27,649 --> 00:00:29,690
They're also referred to as ex variables
13
00:00:29,690 --> 00:00:32,649
and by values. We then got a big picture
14
00:00:32,649 --> 00:00:35,000
understanding off all off the processes
15
00:00:35,000 --> 00:00:37,340
involved in the common machine learning
16
00:00:37,340 --> 00:00:40,229
workflow be understood exactly where in
17
00:00:40,229 --> 00:00:42,479
this workflow data, pre processing and
18
00:00:42,479 --> 00:00:45,000
feature engineering fit in. We then saw
19
00:00:45,000 --> 00:00:47,420
how feature engineering could mean many
20
00:00:47,420 --> 00:00:50,079
different things. It's a broad array of
21
00:00:50,079 --> 00:00:52,240
techniques, all of which fall another
22
00:00:52,240 --> 00:00:55,020
feature engineering umbrella And finally,
23
00:00:55,020 --> 00:00:57,409
the rounded our discussion off by talking
24
00:00:57,409 --> 00:00:59,350
off the kinds of data that you will work
25
00:00:59,350 --> 00:01:02,030
with the new building animal models. We
26
00:01:02,030 --> 00:01:04,010
spoke off the importance off training
27
00:01:04,010 --> 00:01:06,530
tests and validation data, and he also
28
00:01:06,530 --> 00:01:09,170
discussed K fold cross validation to build
29
00:01:09,170 --> 00:01:12,209
more robust models in the next model will
30
00:01:12,209 --> 00:01:14,530
get hands on and we see how we can prepare
31
00:01:14,530 --> 00:01:18,000
and pre process all data for machine learning.
2494
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.