estimates () not working at all and compare.fits() not for lmer?

[LauritaInFrance]

Would love the help here. Is it common for others? problem in the code or the version of flexplot?

######https://www.youtube.com/watch?v=jXlzQm0vF5g
model = lmer(darkness ~ anger + emotional_bonds + (anger + emotional_bonds | dojo_id),data=jedi)
Message d'avis :
Dans checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
Model failed to converge with max|grad| = 0.0144489 (tol = 0.002, component 1)
summary(model)
Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: darkness ~ anger + emotional_bonds + (anger + emotional_bonds |
dojo_id)
Data: jedi

REML criterion at convergence: 8789.8

Scaled residuals:
Min 1Q Median 3Q Max
-4.4329 -0.6265 0.0004 0.6542 3.8004

Random effects:
Groups Name Variance Std.Dev. Corr
dojo_id (Intercept) 6.90721 2.6282
anger 0.01302 0.1141 -0.62
emotional_bonds 0.02213 0.1488 -0.78 0.11
Residual 1.56222 1.2499
Number of obs: 2530, groups: dojo_id, 95

Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.84432 0.31358 90.63236 -2.693 0.00845 **
anger 0.37820 0.01492 88.79089 25.344 < 2e-16 ***
emotional_bonds 0.10897 0.01791 89.49654 6.085 2.83e-08 ***

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1

Correlation of Fixed Effects:
(Intr) anger
anger -0.638
emotnl_bnds -0.756 0.071
optimizer (nloptwrap) convergence code: 0 (OK)
Model failed to converge with max|grad| = 0.0144489 (tol = 0.002, component 1)

library(flexplot)
> estimates(model)
Erreur dans str2lang(x) : :1:15: ')' inattendu(e)
1: darkness~1+(1|)
^
###NOT WORKING ############### in none of the scripts i tried

visualize(model, plot="model")

####me playing around according to codes from https://github.com/dustinfife/flexplot
visualize(model)
flexplot(darkness1, data=jedi)
flexplot(darknessemotional_bonds| anger , data=jedi, method="lm", se=F, bins=3)

####back to your visualizations
visualize(model, plot="model",

•       formula=darkness~emotional_bonds+dojo_id| anger, sample=30)
  

It looks like you're trying to plot more than 6 colors/lines/symbols.
I gotta give it to you...you're ambitious. Alas, I can't do that, so I'm removing the colors/lines/symbols.
I hope we can still be friends.

visualize(model, plot="model",

•       formula=darkness~emotional_bonds+dojo_id, sample=30)
  

Note: You didn't choose to plot anger so I am inputting the median

It looks like you're trying to plot more than 6 colors/lines/symbols.
I gotta give it to you...you're ambitious. Alas, I can't do that, so I'm removing the colors/lines/symbols.
I hope we can still be friends.

visualize(model, plot="model",

•       formula=darkness~anger+dojo_id, sample=30)
  

Note: You didn't choose to plot emotional_bonds so I am inputting the median

It looks like you're trying to plot more than 6 colors/lines/symbols.
I gotta give it to you...you're ambitious. Alas, I can't do that, so I'm removing the colors/lines/symbols.
I hope we can still be friends.

###anger seems parallel lines to fixed-so maybe no random effect neeeded
reduced = lmer(darkness ~ emotional_bonds + anger + (emotional_bonds | dojo_id ), data=jedi)
Message d'avis :
Dans checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
Model failed to converge with max|grad| = 0.0232594 (tol = 0.002, component 1)

**> compare.fits(darkness ~ anger | dojo_id,

•          data=jedi,
  

•          model1 = model,
  

•          model2 = reduced,
  

•          re=T,
  

•          clusters=10)
  

Erreur dans model$model : $ operator not defined for this S4 class

###########NOT WORKING ############# compare fits does not work here for me but actually model comparison works,
#however not the same output as yours in the video, no p value or squared change
model.comparison(model,reduced)
$statistics
aic bic bayes.factor
model 8809.836 8868.196 118334497021
reduced 8878.337 8919.189 0

$predicted_differences
0% 25% 50% 75% 100%
0.000 0.028 0.091 0.230 2.062

####compare fits for a lm model instead of lmer is working though adopted from your github example
reduced2 = lm(darkness ~ emotional_bonds + anger, data=jedi)
model2 = lm(darkness ~ anger * emotional_bonds,data=jedi)
compare.fits(darkness~anger|emotional_bonds, data=jedi,model2,reduced2)**