I have a data set in google sheets, for each week of data I have 3 rows. I wish to query the data in every second row to calculate the max value and the last value.
For instance:
ROW
DATA
1
800
2
Text
3
500
4
More text
5
600
6
Blah
7
700
8
Blah
For Max value I have the following which will return 800
MAX(FILTER(QUERY(A1:A,"Select * skipping 2"), QUERY(A1:A,"Select * skipping 2") <> 0))
How do I change it up to return the last value? Which should return 700
try:
=LOOKUP(2^99,FILTER(A:A,A:A<>0))
#rockinfreakshow answer will successfully find the last number.
To filter a range by n amount of rows, you can use:
=FILTER(A:A,MOD(ROW(A:A),n)=1)
Change n with your desired value, and 1 with the number of row you want to get. 1 for the first, 2 for the second, but 0 if you want the nth one. To find MAX, just wrap it in MAX()
To find the last one, even if it's a text or number, you can use SORTN and SEQUENCE:
=SORTN(FILTER(A:A,MOD(ROW(A:A),n)=1,A:A<>""),1,1, SEQUENCE(COUNTA(FILTER(A:A,MOD(ROW(A:A),n)=1,A:A<>""))),0)
It orders the elements in reverse order and only chooses the first one
Remember to change n with the number of rows and =1 with the number of row you want to choose
I'm not sure how to explain this, but hopefully this makes sense. If I require 10 items of stock to be removed from this table data below, is it possible to send the number 10 into a query
and it updates the table based on the sort order, basically, it will take all it can from the first line and the rest from the second line (or third or firth if you know what I mean):
This is the start
componentid
stocklevel
so
991
5
1
1063
16
2
This is what I am trying to achieve
componentid
stocklevel
so
991
0
1
1063
11
2
Almost like, if it gets to zero, move to the next row
I'm trying to automate billing for my boss. I have to choose the highest quantity for an invoice date and client, then print that quantity in a separate column and a 0 (or blank) for the second row associated with that client. I'm trying to recreate this example:
Billing Snippet
I'm having trouble using Pandas to do this. I used a pivot table to get the max quantity for each client, then merged that data with the original to get a "max" column. That looks like this:
Dataframe snippet
My plan is to use indexes to essentially say "if the Qty is not equal to Max, then change the value to 0"
Here's my code, but I get the error "A value is trying to be set on a copy of a slice from a DataFrame" :
ad2[ad2['Qty'] != ad2['max']]['Qtrly Billing Count']=0
Any advice on how to tackle this?
Update: Tried turning off the setting that gives me the index error, but the column I want to update isn't changing. Help!
Recreating you df:
ad2 = pd.DataFrame({'Qty':[33, 47],'max':[47,47], 'Qtrly':[47,47] })
Qtrly Qty max
0 47 33 47
1 47 47 47
using loc:
ad2.loc[ad2['Qty'] != ad2['max'], 'Qtrly']=0
result:
Qtrly Qty max
0 0 33 47
1 47 47 47
Let's say I have two columns.
3.5463 11
4.5592 12
1.6993 111
0.92521 112
1.7331 121
2.1407 122
1.4082 1111
2.0698 1112
2.3973 1121
2.4518 1122
1.1719 1211
1.153 1212
0.67139 1221
0.64744 1222
1.3705 11111
0.9557 11112
0.64868 11121
0.7325 11211
0.58874 11212
0.86673 11221
0.17075 11222
0.64026 12111
0.80229 12112
0.43422 12122
1.0405 12211
0.63376 12212
0.56491 12221
0.34626 12222
0.81631 111111
0.91837 111112
0.70013 111121
0.87384 111122
1.1474 111211
0.47411 111221
0.12249 111222
0.56728 112111
0.88169 112112
0.14509 112121
0.68655 112211
0.36274 112212
1.1652 121111
0.99314 121112
0.42024 121121
0.23937 121122
1.0346 122111
0.64642 122112
0.15632 122121
0.41725 122122
0.40793 122211
In the first column, there is a number. With every one of those numbers, in the second column, is an associated ID. Now, there are some blank rows that do not contain any numbers in them.
Define one of these numbers to be a "daughter" of another number if the ID of the first number is the same as the ID of the second, with an extra digit on the end. For example, both IDs 11211 and 11212 are daughters of 1121, because the ID of 1121 has an extra digit, either a 1 or a 2, added onto the end to form the ID of its daughters. Thus, 1121 is the parent of both 11211 and 11212.
