I have the following DataFrame :
Police Product PV1 PV2 PV3 PM1 PM2 PM3
0 1 AA 10 8 14 150 145 140
1 2 AB 25 4 7 700 650 620
2 3 AA 13 22 5 120 80 60
3 4 AA 12 6 12 250 170 120
4 5 AB 10 13 5 500 430 350
5 6 BC 7 21 12 1200 1000 900
PV1 is the item PV for year 1, PV2 for year 2, ....
I would like to combine reshaping and group by operations + some renaming stuffs to obtain the DF below :
Product Item Year1 Year2 Year3
0 AA PV 35 36 31
1 AA PM 520 395 320
2 AB PV 35 17 12
3 AB PM 1200 1080 970
4 BC PV 7 21 12
5 BC PM 1200 1000 900
It makes a group by operation on product name and reshape the DF to pass the item as a column and put the sum of each in new columns years.
I found a way to do it in Python but I am now looking for a solution passing my code in Julia.
No problem for the groupby operation, but I have more issues with the reshaping / renaming part.
If you have any idea, I would be very grateful.
Thanks for any help
Edit :
As you recommended, I have installed Julia 1.5 and updated the DataFrames pkg to 0.22 version. As a result, the code runs well. The only remaining issue is related to the non constant lenght of column names in my real DF, which makes the transform part of the code not completly suitable. I will search for a way to split char/num with regular expression.
Thanks a lot for your time and sorry for the mistakes on editing.
There are probably several ways how you can do it. Here is an example using in-built functions (also taking advantage of several advanced features at once, so if you have any questions regarding the code please comment and I can explain):
julia> using CSV, DataFrames, Chain
julia> str = """
Police Product PV1 PV2 PV3 PM1 PM2 PM3
1 AA 10 8 14 150 145 140
2 AB 25 4 7 700 650 620
3 AA 13 22 5 120 80 60
4 AA 12 6 12 250 170 120
5 AB 10 13 5 500 430 350
6 BC 7 21 12 1200 1000 900""";
julia> #chain str begin
IOBuffer
CSV.read(DataFrame, ignorerepeated=true, delim=" ")
groupby(:Product)
combine(names(df, r"\d") .=> sum, renamecols=false)
stack(Not(:Product))
transform!(:variable => ByRow(x -> (first(x, 2), last(x, 1))) => [:Item, :Year])
unstack([:Product, :Item], :Year, :value, renamecols = x -> Symbol("Year", x))
sort!(:Product)
end
6×5 DataFrame
Row │ Product Item Year1 Year2 Year3
│ String String Int64? Int64? Int64?
─────┼─────────────────────────────────────────
1 │ AA PV 35 36 31
2 │ AA PM 520 395 320
3 │ AB PV 35 17 12
4 │ AB PM 1200 1080 970
5 │ BC PV 7 21 12
6 │ BC PM 1200 1000 900
I used Chain.jl just to show how it can be employed in practice (but of course it is not needed).
You can add #aside show(_) annotation after any stage of the processing to see the results of the processing steps.
Edit:
Is this the regex you need (split non-digit characters followed by digit characters)?
julia> match(r"([^\d]+)(\d+)", "fsdfds123").captures
2-element Array{Union{Nothing, SubString{String}},1}:
"fsdfds"
"123"
Then just write:
ByRow(x -> match(r"([^\d]+)(\d+)", x).captures)
as your transformation
Select rows after the maximum minimum valor found
Input file
22 101 5
23 102 5
24 103 5
25 104 23
26 105 25
27 106 21
28 107 20
29 108 8
30 109 6
31 110 7
To figure out my problem, I tried to subtract column 3 and print the lines after the minimum value found in column 4. In this case after row 7
awk '{$4 = $3 - prev3; prev3 = $3; print $0}' file
22 101 5
23 102 5 0
24 103 5 0
25 104 2 18
26 105 2 2
27 106 2 -4
28 107 2 -1
29 108 8 -12
30 109 6 -2
31 110 7 1
Desired Output
29 108 8
30 109 6
31 110 7
I believe there is better and easy way to get same output.
Thanks in advance
You need to process the same file twice:
Find out the line number of the min value
Print the line and the lines after it
Like this:
awk 'NR==FNR{v=$3-prev3;prev3=$3;if(NR==2||v<m){m=v;ln=NR};next}FNR>=ln' file file
Explanation:
# This condition is true as long as we process the file the first time
NR==FNR {
# Your calculation
v=$3-prev3
prev3=$3
# If NR==2, meaning in row 2 we initialize m and ln.
# Otherwise check if v is the new minimum and set m and ln.
if(NR==2 || v<m){
# Set m and ln when v is the new minimum
m=v
ln=NR
}
next # Skip the conditional below
}
# This condition will be only evaluated when we parse the file
# the second time. (because of the "next" statement above)
# When the line number is greater or equal than "ln" print it.
# (print is the default action)
FNR>=ln
Tensorflow Graph Transforms page https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/graph_transforms/README.md shows how to use strip_unused_nodes.
But how to know the right values of X and Y in strip_unused_nodes(type=X, shape="y0,y1,y3,3") for my model?
Output of summarize_graph on my MobileNetV2 model :
Found 1 possible inputs: (name=image_tensor, type=uint8(4), shape=[?,?,?,3])
No variables spotted.
