1 00:00:00,900 --> 00:00:01,780 Right. 2 00:00:01,800 --> 00:00:02,370 Welcome back. 3 00:00:02,610 --> 00:00:04,110 So this might I do. 4 00:00:04,380 --> 00:00:07,140 I'm going to create a new script so I'll come to follow over here. 5 00:00:07,140 --> 00:00:10,590 New and then I'm going to press control. 6 00:00:10,610 --> 00:00:11,440 Yes. 7 00:00:11,650 --> 00:00:13,940 And I'm going to navigate. 8 00:00:14,290 --> 00:00:20,110 I'm going to navigate to uh um my project folder is cut or not over here. 9 00:00:20,290 --> 00:00:23,020 I'm going to call this helper. 10 00:00:23,290 --> 00:00:28,540 There's going to be the name of the script and I'm going to write a function in this script to help 11 00:00:28,540 --> 00:00:29,840 us load a war. 12 00:00:30,310 --> 00:00:32,230 What data what in your network. 13 00:00:32,230 --> 00:00:36,910 I'll start off by importing num pi import num pi 14 00:00:40,500 --> 00:00:41,450 S and P. 15 00:00:41,570 --> 00:00:42,260 Like this. 16 00:00:42,410 --> 00:00:47,290 Once that is done I'm going to import H5 pi. 17 00:00:47,840 --> 00:00:51,940 Remember there's the format of what image data once that's done. 18 00:00:51,950 --> 00:00:53,590 I'm going to write my function I'll say it. 19 00:00:53,600 --> 00:01:02,780 Def load data set and then come here like this. 20 00:01:03,240 --> 00:01:03,530 Okay. 21 00:01:03,550 --> 00:01:05,900 I'm gonna create a variable quote train data set. 22 00:01:05,900 --> 00:01:08,100 We're gonna have train and test data set. 23 00:01:08,330 --> 00:01:14,000 We're going to have the training set is going to have its x values its y values to test set is going 24 00:01:14,000 --> 00:01:15,090 to have the same thing. 25 00:01:15,220 --> 00:01:16,520 I'll say train data set 26 00:01:19,990 --> 00:01:22,990 and all we have to do is read I far so see 27 00:01:27,690 --> 00:01:34,850 H5 pi data file and then we find a file in our food our code data set 28 00:01:37,650 --> 00:01:42,460 in the name of the file this is trained you makes data Delta H5 29 00:01:45,710 --> 00:01:49,150 and we said we are reading it. 30 00:01:49,190 --> 00:01:53,600 So we pass are over here like this. 31 00:01:53,600 --> 00:01:54,050 Okay. 32 00:01:54,110 --> 00:02:01,590 So once we've read a file we are going to take the x values and story tonight a separate variable and 33 00:02:01,590 --> 00:02:12,920 a wife I'll use in a separate variable to do it I'm gonna see cranes set X original because then I'll 34 00:02:12,920 --> 00:02:21,530 see an umpire ray which we've renamed to NDP num pi dot array and then obviously train data which is 35 00:02:21,530 --> 00:02:23,090 this variable we've created here 36 00:02:27,160 --> 00:02:28,590 what I want to do is say 37 00:02:33,260 --> 00:02:44,390 Crane sets X and then there's the first index and then the second one we see the semicolon for all right 38 00:02:44,750 --> 00:02:50,810 what is this going to do is this is going to get the features this is gonna get a features of the training 39 00:02:50,820 --> 00:03:00,740 sets which are the x values and then I'm gonna do for the y values as well as a train set y or rage 40 00:03:01,120 --> 00:03:06,260 I'm going to straighten this variable and then we'll see and Peter s a ray 41 00:03:09,980 --> 00:03:10,790 Crane set 42 00:03:17,380 --> 00:03:27,130 and then when a pass Crane set y I'm typing this because this the way I want H5 data is arranged internally 43 00:03:28,260 --> 00:03:30,000 right. 44 00:03:30,250 --> 00:03:30,660 Right. 