Here is what I want the macro to do. It must output a third column which contains, for every row, a cumulative sum of the number of the first column in that row, plus the parent number of that number, and the parent number of the parent number, etc. all the way up until it reachers either 11 or 12. It will begin by simply outputting the numbers in column 1 for 11 and 12 in the third column. Then, in a loop beginning with 111, it will add up the cumulative sum of every row (the number in that row plus the third column output of the parent), only if that row has a number and an id, and only if the parent exists and has an output in column 3. So for example, the number in the 3rd column of the row with ID 11222 should be the number in column 1 of that row, plus that of 1122, plus that of 112, plus that of 11. So, 0.17075+2.4518+0.92521+3.5463, or 7.09406. However, if you try to do this for ID 111221, you will notice that the row where the parent 11122 should be is empty. Thus, the parent does not exist, and no value will be outputted in column 3 for 111221.
I would greatly appreciate it if someone has some time on their hands to code up this VBA macro for me in exchange for an accepted solution.
Thanks
I don't think a macro is needed, just some formulas. First, I put a header on my columns of data, such as "value," and "id." If you then highlight the column labels (i.e., A and B) and sort by B ("id") then A ("value"), you'll group your blank rows. You can then delete those rows. Now you have the data almost ready. When I did this, I converted the id column to text, as opposed to a number value, so if I sort the table by id, the pattern will be, "11, 111, 1111," and so on, instead of, "11, 12, 111, 112, 121." Then, I added columns to separate the separate characters or levels of the ids. This is to help with parents and children. You can use text-to-columns, or a MID formula, but what I did was have 6 more columns to the right. For each id row, each column would either have a "1," a "2," or a blank (null) value. Then I added another column, calling it "level." I used a formula like COUNTA across all my id splitting columns. So, for 11, my level value was 2. 111 would be 3, 11221 would be 5, and so on. This gives me the id level (parent, child, grandchild, etc). Then I added my final column to the right to compute my cumulative sum of the values. In concept I have one big nested IF statement, but in practice, I needed two. My formula says, if the row above me has a lower level number (i.e., it is some kind of parent), add the value of the current row to the value of the above row. Otherwise, keep going up a row till I do get a parent, and add the current row value to that number.
My final formula for all but the first 5 rows of data was (in the 6th row of data):
=if(K6
rest of answer is below
=if(K6<K7,L6+C7,if(K5<K7,L5+C7,if(K4<K7,L4+C7,if(K3<K7,L3+C7,if(K2<K7,L2+C7,C7)))))
The values were column C, the original id in column D, the id split columns were E through J, the level column was K, and my formula was in L. This formula can be copied down the table. For the first 4 rows, you just need 1 less IF statement each row you go up. The fifth row of data might take the above formula; it depends how it will deal with the column headers in row one. The formula on the 4 row of data might be:
=if(K4<K5,L4+C5,if(K3<K5,L3+C5,if(K2<K5,L2+C5,if(K1<K5,L1+C5,C5))))
I'm still learning how to format these comments, so I'll try to provide a sample of the layout I have...
C D E F G H I J K L
1 value id 1 2 3 4 5 6 lvl cumul_sum
2 3.546300 11 1 1 2 3.546300
3 1.699300 111 1 1 1 3 5.245600
4 1.408200 1111 1 1 1 1 4 6.653800
5 1.370500 11111 1 1 1 1 1 5 8.024300
6 0.816310 111111 1 1 1 1 1 1 6 8.840610
7 0.918370 111112 1 1 1 1 1 2 6 8.942670
8 0.955700 11112 1 1 1 1 2 5 7.609500
So for example, the number in the 3rd column of the row with ID 11222 should be the number in column 1 of that row, plus that of 1122, plus that of 112, plus that of 11. So, 0.17075+2.4518+0.92521+3.5463, or 7.09406.However, if you try to do this for ID 111221, you will notice that the row where the parent 11122 should be is empty. Thus, the parent does not exist, and no value will be outputted in column 3 for 111221.
