Meeting chat log
00:11:35 Ben G: Anyone know how to re-size an image for a Markdown slides presentation
00:19:03 Federica Gazzelloni: ops sorry just seen it
00:20:15 Federica Gazzelloni: anyway you can resize it using fig.dim=“some percentage” in the chunk
00:30:51 Ricardo Serrano: Cohen's Kappa metric https://thedatascientist.com/performance-measures-cohens-kappa-statistic/
00:33:22 Federica Gazzelloni: thanks @ricardo
00:37:32 Ricardo Serrano: H2O Metrics Masterclass https://youtu.be/0bZeTUOPDTQ
00:37:47 Federica Gazzelloni: {r pressure, echo=FALSE, fig.dim="50%"}
plot(pressure)
00:40:31 Jim Gruman: roll_rs <- sliding_period(
train_df,
date,
period = "month",
every = 6L,
lookback = Inf,
skip = 2L
)
roll_rs |>
tidy() |>
ggplot(aes(x = Resample,
y = Row,
fill = Data)) +
geom_tile() +
scale_fill_viridis_d() +
theme_bw()
00:46:17 ethan tenison: In the book it recommends doing 10-fold cross validation plus 5 iterations for small datasets. Would bootstrapping still be better?
01:05:34 Jim Gruman: thank you Ben! I'm going to have to step away. See you all next week