diff --git a/articles/aorsf.html b/articles/aorsf.html index baf35a4f..166911fb 100644 --- a/articles/aorsf.html +++ b/articles/aorsf.html @@ -174,11 +174,11 @@

Variable importance orsf_vi_negate(orsf_fit) #> bili copper sex age protime stage -#> 0.121621470 0.047409632 0.036160756 0.023453468 0.023036042 0.021079271 +#> 0.121610913 0.047724351 0.035987586 0.023472835 0.023013176 0.020959499 #> albumin ascites ast chol edema trt -#> 0.019171998 0.014530950 0.011655566 0.011520295 0.007794807 0.007008191 +#> 0.019390481 0.014457412 0.011659769 0.011544114 0.007728580 0.006902781 #> spiders alk.phos hepato trig platelet -#> 0.004761188 0.003541539 0.003034140 0.002904610 0.001811919 +#> 0.004752062 0.003464822 0.003092596 0.002763978 0.001742350

  • You can also compute variable importance using permutation, a @@ -187,13 +187,13 @@

    Variable importance orsf_vi_permute(orsf_fit) #> bili copper stage age protime -#> 0.0564823961 0.0223855853 0.0142026653 0.0135127275 0.0125608069 +#> 0.0563410637 0.0223514680 0.0140882841 0.0136856141 0.0125390739 #> ascites albumin ast edema chol -#> 0.0115005317 0.0111892767 0.0062583603 0.0059328841 0.0049633185 +#> 0.0114579347 0.0112609663 0.0063723451 0.0058601940 0.0049876670 #> spiders sex alk.phos trig hepato -#> 0.0047712233 0.0029539032 0.0021892028 0.0012938183 0.0009293975 +#> 0.0046062839 0.0029547526 0.0020493151 0.0013490654 0.0009248580 #> platelet trt -#> 0.0003236044 -0.0010373711 +#> 0.0002087435 -0.0010805504

  • A faster alternative to permutation and negation importance is @@ -203,11 +203,11 @@

    Variable importance orsf_vi_anova(orsf_fit) #> ascites bili edema copper age albumin protime -#> 0.42817294 0.30059141 0.28426857 0.21943005 0.20141521 0.19357326 0.18584305 +#> 0.42926829 0.30005133 0.28307417 0.21974605 0.20161698 0.19367284 0.18475384 #> stage ast chol sex spiders hepato trig -#> 0.16470910 0.15964407 0.15792194 0.15172667 0.14802844 0.13351750 0.12661567 +#> 0.16479913 0.15931052 0.15823438 0.15274199 0.14765751 0.13382443 0.12667896 #> alk.phos platelet trt -#> 0.12163561 0.10985915 0.09887394 +#> 0.12193232 0.10931900 0.09903714

    • @@ -259,7 +259,7 @@

    What about the original ORSF?) ) #> user system elapsed -#> 4.722 0.004 4.728 +#> 4.175 0.000 4.175 # and how long it takes to fit 50 cph trees print( @@ -271,11 +271,11 @@

    What about the original ORSF?) ) #> user system elapsed -#> 0.056 0.000 0.056 +#> 0.051 0.000 0.051 t1['elapsed'] / t2['elapsed'] #> elapsed -#> 84.42857 +#> 81.86275

    aorsf and other machine learning software diff --git a/articles/fast.html b/articles/fast.html index 9e827451..f86e102c 100644 --- a/articles/fast.html +++ b/articles/fast.html @@ -150,7 +150,7 @@

    Use orsf_control_fast()# control_fast() is much faster time_net['elapsed'] / time_fast['elapsed'] #> elapsed -#> 38.38776

    +#> 33.74545

    Use n_thread @@ -183,10 +183,10 @@

    Use n_threadtime_1_thread['elapsed'] / time_5_thread['elapsed'] #> elapsed -#> 0.3702316 +#> 0.3305418 time_1_thread['elapsed'] / time_auto_thread['elapsed'] #> elapsed -#> 0.7542064

    +#> 0.7340032

    Because R is a single threaded language, multi-threading cannot be applied when orsf() needs to call R functions from C++, which occurs when a customized R function is used to find linear @@ -222,8 +222,8 @@

    Do less # about two times faster than auto thread with defaults time_auto_thread['elapsed'] / time_lightweight['elapsed'] -#> elapsed -#> 2.560515 +#> elapsed +#> 2.53372

    While these default values do make orsf() run slower, they also usually make its predictions more accurate or make the fit easier to interpret.

