WEBVTT 00:00.840 --> 00:06.040 In this video we are going to impart our very first time series data and then our fights. 00:06.060 --> 00:12.330 We have a CSG fight called temperature or temp and it contains temperature information for New York 00:12.360 --> 00:14.310 and Los Angeles for the yes. 00:14.370 --> 00:19.800 2013 to 2016 and in a very first step. 00:19.840 --> 00:27.240 Is actually no difference between importing longtime serious data and time serious data from CSP fires. 00:27.240 --> 00:33.040 So first of all we import pandas and then we are using the method. 00:33.090 --> 00:34.190 Read CSB. 00:34.880 --> 00:41.730 So we are importing our CSP fire temperature our temp and we actually store our new data frame then 00:42.030 --> 00:51.060 in the web temperature and let's have a very first look at our data frame so we we can see the first 00:51.060 --> 00:54.420 five rows and actually we have a sweet column. 00:54.420 --> 01:01.650 So we have the column date time we have L.A. for Los Angeles and we have New York and apparently here 01:01.680 --> 01:08.820 as well use of our data frame we have actually the temperature at this specific time in these specific 01:08.970 --> 01:14.560 city and the temperature is actually in degrees Celsius. 01:14.560 --> 01:21.760 So for example here we have January eleven point seven in L.A. and minus one point one degrees Celsius 01:21.790 --> 01:23.290 in New York. 01:23.290 --> 01:30.760 So let's get some meta information with the info method and on the left hand side we have a range index 01:30.820 --> 01:34.770 and we have actually thirty five thousand sixty four rows. 01:34.840 --> 01:42.260 So we have over thirty five thousand different the timestamps here and with actually three columns. 01:42.260 --> 01:48.050 So we have the date time column and obviously here we have string data type. 01:48.050 --> 01:55.190 So here we have object and this is a clear indication that to we to the strings in the column then we 01:55.190 --> 01:59.130 have the column in The Los Angeles service the flow data types. 01:59.150 --> 02:04.130 So these are your degrees Celsius values and therefore float the data type. 02:04.130 --> 02:10.000 Makes sense and also here for the New York column we have also a float data type. 02:10.730 --> 02:17.870 So the first problem is here that in the daytime column we actually do not have a specific date time 02:17.870 --> 02:18.900 data type. 02:19.040 --> 02:25.400 So we have a string that I type and this might not be really helpful and therefore on a first step we 02:25.400 --> 02:32.030 can try to transform data information that this actually starts here and strings into a date time data 02:32.060 --> 02:39.260 type of format and we can do this immediately when we import the data frame so they read CSP method 02:39.260 --> 02:47.630 provide some parameters that helps to import the date time information and let's go inside and if we 02:47.630 --> 02:54.770 can see that we have the pyramid up past dates and by default it set to false but let's have a look 02:54.770 --> 02:55.480 here. 02:55.580 --> 03:02.720 So apostates and Tippett can actually pass column labels for columns that we want to transform from 03:02.840 --> 03:11.420 string data type to a date time data type and we have to do this as a list and obviously our column 03:11.450 --> 03:14.460 that we want to transform a the daytime column. 03:14.540 --> 03:23.250 So we pass the time and we can also look at the documentation for the past dates power Mehta 03:27.970 --> 03:29.450 so weird us. 03:29.560 --> 03:36.280 And for example here we can pass the Boolean value true and then pan out tries to pass them the index 03:36.280 --> 03:37.940 into a date time format. 03:38.020 --> 03:40.870 But obviously here we have a Range Index. 03:40.870 --> 03:46.710 So we do not have any date time information on the index or we can pass a list of integers or names 03:46.750 --> 03:53.890 so we can pass either a list with the index position of the columns or the column labels and this is 03:53.890 --> 04:00.460 exactly what we do what we do here we pass that the column label daytime and let's see what we got here. 04:00.610 --> 04:07.510 So we actually overwriting our variable temperature and let's have a look here again at the Inform method 04:08.420 --> 04:15.990 and here we can see now that the data type changed from object to a daytime 64 ante and spat brackets 04:16.000 --> 04:17.550 we have an annual seconds. 