As a native worksheet array formula¹ in D1,
=IF(LEN(B1), SUM(SUMIFS(A$1:INDEX(A:A, MATCH(1E+99, A:A)),
B$1:INDEX(B:B, MATCH(1E+99, A:A)), LEFT(B1, ROW(INDIRECT("2:"&LEN(B1)))))), TEXT(,))
The above does not compensate for missing parents (null string). It totals everything it can find and uses zero for missing parents.
As a VBA UDF² in E1,
Function conditionalCumulativeSum(nums As Range, _
ids As Range, sib As Range, _
Optional nullOnBlank As Boolean = True)
Dim i As Integer
'truncate any full column reference to the UsedRange
Set nums = Intersect(nums, nums.Parent.UsedRange)
'match the nums and ids ranges
Set ids = ids.Resize(nums.Rows.Count, nums.Columns.Count)
For i = Len(sib.Value2) To 2 Step -1
If nullOnBlank And IsError(Application.Match(--Left(sib, i), ids, 0)) Then
conditionalCumulativeSum = vbNullString
Exit For
End If
conditionalCumulativeSum = conditionalCumulativeSum + _
Application.SumIfs(nums, ids, Left(sib, i))
Next i
If i = 0 Then conditionalCumulativeSum = vbNullString
End Function
The above defaults to return a null string when it encounters any missing parent through the hereditary chain. This can be turned off by adding FALSE as the optional fourth parameter and then the UDF will behave identically to the native formula.
Results from sample data
¹ Array formulas need to be finalized with Ctrl+Shift+Enter↵. If entered correctly, Excel with wrap the formula in braces (e.g. { and }). You do not type the braces in yourself. Once entered into the first cell correctly, they can be filled or copied down or right just like any other formula. Try and reduce your full-column references to ranges more closely representing the extents of your actual data. Array formulas chew up calculation cycles logarithmically so it is good practise to narrow the referenced ranges to a minimum. See Guidelines and examples of array formulas for more information.
² A User Defined Function (aka UDF) is placed into a standard module code sheet. Tap Alt+F11 and when the VBE opens, immediately use the pull-down menus to Insert ► Module (Alt+I,M). Paste the function code into the new module code sheet titled something like Book1 - Module1 (Code). Tap Alt+Q to return to your worksheet(s).
I have to retrieve only particular records whose sum value of size field is <=150.
I have table like below ...
userid size
1 70
2 100
3 50
4 25
5 120
6 90
The output should be ...
userid size
1 70
3 50
4 25
For example, if we add 70,50,25 we get 145 which is <=150.
How would I write a query to accomplish this?
Here's a query which will produce the above results:
SELECT * FROM `users` u
WHERE (select sum(size) from `users` where size <= u.size order by size) < 150
ORDER BY userid
However, the problem you describe of wanting the selection of users which would most closely fit into a given size, is a bin packing problem. This is an NP-Hard problem, and won't be easily solved with ANSI SQL. However, the above seems to return the right result, but in fact it simply starts with the smallest item, and continues to add items until the bin is full.
A general, more effective bin packing algorithm would is to start with the largest item and continue to add smaller ones as they fit. This algorithm would select users 5 and 4.
What you're looking for is a greedy algorithm. You can't really do this with one SQL statement.
It's similar to the subset sum problem. You are definitely going to be into exponential time ...
There are several ways to solve subset
sum in time exponential in N. The most
naïve algorithm would be to cycle
through all subsets of N numbers and,
for every one of them, check if the
subset sums to the right number. The
running time is of order O(2^N*N), since
there are 2N subsets and, to check
each subset, we need to sum at most N
elements.
Unless you can constrain the problem to smaller subsets.
According to your definition as it stands you could get any of these tables:
userid size userid size
1 70 2 100
userid size userid size
3 50 4 25
userid size userid size
5 120 6 90
userid size userid size
1 70 2 100
3 50 3 50
userid size userid size
1 70 2 100
4 25 4 25
userid size userid size
1 70 4 25
3 50 6 90
4 25
userid size userid size
4 25 3 50
5 120 6 90
SQL sucks at guessing. Do you mean to say you want the most users who's total size is under a certain limit? You'll need to create a temp table of all the combinations of users, then select the ones who's total size is less then the limit, then select the one with the most users, and possibly the lowest user ID or something. Either way, it won't be fast due to the first step.
But do you want to maximize the number of results or minimize or you simply don't care? first two cases is constraints optimization for which there should be solution using SQL, the latter (as mentioned above) requires greedy strategy.