    @@ -241,11 +241,11 @@

    Show progress= 0, n_tree = 500, verbose_progress = TRUE) -#> Growing trees: 18%. ~ time remaining: 8 seconds. -#> Growing trees: 39%. ~ time remaining: 6 seconds. -#> Growing trees: 56%. ~ time remaining: 4 seconds. -#> Growing trees: 75%. ~ time remaining: 2 seconds. -#> Growing trees: 96%. ~ time remaining: 0 seconds. +#> Growing trees: 17%. ~ time remaining: 9 seconds. +#> Growing trees: 33%. ~ time remaining: 8 seconds. +#> Growing trees: 50%. ~ time remaining: 6 seconds. +#> Growing trees: 68%. ~ time remaining: 3 seconds. +#> Growing trees: 86%. ~ time remaining: 1 seconds. #> Growing trees: 100%. #> Computing predictions: 100%. diff --git a/articles/oobag.html b/articles/oobag.html index 17acd8a0..9ebb5d39 100644 --- a/articles/oobag.html +++ b/articles/oobag.html @@ -139,9 +139,9 @@

    Out-of-bag predictions and error# what is the output from this function? fit$eval_oobag$stat_values #> [,1] -#> [1,] 0.8415542 +#> [1,] 0.8403563

    The out-of-bag estimate of Harrell’s C-statistic (the default method -to evaluate out-of-bag predictions) is 0.8415542.

    +to evaluate out-of-bag predictions) is 0.8403563.

    Monitoring out-of-bag error @@ -204,7 +204,7 @@

    User-supplied out-of-bag oobag_fun_brier(y_mat = pbc_orsf[,c('time', 'status')], s_vec = fit$pred_oobag) -#> [1] 0.113788

    +#> [1] 0.112774

    Second, you can pass your function into orsf(), and it will be used in place of Harrell’s C-statistic:

    @@ -327,12 +327,12 @@ 
    User-supplied function
                        importance = 'negate')
 
 fit_tdep_cstat$importance
-#>        bili      copper         sex       stage     albumin         age 
-#> 0.109547520 0.048633340 0.031563870 0.024989400 0.023426140 0.020986680 
-#>     protime        chol     ascites         ast       edema         trt 
-#> 0.019175230 0.018283250 0.017884150 0.015850050 0.008054587 0.007685040 
-#>      hepato     spiders    alk.phos    platelet        trig 
-#> 0.006622980 0.006348670 0.004479150 0.002303970 0.002085840
    +#> bili copper sex stage albumin protime age +#> 0.13229516 0.04918793 0.03788613 0.02751833 0.02658144 0.02440104 0.02172651 +#> ascites chol ast edema trt spiders hepato +#> 0.01733133 0.01295822 0.00851655 0.00793299 0.00534734 0.00429056 0.00415910 +#> trig platelet alk.phos +#> 0.00274328 0.00204262 0.00120536

    Notes diff --git a/articles/oobag_files/figure-html/unnamed-chunk-2-1.png b/articles/oobag_files/figure-html/unnamed-chunk-2-1.png index ea384204..b2d9ef89 100644 Binary files a/articles/oobag_files/figure-html/unnamed-chunk-2-1.png and b/articles/oobag_files/figure-html/unnamed-chunk-2-1.png differ diff --git a/articles/oobag_files/figure-html/unnamed-chunk-4-1.png b/articles/oobag_files/figure-html/unnamed-chunk-4-1.png index 22eb2e83..a06b106a 100644 Binary files a/articles/oobag_files/figure-html/unnamed-chunk-4-1.png and b/articles/oobag_files/figure-html/unnamed-chunk-4-1.png differ diff --git a/articles/oobag_files/figure-html/unnamed-chunk-7-1.png b/articles/oobag_files/figure-html/unnamed-chunk-7-1.png index 94227b70..a67994f2 100644 Binary files a/articles/oobag_files/figure-html/unnamed-chunk-7-1.png and b/articles/oobag_files/figure-html/unnamed-chunk-7-1.png differ diff --git a/articles/oobag_files/figure-html/unnamed-chunk-8-1.png b/articles/oobag_files/figure-html/unnamed-chunk-8-1.png index b8ba3103..ddee37c7 100644 Binary files a/articles/oobag_files/figure-html/unnamed-chunk-8-1.png and b/articles/oobag_files/figure-html/unnamed-chunk-8-1.png differ diff --git a/articles/pd.html b/articles/pd.html index 06e0c2ae..414f4ba0 100644 --- a/articles/pd.html +++ b/articles/pd.html @@ -153,12 +153,12 @@