Found 4 possible outputs: (name=detection_boxes, op=Identity) (name=detection_scores, op=Identity) (name=detection_classes, op=Identity) (name=num_detections, op=Identity)
Found 3457096 (3.46M) const parameters, 0 (0) variable parameters, and 623 control_edges
Op types used: 1707 Const, 525 Identity, 277 Mul, 194 Add, 170 Reshape, 147 GatherV2, 133 Sub, 117 Minimum, 98 Slice, 92 Maximum, 77 ConcatV2, 77 Cast, 64 Rsqrt, 60 StridedSlice, 59 Relu6, 55 Conv2D, 54 Pack, 52 Greater, 49 Shape, 46 Split, 46 Where, 45 ExpandDims, 40 Fill, 37 Tile, 33 RealDiv, 33 DepthwiseConv2dNative, 30 Range, 29 Switch, 27 Unpack, 26 Enter, 25 Squeeze, 25 ZerosLike, 23 NonMaxSuppressionV2, 14 Merge, 12 BiasAdd, 12 FusedBatchNorm, 11 TensorArrayV3, 8 NextIteration, 6 TensorArrayWriteV3, 6 TensorArraySizeV3, 6 Sqrt, 6 Exit, 6 TensorArrayGatherV3, 5 TensorArrayScatterV3, 5 TensorArrayReadV3, 3 Rank, 3 Equal, 3 Transpose, 3 Assert, 2 Exp, 2 Less, 2 LoopCond, 1 All, 1 TopKV2, 1 Size, 1 Sigmoid, 1 ResizeBilinear, 1 Placeholder
To use with tensorflow/tools/benchmark:benchmark_model try these arguments:
bazel run tensorflow/tools/benchmark:benchmark_model -- --graph=/home/ubuntu/model-optimization/frozen_inference_graph.pb --show_flops --input_layer=image_tensor --input_layer_type=uint8 --input_layer_shape=-1,-1,-1,3 --output_layer=detection_boxes,detection_scores,detection_classes,num_detections
I believe you should copy the input layer dims, you can find in the .ascii file of your model
I am trying to make an auctions system but can not figure out the logical conditions for doing so..
Lets say that I have 10 credit
$credit
I have already bet 5 credits on another auction... so I owe 5 from 10 $owe
I thus have 5 available... $available = $credit - $owe (=5)
I bet 3 from available (on a different item)...
I wish to bet again 4 (cancel 3, update to 4), but credit available is now $available - 3 (=2)
Can't find a logical solution.... written in code.
What is the condition for setting a bet???
Made up a matrix with the dependence between variables:
bet available owe lastbet
1 10 10 0
2 9 11 1
3 7 13 2
4 4 16 3
5 0 20 4
6 -5 25 5
7 -11 31 6
8 -18 38 7
9 -26 46 8
10 -35 55 9
11 -45 65 10
Need to translate it into a condition statement.... (the next row would not meet the conditions)
The condition should fail on the 11th row....
Based on the Matrix... I found out that the condition is:
if ($bet <= (($owe + $available) / 2)) {}
Not very intuitive......
In this dataframe, column key values correspond to integer notation of each song key.
df
track key
0 Last Resort 4
1 Casimir Pulaski Day 8
2 Glass Eyes 8
3 Ohio - Live At Massey Hall 1971 7
4 Ballad of a Thin Man 11
5 Can You Forgive Her? 11
6 The Only Thing 3
7 Goodbye Baby (Baby Goodbye) 4
8 Heart Of Stone 0
9 Ohio 0
10 the gate 2
11 Clampdown 2
12 Cry, Cry, Cry 4
13 What's Happening Brother 8
14 Stupid Girl 11
15 I Don't Wanna Play House 7
16 Inner City Blues (Make Me Wanna Holler) 11
17 The Lonesome Death of Hattie Carroll 4
18 Paint It, Black - (Original Single Mono Version) 5
19 Let Him Run Wild 11
20 Undercover (Of The Night) - Remastered 5
21 Between the Bars 7
22 Like a Rolling Stone 0
23 Once 2
24 Pale Blue Eyes 5
25 The Way You Make Me Feel - 2012 Remaster 1
26 Jeremy 2
27 The Entertainer 7
28 Pressure 9
29 Play With Fire - Mono Version / Remastered 2002 2
30 D-I-V-O-R-C-E 9
31 Big Shot 0
32 What's Going On 1
33 Folsom Prison Blues - Live 0
34 American Woman 1
35 Cocaine Blues - Live 8
36 Jesus, etc. 5
the notation is as follows:
'C' --> 0
'C#'--> 1
'D' --> 2
'Eb'--> 3
'E' --> 4
'F' --> 5
'F#'--> 6
'G' --> 7
'Ab'--> 8
'A' --> 9
'Bb'--> 10
'B' --> 11
what is specific about this notation is that 11 is closer to 0 than 2, for instance.
GOAL:
given an input_notation = 0, I would like to sort according to closeness to key 0, or 'C'.
you can get closest value by doing:
closest_key = (input_notation -1) % 12
so I would like to sort according to this logic, having on top input_notation values and then closest matches, like so:
8 Heart Of Stone 0
9 Ohio 0
22 Like a Rolling Stone 0
31 Big Shot 0
33 Folsom Prison Blues - Live 0
(...)
I have tried:
v = df[['key']].values
df = df.iloc[np.lexsort(np.abs(v - (input_notation - 1) %12 ).T)]
but this does not work..
any clues?
You can define the closeness firstly and then use argsort with iloc to sort the data frame:
input_notation = 0
# define the closeness or distance
diff = (df.key - input_notation).abs()
closeness = np.minimum(diff, 12 - diff)
# use argsort to calculate the sorting index, and iloc to reorder the data frame
closest_to_input = df.iloc[closeness.argsort(kind='mergesort')]
closest_to_input.head()
# track key
#8 Heart Of Stone 0
#9 Ohio 0
#22 Like a Rolling Stone 0
#31 Big Shot 0
#33 Folsom Prison Blues - Live 0