04:17.890 --> 04:22.800 So the position is nanoseconds if we want to have this. 04:22.900 --> 04:27.000 So now we can also update you on the view on the first five rows. 04:27.650 --> 04:32.930 And now we have had the column date time with the data type date times 64. 04:33.290 --> 04:40.000 And we can also slice your fun element in our daytime column and we can select here the very first element 04:40.450 --> 04:42.360 myth of the ILO operator. 04:44.650 --> 04:50.780 And here we have a so-called timestamp and a time stamp is actually a single point in time. 04:50.800 --> 04:58.600 So for example here we have the year 2013 we have a 10 year Ria and we have it and you read the first 04:59.080 --> 05:01.600 and then also we have information on the time. 05:01.600 --> 05:03.660 So we have at midnight here. 05:03.700 --> 05:04.980 So these are the hours. 05:04.990 --> 05:08.380 Then we have minutes and then seconds. 05:08.560 --> 05:13.530 And as you can see above our data frame provides hourly temperature information. 05:13.660 --> 05:21.730 So the increment this one hour and we're starting here with the January the 1st at midnight then our 05:21.730 --> 05:25.060 next timestamp is this third January the 1st. 05:25.060 --> 05:33.640 1:00 a.m. Then we have 2:00 a.m. 3:00 and so on and we can also check here the data type off the time 05:33.640 --> 05:35.670 stamp of that type of method 05:40.270 --> 05:43.510 and this is here a panda's timestamp. 05:43.780 --> 05:50.140 So the timestamp is a specific Panda's data type to store daytime information. 05:50.140 --> 05:56.290 And whenever we have daytime information like this it makes sense to have the daytime as the index of 05:56.290 --> 05:57.520 our data frame. 05:57.760 --> 06:01.600 And by doing so this gives us a lot of additional functionality. 06:02.660 --> 06:09.200 So we can already do this when we import our data obviously read CSA method and we've already seen that 06:09.220 --> 06:11.540 that is that the parameter index calls 06:15.450 --> 06:20.850 and Tippecanoe past the column labor for the column that you want to have as index and in this case 06:20.850 --> 06:23.130 it's your day time. 06:23.160 --> 06:29.250 So this is actually a very common workflow with time series that when we import the data we first of 06:29.250 --> 06:36.950 all pass the columns that contain the daytime information with the parent to pass dates and then being 06:36.950 --> 06:39.840 can to these columns as the index. 06:39.840 --> 06:42.870 So let's do this here and be all right. 06:42.870 --> 06:50.640 Our variable temperature and have again a look at the first five rows and now we can see here that we 06:50.640 --> 06:52.150 have our daytime column. 06:52.150 --> 07:01.630 Now as the index and we can also check this with the informal method so we can see here that we have 07:01.660 --> 07:11.980 a so-called daytime index and we have 35000 entries and these range from January the 1st 2013 midnight 07:12.220 --> 07:18.980 until December thirty first of 2016 11:00 p. m.. 07:19.060 --> 07:26.500 So we feel all the temperature information for New York and Los Angeles for the year 2000 and 30 into 07:26.500 --> 07:28.090 2016. 07:28.990 --> 07:34.610 And we can also call you to the index attribute to further analyze our new index. 07:34.720 --> 07:39.980 And here we have our daytime index until we have all of the timestamps. 07:40.060 --> 07:46.180 So a daytime index is actually the collection of many timestamps and we can see at the bottom that we 07:46.180 --> 07:54.940 have the data type daytime 64 and index label we have daytime and we have thirty five thousand entries 07:56.020 --> 08:02.620 and we can also select a single element in our index so let's select the very first element and this 08:02.620 --> 08:04.730 is here the very first timestamp. 08:04.930 --> 08:09.030 So January of the 1st 2013 at midnight. 08:09.070 --> 08:09.550 All right. 08:09.550 --> 08:15.670 This were the first steps with time series and I hope to see you also in the next video by.