    Three ways to compute PDpd_inb <- orsf_pd_inb(fit, pred_spec = list(bili = 1:5)) pd_inb -#> pred_horizon bili mean lwr medn upr -#> 1: 1826.25 1 0.2185157 0.01435497 0.0970635 0.8244006 -#> 2: 1826.25 2 0.2537791 0.03094375 0.1375934 0.8443459 -#> 3: 1826.25 3 0.2980089 0.05332399 0.1956020 0.8578012 -#> 4: 1826.25 4 0.3532776 0.09828527 0.2759155 0.8699563 -#> 5: 1826.25 5 0.3950478 0.14481534 0.2998934 0.8782313

    +#> pred_horizon bili mean lwr medn upr +#> 1: 1826.25 1 0.2189551 0.01432475 0.09631307 0.8249006 +#> 2: 1826.25 2 0.2542372 0.03084190 0.13608413 0.8448459 +#> 3: 1826.25 3 0.2984207 0.05324065 0.19420673 0.8583845 +#> 4: 1826.25 4 0.3537711 0.09798050 0.27685533 0.8704563 +#> 5: 1826.25 5 0.3955910 0.14622431 0.30021043 0.8776599

  • using out-of-bag predictions for the training data

    @@ -168,11 +168,11 @@

    Three ways to compute PD pd_oob #> pred_horizon bili mean lwr medn upr -#> 1: 1826.25 1 0.2181099 0.01218789 0.1004867 0.8311560 -#> 2: 1826.25 2 0.2540728 0.02447359 0.1455362 0.8494611 -#> 3: 1826.25 3 0.2980036 0.04854875 0.1998871 0.8641245 -#> 4: 1826.25 4 0.3550648 0.09984838 0.2693004 0.8647180 -#> 5: 1826.25 5 0.3958124 0.14627606 0.3262588 0.8741973 +#> 1: 1826.25 1 0.2183177 0.01218789 0.1004630 0.8304537 +#> 2: 1826.25 2 0.2542048 0.02447359 0.1446725 0.8484741 +#> 3: 1826.25 3 0.2981275 0.04854875 0.2004516 0.8641502 +#> 4: 1826.25 4 0.3552433 0.10141417 0.2666887 0.8647438 +#> 5: 1826.25 5 0.3959304 0.14768055 0.3222995 0.8742230

  • using predictions for a new set of data

    @@ -184,11 +184,11 @@

    Three ways to compute PD pd_test #> pred_horizon bili mean lwr medn upr -#> 1: 1826.25 1 0.2643299 0.01758300 0.2098936 0.8408190 -#> 2: 1826.25 2 0.2989177 0.04071722 0.2510869 0.8551052 -#> 3: 1826.25 3 0.3431601 0.06852192 0.3050132 0.8668560 -#> 4: 1826.25 4 0.3967849 0.11810058 0.3592089 0.8724041 -#> 5: 1826.25 5 0.4387927 0.16022391 0.4094224 0.8806860 +#> 1: 1826.25 1 0.2644314 0.01758300 0.2098936 0.8412690 +#> 2: 1826.25 2 0.2990618 0.04063388 0.2514774 0.8555552 +#> 3: 1826.25 3 0.3432566 0.06843859 0.3059299 0.8673060 +#> 4: 1826.25 4 0.3968302 0.11801725 0.3595564 0.8729358 +#> 5: 1826.25 5 0.4388773 0.16038177 0.4095296 0.8811360

  • in-bag PD indicates relationships that the model has learned during training. This is helpful if your goal is to interpret the @@ -222,8 +222,8 @@

    One variable, one horizon pd_sex #> pred_horizon sex mean lwr medn upr -#> 1: 1826.25 m 0.3636796 0.048601116 0.2450203 0.9536765 -#> 2: 1826.25 f 0.3043006 0.009488887 0.1552474 0.9558234 +#> 1: 1826.25 m 0.3640961 0.048985800 0.2463878 0.9537099 +#> 2: 1826.25 f 0.3051194 0.009487996 0.1574919 0.9562044

    The output shows that the expected predicted mortality risk for men is substantially higher than women at 5 years after baseline.

