Skip to content

Commit 8d63d49

Browse files
dajmcdondajmcdon
committed
fix README options for proper output print width
1 parent b8ee54e commit 8d63d49

File tree

2 files changed

+89
-134
lines changed

2 files changed

+89
-134
lines changed

‎README.Rmd

+2-1
Original file line numberDiff line numberDiff line change
@@ -5,8 +5,9 @@ output: github_document
55
<!-- README.md is generated from README.Rmd. Please edit that file -->
66

77
```{r, include = FALSE}
8+
options(width = 76)
89
knitr::opts_chunk$set(
9-
collapse = FALSE,
10+
collapse = TRUE,
1011
comment = "#>",
1112
fig.path = "man/figures/README-",
1213
out.width = "100%"

‎README.md

+87-133
Original file line numberDiff line numberDiff line change
@@ -72,28 +72,27 @@ processed using
7272
``` r
7373
library(epipredict)
7474
case_death_rate_subset
75+
#> An `epi_df` object, 20,496 x 4 with metadata:
76+
#> * geo_type = state
77+
#> * time_type = day
78+
#> * as_of = 2022-05-31 12:08:25.791826
79+
#>
80+
#> # A tibble: 20,496 × 4
81+
#> geo_value time_value case_rate death_rate
82+
#> * <chr> <date> <dbl> <dbl>
83+
#> 1 ak 2020-12-31 35.9 0.158
84+
#> 2 al 2020-12-31 65.1 0.438
85+
#> 3 ar 2020-12-31 66.0 1.27
86+
#> 4 as 2020-12-31 0 0
87+
#> 5 az 2020-12-31 76.8 1.10
88+
#> 6 ca 2020-12-31 96.0 0.751
89+
#> 7 co 2020-12-31 35.8 0.649
90+
#> 8 ct 2020-12-31 52.1 0.819
91+
#> 9 dc 2020-12-31 31.0 0.601
92+
#> 10 de 2020-12-31 65.2 0.807
93+
#> # ℹ 20,486 more rows
7594
```
7695

77-
#> An `epi_df` object, 20,496 x 4 with metadata:
78-
#> * geo_type = state
79-
#> * time_type = day
80-
#> * as_of = 2022-05-31 12:08:25.791826
81-
#>
82-
#> # A tibble: 20,496 × 4
83-
#> geo_value time_value case_rate death_rate
84-
#> * <chr> <date> <dbl> <dbl>
85-
#> 1 ak 2020-12-31 35.9 0.158
86-
#> 2 al 2020-12-31 65.1 0.438
87-
#> 3 ar 2020-12-31 66.0 1.27
88-
#> 4 as 2020-12-31 0 0
89-
#> 5 az 2020-12-31 76.8 1.10
90-
#> 6 ca 2020-12-31 96.0 0.751
91-
#> 7 co 2020-12-31 35.8 0.649
92-
#> 8 ct 2020-12-31 52.1 0.819
93-
#> 9 dc 2020-12-31 31.0 0.601
94-
#> 10 de 2020-12-31 65.2 0.807
95-
#> # ℹ 20,486 more rows
96-
9796
To create and train a simple auto-regressive forecaster to predict the
9897
death rate two weeks into the future using past (lagged) deaths and
9998
cases, we could use the following function.
@@ -109,40 +108,24 @@ two_week_ahead <- arx_forecaster(
109108
)
110109
)
111110
two_week_ahead
111+
#> ══ A basic forecaster of type ARX Forecaster ═══════════════════════════════
112+
#>
113+
#> This forecaster was fit on 2023-12-23 09:12:46.
114+
#>
115+
#> Training data was an <epi_df> with:
116+
#> • Geography: state,
117+
#> • Time type: day,
118+
#> • Using data up-to-date as of: 2022-05-31 12:08:25.
119+
#>
120+
#> ── Predictions ─────────────────────────────────────────────────────────────
121+
#>
122+
#> A total of 56 predictions are available for
123+
#> • 56 unique geographic regions,
124+
#> • At forecast date: 2021-12-31,
125+
#> • For target date: 2022-01-14.
126+
#>
112127
```
113128