45 00:03:30,730 --> 00:03:33,580 So what we've just done is for Crane data set. 46 00:03:33,580 --> 00:03:35,300 This is for the training set. 47 00:03:35,410 --> 00:03:42,850 Next we going to take the test set and put it into it's X and then it's y. 48 00:03:43,990 --> 00:03:55,900 So I'm gonna come down here and see test data set because each 5 pi that's fine and is the same file 49 00:03:58,690 --> 00:04:05,110 is not the same file that we have quite different file for test of course we'll see a data sets 50 00:04:07,780 --> 00:04:23,740 slash test AMG data that H5 then a POS are here for read once that is done I'm going to see test set 51 00:04:25,620 --> 00:04:26,410 next original 52 00:04:30,540 --> 00:04:33,460 and I'm simply going to copy what I have here and rename it 53 00:04:38,570 --> 00:04:41,150 this over here becomes test set. 54 00:04:46,210 --> 00:04:48,040 This becomes test as well. 55 00:04:53,510 --> 00:04:56,420 Then I'm going to do for why put this 56 00:05:02,480 --> 00:05:03,530 this becomes test 57 00:05:06,270 --> 00:05:07,370 this becomes test. 58 00:05:09,570 --> 00:05:10,840 This becomes test as well. 59 00:05:12,720 --> 00:05:19,010 So once this is done we going to grab a list of classes and storage in a variable called classes. 60 00:05:19,030 --> 00:05:27,570 Come over here C classes equals empty array and then I'm gonna say 61 00:05:30,600 --> 00:05:31,620 test data set 62 00:05:36,030 --> 00:05:39,480 and then list classes 63 00:05:45,390 --> 00:05:45,750 right. 64 00:05:45,750 --> 00:05:47,980 I'm gonna grab all storage and classes. 65 00:05:48,000 --> 00:05:48,400 Okay. 66 00:05:48,420 --> 00:05:49,260 We're almost done. 67 00:05:49,260 --> 00:05:57,300 What we have to do now is simply reshape the DIY set for the training set and the test set. 68 00:05:57,330 --> 00:06:01,650 So I'm gonna come here and see cranes set. 69 00:06:02,460 --> 00:06:07,410 Y RH which is the original y set. 70 00:06:07,430 --> 00:06:22,980 I see a train set y original to reshape and then what I want to do is one by and I'm going to grab the 71 00:06:24,750 --> 00:06:25,680 one I grew up this 72 00:06:30,040 --> 00:06:41,520 dot shape index 0 right and I'm going to do the same for the test set. 73 00:06:42,150 --> 00:06:46,260 Come over here and then this becomes test 74 00:06:49,960 --> 00:06:51,030 just becomes test 75 00:06:55,700 --> 00:06:59,540 this becomes test right. 76 00:06:59,550 --> 00:07:07,080 Once this is done we can complete our function by returning the data we've processed or return to the 77 00:07:07,200 --> 00:07:17,620 sets we have processed and rearranged so I'll say return over here we can return train set X or rage 78 00:07:19,770 --> 00:07:26,220 and then we can return train set y or regional. 79 00:07:26,700 --> 00:07:35,610 So break and then test set X brake. 80 00:07:39,100 --> 00:07:41,330 I forgot the word set over here 81 00:07:44,300 --> 00:07:50,050 and then test set y a rake. 82 00:07:51,360 --> 00:07:53,330 Um we have the last one. 83 00:07:53,370 --> 00:08:01,090 Classes over here we can return to as well put over here to return. 84 00:08:01,940 --> 00:08:02,380 Right. 85 00:08:02,390 --> 00:08:03,770 So there is going to be a Y. 86 00:08:03,770 --> 00:08:08,540 Help File we shall be adding new functions to this as we go through the course. 87 00:08:09,080 --> 00:08:12,110 So now um we are done with this function. 88 00:08:12,110 --> 00:08:18,670 We shall use this function in a 1 your network script um and we cannot test it now. 89 00:08:18,710 --> 00:08:23,610 We test it when we load it during the um the neural network scripting. 90 00:08:24,080 --> 00:08:25,730 So if you have any questions just let me know. 91 00:08:25,760 --> 00:08:28,350 But for now this all there is for this particular lesson. 92 00:08:28,370 --> 00:08:29,090 I'll see you later.