    @@ -276,13 +276,13 @@

    Multiple variables, marginallypd_two_vars #> pred_horizon variable value level mean lwr medn upr -#> 1: 1826.25 sex NA m 0.3636796 0.048601116 0.2450203 0.9536765 -#> 2: 1826.25 sex NA f 0.3043006 0.009488887 0.1552474 0.9558234 -#> 3: 1826.25 bili 1 <NA> 0.2485802 0.011564981 0.1291795 0.9002107 -#> 4: 1826.25 bili 2 <NA> 0.2998066 0.037689685 0.1917402 0.9201019 -#> 5: 1826.25 bili 3 <NA> 0.3550793 0.065043688 0.2640451 0.9320806 -#> 6: 1826.25 bili 4 <NA> 0.4015254 0.092934248 0.3215990 0.9422651 -#> 7: 1826.25 bili 5 <NA> 0.4369651 0.113674025 0.3658016 0.9445660 +#> 1: 1826.25 sex NA m 0.3640961 0.048985800 0.2463878 0.9537099 +#> 2: 1826.25 sex NA f 0.3051194 0.009487996 0.1574919 0.9562044 +#> 3: 1826.25 bili 1 <NA> 0.2488347 0.011917581 0.1329502 0.9002107 +#> 4: 1826.25 bili 2 <NA> 0.2998894 0.036872735 0.1917267 0.9201374 +#> 5: 1826.25 bili 3 <NA> 0.3550726 0.064920296 0.2624318 0.9317436 +#> 6: 1826.25 bili 4 <NA> 0.4017419 0.093708482 0.3227094 0.9422651 +#> 7: 1826.25 bili 5 <NA> 0.4373928 0.114171537 0.3651643 0.9445660

    Now would it be tedious if you wanted to do this for all the variables? You bet. That’s why we made a function for that. As a bonus, the printed output is sorted from most to least important variables.

