]> code.communitydata.science - ml_measurement_error_public.git/blob - simulations/.Rhistory
Unstash work on misclassification models
[ml_measurement_error_public.git] / simulations / .Rhistory
1 ls()
2 weight
3 weight
4 lablr
5 labelr
6 nrow(labelr)
7 names(labelr)
8 names(labelr$data)
9 labelr$data
10 labelr
11 names(labelr)
12 labelr$labelr
13 labelr$toxic
14 setwd("..")
15 q()
16 n
17 summary(w2)
18 summary(w2)
19 q()
20 n
21 summary(fit1)
22 0.5*(dat$x1 + sapply(dat$sdx, function(sd) rnorm(1,0,sd)))
23 summary(fit1)
24 summary(fit2)
25 summary(fit2)
26 conditional_effects(fit2,resp='y')
27 plot(conditional_effects(fit2,resp='y'))
28 stancode(fit2)
29 stancode(fit1)
30 sessionInfo()
31 q()
32 y
33 p.y
34 range(p.y)
35 rbinom
36 df2
37 df2
38 df2
39 brms.corrected.logit
40 q()
41 n
42 summary(brms.corrected.logit)
43 summary(brms.corrected.logit)
44 p.y
45 q()
46 n
47 mw
48 summary(mw)
49 )
50 summary(true.model)
51 true.model
52 true.model$R
53 true.model$null.deviance
54 true.model$deviance
55 getwd()
56 setwd("../../)
57 setwd("../../)
58
59 setwd("../../partitioning_reddit")
60 ls
61 getwd()
62 list.files()
63 install.packages("filelock")
64 q()
65 n
66 df
67 df
68 outcome_formula <- y ~ x + z
69 outcome_family=gaussian()
70 proxy_formula <- w_pred ~ x
71 truth_formula <- x ~ z
72 params <- start
73 ll.y.obs.x0
74 ll.y.obs.x1
75 rater_formula <- x.obs ~ x
76 rater_formula
77 rater.modle.matrix.obs.x0
78 rater.model.matrix.obs.x0
79 names(rater.model.matrix.obs.x0)
80 head(rater.model.matrix.obs.x0)
81 df.obs
82 ll.x.obs.0
83 rater.params
84 rater.params %*% t(rater.model.matrix.x.obs.0[df.obs$xobs.0==1])
85 df.obs$xobs.0==1
86 df.obs$x.obs.0==1
87 ll.x.obs.0[df.obs$x.obs.0==1]
88 rater.model.matrix.x.obs.0[df.obs$x.obs.0==1,]
89 df.obs$x.obs.0==1
90 n.rater.model.covars <- dim(rater.model.matrix.x.obs.0)[2]
91         rater.params <- params[param.idx:n.rater.model.covars]
92 rater.params
93         ll.x.obs.0[df.obs$x.obs.0==1] <- plogis(rater.params %*% t(rater.model.matrix.x.obs.0[df.obs$x.obs.0==1,]), log=TRUE)
94 t(rater.model.matrix.x.obs.0[df.obs$x.obs.0==1,]
95 )
96 dimt(rater.model.matrix.x.obs.0[df.obs$x.obs.0==1,])
97 dim(t(rater.model.matrix.x.obs.0[df.obs$x.obs.0==1,]))
98 dim(ll.x.obs.0[df.obs$x.obs.0==1])
99 rater.params
100 rater.params
101 rater.params
102 rater_formula
103 rater.params
104 )
105 1+1
106 q()
107 n
108 outcome_formula <- y ~ x + z
109 proxy_formula <- w_pred ~ x + z + y
110 truth_formula <- x ~ z
111 proxy_formula
112 eyboardio Model 01 - Kaleidoscope locally built
113 df <- df.triple.proxy.mle
114 outcome_family='gaussian'
115 outcome_family=gaussian()
116 proxy_formulas=list(proxy_formula,x.obs.0~x, x.obs.1~x)
117 proxy_formulas
118 proxy_familites <- rep(binomial(link='logit'),3)
119 proxy_families = rep(binomial(link='logit'),3)
120 proxy_families
121 proxy_families = list(binomial(link='logit'),binomial(link='logit'),binomial(link='logit'))
122 proxy_families
123 proxy_families[[1]]
124 proxy.params
125 i
126 proxy_params
127 proxy.params
128 params
129 params <- start
130 df.triple.proxy.mle
131 df
132 coder.formulas <- c(x.obs.0 ~ x, x.obs.1 ~x)
133 outcome.formula
134 outcome_formula
135 depvar(outcome_formula
136 )
137 outcome_formula$terms
138 terms(outcome_formula)
139 q()
140 n
141 df.triple.proxy.mle
142 triple.proxy.mle
143 df
144 df <- df.triple.proxy
145 outcome_family <- binomial(link='logit')
146 outcome_formula <- y ~x+z
147 proxy_formula <- w_pred ~ y
148 coder_formulas=list(y.obs.1~y,y.obs.2~y); proxy_formula=w_pred~y; proxy_family=binomial(link='logit'))
149 coder_formulas=list(y.obs.1~y,y.obs.2~y); proxy_formula=w_pred~y; proxy_family=binomial(link='logit')
150 coder_formulas=list(y.obs.0~y,y.obs.1~y)
151 traceback()
152 df
153 df
154 outcome.model.matrix
155 q()
156 n

Community Data Science Collective || Want to submit a patch?