Is nested Table an efficient way to generate nested list?

Question

If I want to have a nested list like 10 times of {1,2..,9,10}:

{{1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1,
2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 
2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2,
3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 
3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}}

I can easily understand that Table[i,{i,1,10}] will give me a sub list {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, so I can imagine if I want to have ten same sub lists I need to apply this table function for ten times, so I use Table[Table[i,{i,1,10}],{10}] to get what I want.

But I know for the language of Mathematica, things could happen a lot of miraculous than I thought, there must be some more elegant ways to manipulate lists. My question is asking do you think 'Table[Table[expr, ],{conditions}]' stuff are good implementations in terms of efficiency in my case?

Answer 1

Extract is much faster than all alternatives posted/mentioned so far.

Extract[{Range @ 5}, ConstantArray[{1}, 5]] // TeXForm

$\left( \begin{array}{ccccc} 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ \end{array} \right)$

Timings: Using @BobHanlon's timing setup:

n = 10^3;
{t1 = AbsoluteTiming[tab1 = Table[i, {n}, {i, 1, n}];][[1]], 
 t2 = AbsoluteTiming[tab2 = Table[Range[n], {n}];][[1]], 
 t3 = AbsoluteTiming[tab3 = Table[Table[i, {i, 1, n}], {n}];][[1]], 
 t4 = AbsoluteTiming[tab4 = ConstantArray[Range[n], n];][[1]],
 t5 = AbsoluteTiming[tab5 = Extract[{Range[n]}, ConstantArray[{1}, n]];][[1]]}

{5.844533, 0.104275, 0.750996, 0.095255, 0.001003}

% / t5 // Round[#, 1]&

{5825, 104, 748, 95, 1}

 tab1 == tab2 == tab3 == tab4 == tab5

True

For n = 10^4; we get

{t1, t2, t3, t4, t5}

{24.885165, 1.786752, 2.813479, 0.323860, 0.001003}

% / t5 // Round[#, 1]&

{24801, 1781, 2804, 323, 1}

Update: Part + ConstantArray

{Range[5]}[[ConstantArray[1, 5]]] // TeXForm

$\left( \begin{array}{ccccc} 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 2 & 3 & 4 & 5 \\ \end{array} \right)$

This is as fast or faster than Extract:

n = 10^4;
{t5 = AbsoluteTiming[tab5 = Extract[{Range[n]}, ConstantArray[{1}, n]];][[1]],
 t6 = AbsoluteTiming[tab6 = {Range[n]}[[ConstantArray[1, n]]];][[1]]}

{0.003003, 0.001006}

tab5 == tab6

True

Answer 2

The list can be constructed by more specific functions than Table:

ConstantArray[Range[10], 10]

(*{{1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1,
2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 
2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2,
3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 
3, 4, 5, 6, 7, 8, 9, 10}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}}*)

Answer 3

The relative timings can change significantly with n

n = 10^2;

{t1 = RepeatedTiming[tab1 = Table[i, {n}, {i, 1, n}];][[1]],
 t2 = RepeatedTiming[tab2 = Table[Range[n], {n}];][[1]],
 t3 = RepeatedTiming[tab3 = Table[Table[i, {i, 1, n}], {n}];][[1]], 
 t4 = RepeatedTiming[tab4 = ConstantArray[Range[n], n];][[1]]}

(* {0.00069, 0.00008, 0.00069, 4.*10^-6} *)

%/t4 // N

(* {180.947, 19.9453, 182.789, 1.} *)

Note that the nested tables is the slowest.

Verifying that all four approaches give the same result

tab1 === tab2 === tab3 === tab4

(* True *)

Increasing n

n = 10^3;

{t1 = RepeatedTiming[tab1 = Table[i, {n}, {i, 1, n}];][[1]],
 t2 = RepeatedTiming[tab2 = Table[Range[n], {n}];][[1]],
 t3 = RepeatedTiming[tab3 = Table[Table[i, {i, 1, n}], {n}];][[1]], 
 t4 = RepeatedTiming[tab4 = ConstantArray[Range[n], n];][[1]]}

(* {0.064, 0.010, 0.0099, 0.00075} *)

%/t4 // N

(* {85.5733, 13.388, 13.181, 1.} *)

Note that the nested tables is no longer the slowest.

tab1 === tab2 === tab3 === tab4

(* True *)

Further increasing n

n = 10^4;

{t1 = RepeatedTiming[tab1 = Table[i, {n}, {i, 1, n}];][[1]],
 t2 = RepeatedTiming[tab2 = Table[Range[n], {n}];][[1]],
 t3 = RepeatedTiming[tab3 = Table[Table[i, {i, 1, n}], {n}];][[1]], 
 t4 = RepeatedTiming[tab4 = ConstantArray[Range[n], n];][[1]]}

(* {10.3, 1.8, 0.93, 0.51} *)

%/t4 // N

(* {20.1379, 3.51058, 1.83164, 1.} *)

tab1 === tab2 === tab3 === tab4

(* True *)