    @@ -296,133 +296,133 @@

    Multiple variables, marginally#> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 0.80 0.2426718 0.1279552 0.04984847 0.3943018 -#> 1.4 0.2650919 0.1511846 0.06806841 0.4204671 -#> 3.5 0.3813791 0.3006846 0.17681607 0.5728243 +#> 0.80 0.2429676 0.1331339 0.05061303 0.3945576 +#> 1.4 0.2653392 0.1505044 0.06736587 0.4191150 +#> 3.5 0.3816531 0.3008428 0.17477323 0.5710884 #> #> -- copper (VI Rank: 2) ----------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 43 0.2711192 0.1364832 0.05085395 0.4742508 -#> 74 0.2919267 0.1648166 0.06045514 0.5024584 -#> 129 0.3474690 0.2264488 0.11666296 0.5862955 +#> 43 0.2714431 0.1379538 0.05205695 0.4730195 +#> 74 0.2921292 0.1641790 0.06105152 0.5005068 +#> 129 0.3475959 0.2266914 0.11717755 0.5878593 #> #> -- sex (VI Rank: 3) -------------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> m 0.3636796 0.2450203 0.13134161 0.5913622 -#> f 0.3043006 0.1552474 0.05258204 0.5581216 +#> m 0.3640961 0.2463878 0.13204014 0.5879171 +#> f 0.3051194 0.1574919 0.05286845 0.5557254 #> #> -- age (VI Rank: 4) -------------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 42 0.2776103 0.1367878 0.04780102 0.4702551 -#> 50 0.3111019 0.1894254 0.05323669 0.5350162 -#> 57 0.3463241 0.2451927 0.07903236 0.5834995 +#> 42 0.2778325 0.1336068 0.04734564 0.4720067 +#> 50 0.3113836 0.1901663 0.05411322 0.5347301 +#> 57 0.3467609 0.2463643 0.07981925 0.5858235 #> -#> -- protime (VI Rank: 5) ---------------------------------- +#> -- stage (VI Rank: 5) ------------------------------------ #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 10 0.2879205 0.1465481 0.05326982 0.5140322 -#> 11 0.3046243 0.1672641 0.05801802 0.5359534 -#> 11 0.3293300 0.1909532 0.07686704 0.5723693 +#> 1 0.2685206 0.1421588 0.04919176 0.4695496 +#> 2 0.2802496 0.1363632 0.04701566 0.4930937 +#> 3 0.3033486 0.1666220 0.05623057 0.5624443 +#> 4 0.3504840 0.2123025 0.09043543 0.6089733 #> -#> -- stage (VI Rank: 6) ------------------------------------ +#> -- protime (VI Rank: 6) ---------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 1 0.2681668 0.1380835 0.05029263 0.4674358 -#> 2 0.2796755 0.1378881 0.04858443 0.4914603 -#> 3 0.3026613 0.1627288 0.05474229 0.5632632 -#> 4 0.3496420 0.2114635 0.09079857 0.6061424 +#> 10 0.2883981 0.1473313 0.05315309 0.5132458 +#> 11 0.3051930 0.1685960 0.05892726 0.5340487 +#> 11 0.3300183 0.1907082 0.07462220 0.5758706 #> #> -- albumin (VI Rank: 7) ---------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 3.3 0.3261637 0.2004845 0.05692253 0.6069932 -#> 3.5 0.3034620 0.1750030 0.05259209 0.5646301 -#> 3.8 0.2884844 0.1592976 0.05317897 0.5088161 +#> 3.3 0.3263344 0.2038157 0.05762392 0.6060690 +#> 3.5 0.3036941 0.1778324 0.05255517 0.5689823 +#> 3.8 0.2888047 0.1585038 0.05334673 0.5117978 #> #> -- ascites (VI Rank: 8) ---------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 0 0.3050133 0.1684231 0.05309663 0.5701751 -#> 1 0.4807880 0.3999962 0.27558768 0.6522981 +#> 0 0.3059013 0.1698154 0.05368955 0.5671727 +#> 1 0.4817274 0.4021561 0.27869190 0.6563498 #> #> -- ast (VI Rank: 9) -------------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 82 0.2908882 0.1513240 0.04828883 0.5229041 -#> 117 0.3089912 0.1725868 0.05538041 0.5554213 -#> 153 0.3310741 0.1871842 0.06924725 0.5943776 +#> 82 0.2913576 0.1520544 0.04950740 0.5218531 +#> 117 0.3094541 0.1740707 0.05559358 0.5539702 +#> 153 0.3313814 0.1885225 0.06812852 0.5950395 #> #> -- chol (VI Rank: 10) ------------------------------------ #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 250 0.2947493 0.1517498 0.04584494 0.4921210 -#> 310 0.3043914 0.1701447 0.05171725 0.5278263 -#> 401 0.3273137 0.1931407 0.07655706 0.5632925 +#> 250 0.2952563 0.1562400 0.04590143 0.4954350 +#> 310 0.3047719 0.1722904 0.05162941 0.5272168 +#> 401 0.3276568 0.1979026 0.07516577 0.5628684 #> #> -- trt (VI Rank: 11) ------------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> d_penicill_main 0.3169260 0.1832524 0.06007976 0.5796638 -#> placebo 0.3103546 0.1698117 0.05309663 0.5662393 +#> d_penicill_main 0.3174049 0.1857576 0.06024488 0.5767564 +#> placebo 0.3108586 0.1704020 0.05324601 0.5612475 #> #> -- edema (VI Rank: 12) ----------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 0 0.3002624 0.1504964 0.05258204 0.5595393 -#> 0.5 0.3654657 0.2587416 0.10582704 0.6117316 -#> 1 0.4734729 0.3835527 0.27861278 0.6677314 +#> 0 0.3010149 0.1554307 0.05306503 0.5553444 +#> 0.5 0.3657927 0.2577948 0.10587652 0.6128551 +#> 1 0.4738360 0.3857446 0.27691332 0.6636638 #> #> -- hepato (VI Rank: 13) ---------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 0 0.2961537 0.1655051 0.04855092 0.5286287 -#> 1 0.3283298 0.1921262 0.06411885 0.5836396 +#> 0 0.2962090 0.1662321 0.05017602 0.5300380 +#> 1 0.3291752 0.1966199 0.06516627 0.5814178 #> #> -- trig (VI Rank: 14) ------------------------------------ #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 85 0.3044697 0.1600411 0.05319142 0.5347385 -#> 108 0.3106696 0.1654541 0.05589819 0.5305391 -#> 151 0.3251355 0.1923483 0.06240310 0.5549847 +#> 85 0.3049929 0.1612779 0.05452157 0.5375834 +#> 108 0.3111416 0.1656046 0.05702487 0.5315042 +#> 151 0.3253408 0.1930412 0.06238556 0.5552340 #> #> -- spiders (VI Rank: 15) --------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 0 0.3015973 0.1572040 0.05135863 0.5571251 -#> 1 0.3467316 0.2258213 0.09670998 0.5929005 +#> 0 0.3021023 0.1605935 0.05133804 0.5568563 +#> 1 0.3473323 0.2252972 0.09608644 0.5944446 #> #> -- alk.phos (VI Rank: 16) -------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 922 0.3126952 0.1742767 0.05466852 0.5832064 -#> 1278 0.3129036 0.1773399 0.05205471 0.5760135 -#> 2068 0.3162769 0.1753477 0.05602455 0.5857620 +#> 922 0.3132858 0.1759020 0.05488385 0.5859220 +#> 1278 0.3134748 0.1769519 0.05406768 0.5755742 +#> 2068 0.3167539 0.1777876 0.05640827 0.5830347 #> #> -- platelet (VI Rank: 17) -------------------------------- #> #> |---------------- Risk ----------------| #> Value Mean Median 25th % 75th % -#> 200 0.3182309 0.1720752 0.05991039 0.5820997 -#> 257 0.3120849 0.1677648 0.05375493 0.5762923 -#> 318 0.3073012 0.1738184 0.05505376 0.5522198 +#> 200 0.3188386 0.1745985 0.05972147 0.5792435 +#> 257 0.3128024 0.1697741 0.05445992 0.5779811 +#> 318 0.3081249 0.1759292 0.05602966 0.5525766 #> #> Predicted risk at time t = 1826.25 for top 17 predictors

