9 You want to measure the running time of a tactic or function.\smallskip |
9 You want to measure the running time of a tactic or function.\smallskip |
10 |
10 |
11 {\bf Solution:} Time can be measured using the function |
11 {\bf Solution:} Time can be measured using the function |
12 @{ML start_timing} and @{ML end_timing}.\smallskip |
12 @{ML start_timing} and @{ML end_timing}.\smallskip |
13 |
13 |
14 Assume the following function defined in Isabelle. |
14 Suppose you defined the Ackermann function inside Isabelle. |
15 *} |
15 *} |
16 |
16 |
17 fun |
17 fun |
18 ackermann:: "(nat * nat) \<Rightarrow> nat" |
18 ackermann:: "(nat \<times> nat) \<Rightarrow> nat" |
19 where |
19 where |
20 "ackermann (0, n) = n + 1" |
20 "ackermann (0, n) = n + 1" |
21 | "ackermann (m, 0) = ackermann (m - 1, 1)" |
21 | "ackermann (m, 0) = ackermann (m - 1, 1)" |
22 | "ackermann (m, n) = ackermann (m - 1, ackermann (m, n - 1))" |
22 | "ackermann (m, n) = ackermann (m - 1, ackermann (m, n - 1))" |
23 |
23 |
24 text {* |
24 text {* |
25 We can now measure how long the simplifier takes to verify a datapoint |
25 You can measure how long the simplifier takes to verify a datapoint |
26 of this function. The timing can be done using the following wrapper function: |
26 of this function. The timing can be done using the following wrapper function: |
27 *} |
27 *} |
28 |
28 |
29 ML{*fun timing_wrapper tac st = |
29 ML{*fun timing_wrapper tac st = |
30 let |
30 let |
34 in |
34 in |
35 (warning (#message t_end); res) |
35 (warning (#message t_end); res) |
36 end*} |
36 end*} |
37 |
37 |
38 text {* |
38 text {* |
39 Note that this function also takes a state @{text "st"} as an argument and |
39 Note that this function, in addition to a tactic for which it measures the |
40 applies this state to the tactic. This is because tactics are lazy functions |
40 time, also takes a state @{text "st"} as argument and applies this state to |
41 and we need to force them to run, otherwise the timing will be meaningless. |
41 the tactic. The reason is that tactics are lazy functions and you need to force |
42 The used time will be calculated as the end time minus the start time. |
42 them to run, otherwise the timing will be meaningless. The time between start |
43 The wrapper can now be used in the proof |
43 and finish of the tactic will be calculated as the end time minus the start time. |
|
44 An example for the wrapper is the proof |
|
45 |
44 *} |
46 *} |
45 |
47 |
46 lemma "ackermann (3, 4) = 125" |
48 lemma "ackermann (3, 4) = 125" |
47 apply(tactic {* |
49 apply(tactic {* |
48 timing_wrapper (simp_tac (@{simpset} addsimps @{thms "nat_number"}) 1) *}) |
50 timing_wrapper (simp_tac (@{simpset} addsimps @{thms "nat_number"}) 1) *}) |
49 done |
51 done |
50 |
52 |
51 text {* |
53 text {* |
52 where it returns something on the cale of 3 seconds. |
54 where it returns something on the scale of 3 seconds. We choose to return |
|
55 this information as a string, but the timing information is also accessible |
|
56 in number format. |
53 |
57 |
54 \begin{readmore} |
58 \begin{readmore} |
55 Basic functions regarding timing are defined in @{ML_file |
59 Basic functions regarding timing are defined in @{ML_file |
56 "Pure/ML-Systems/polyml_common.ML"} (for the PolyML compiler). Some more |
60 "Pure/ML-Systems/polyml_common.ML"} (for the PolyML compiler). Some more |
57 advanced functions are defined in @{ML_file "Pure/General/output.ML"}. |
61 advanced functions are defined in @{ML_file "Pure/General/output.ML"}. |