    It’s easy enough to turn this ‘summary’ object into a @@ -431,12 +431,12 @@

    Multiple variables, marginally head(as.data.table(pd_smry)) #> variable importance Value Mean Median 25th % 75th % -#> 1: bili 0.11650695 0.80 0.2426718 0.1279552 0.04984847 0.3943018 -#> 2: bili 0.11650695 1.4 0.2650919 0.1511846 0.06806841 0.4204671 -#> 3: bili 0.11650695 3.5 0.3813791 0.3006846 0.17681607 0.5728243 -#> 4: copper 0.04768827 43 0.2711192 0.1364832 0.05085395 0.4742508 -#> 5: copper 0.04768827 74 0.2919267 0.1648166 0.06045514 0.5024584 -#> 6: copper 0.04768827 129 0.3474690 0.2264488 0.11666296 0.5862955 +#> 1: bili 0.11626666 0.80 0.2429676 0.1331339 0.05061303 0.3945576 +#> 2: bili 0.11626666 1.4 0.2653392 0.1505044 0.06736587 0.4191150 +#> 3: bili 0.11626666 3.5 0.3816531 0.3008428 0.17477323 0.5710884 +#> 4: copper 0.04763783 43 0.2714431 0.1379538 0.05205695 0.4730195 +#> 5: copper 0.04763783 74 0.2921292 0.1641790 0.06105152 0.5005068 +#> 6: copper 0.04763783 129 0.3475959 0.2266914 0.11717755 0.5878593 #> pred_horizon level #> 1: 1826.25 <NA> #> 2: 1826.25 <NA> @@ -534,15 +534,15 @@

    Visualizing ICE curves#> id_variable id_row pred_horizon bili pred #> 1: 1 1 1826.25 1 0.9249698 #> 2: 1 2 1826.25 1 0.1039755 -#> 3: 1 3 1826.25 1 0.7438315 -#> 4: 1 4 1826.25 1 0.3883298 +#> 3: 1 3 1826.25 1 0.7436771 +#> 4: 1 4 1826.25 1 0.3805933 #> 5: 1 5 1826.25 1 0.1252893 #> --- -#> 6896: 25 272 1826.25 10 0.3402022 +#> 6896: 25 272 1826.25 10 0.3440760 #> 6897: 25 273 1826.25 10 0.4858345 -#> 6898: 25 274 1826.25 10 0.5194659 +#> 6898: 25 274 1826.25 10 0.5247851 #> 6899: 25 275 1826.25 10 0.2836209 -#> 6900: 25 276 1826.25 10 0.5541267 +#> 6900: 25 276 1826.25 10 0.5550106