114-
#> ══ A basic forecaster of type ARX Forecaster ═══════════════════════════════════
115-
116-
#>
117-
118-
#> This forecaster was fit on 2023-12-23 08:50:59.
119-
120-
#>
121-
122-
#> Training data was an <epi_df> with:
123-
124-
#> • Geography: state,
125-
126-
#> • Time type: day,
127-
128-
#> • Using data up-to-date as of: 2022-05-31 12:08:25.
129-
130-
#>
131-
132-
#> ── Predictions ─────────────────────────────────────────────────────────────────
133-
134-
#>
135-
136-
#> A total of 56 predictions are available for
137-
138-
#> • 56 unique geographic regions,
139-
140-
#> • At forecast date: 2021-12-31,
141-
142-
#> • For target date: 2022-01-14.
143-
144-
#>
145-
146129
In this case, we have used a number of different lags for the case rate,
147130
while only using 3 weekly lags for the death rate (as predictors). The
148131
result is both a fitted model object which could be used any time in the
@@ -152,98 +135,69 @@ last available time value in the data.
152135

153136
``` r
154137
two_week_ahead$epi_workflow
138+
#>
139+
#> ══ Epi Workflow [trained] ══════════════════════════════════════════════════
140+
#> Preprocessor: Recipe
141+
#> Model: linear_reg()
142+
#> Postprocessor: Frosting
143+
#>
144+
#> ── Preprocessor ────────────────────────────────────────────────────────────
145+
#>
146+
#> 6 Recipe steps.
147+
#> 1. step_epi_lag()
148+
#> 2. step_epi_lag()
149+
#> 3. step_epi_ahead()
150+
#> 4. step_naomit()
151+
#> 5. step_naomit()
152+
#> 6. step_training_window()
153+
#>
154+
#> ── Model ───────────────────────────────────────────────────────────────────
155+
#>
156+
#> Call:
157+
#> stats::lm(formula = ..y ~ ., data = data)
158+
#>
159+
#> Coefficients:
160+
#> (Intercept) lag_0_case_rate lag_1_case_rate lag_2_case_rate
161+
#> -0.0073358 0.0030365 0.0012467 0.0009536
162+
#> lag_3_case_rate lag_7_case_rate lag_14_case_rate lag_0_death_rate
163+
#> 0.0011425 0.0012481 0.0003041 0.1351769
164+
#> lag_7_death_rate lag_14_death_rate
165+
#> 0.1471127 0.1062473
166+
#>
167+
#> ── Postprocessor ───────────────────────────────────────────────────────────
168+
#>
169+
#> 5 Frosting layers.
170+
#> 1. layer_predict()
171+
#> 2. layer_residual_quantiles()
172+
#> 3. layer_add_forecast_date()
173+
#> 4. layer_add_target_date()
174+
#> 5. layer_threshold()
175+
#>
155176
```
156177

157-
#>
158-
159-
#> ══ Epi Workflow [trained] ══════════════════════════════════════════════════════
160-
161-
#> Preprocessor: Recipe
162-
163-
#> Model: linear_reg()
164-
165-
#> Postprocessor: Frosting
166-
167-
#>
168-
169-
#> ── Preprocessor ────────────────────────────────────────────────────────────────
170-
171-
#>
172-
173-
#> 6 Recipe steps.
174-
175-
#> 1. step_epi_lag()
176-
177-
#> 2. step_epi_lag()
178-
179-
#> 3. step_epi_ahead()
180-
181-
#> 4. step_naomit()
182-
183-
#> 5. step_naomit()
184-
185-
#> 6. step_training_window()
186-
187-
#>
188-
189-
#> ── Model ───────────────────────────────────────────────────────────────────────
190-
191-
#>
192-
#> Call:
193-
#> stats::lm(formula = ..y ~ ., data = data)
194-
#>
195-
#> Coefficients:
196-
#> (Intercept) lag_0_case_rate lag_1_case_rate lag_2_case_rate
197-
#> -0.0073358 0.0030365 0.0012467 0.0009536
198-
#> lag_3_case_rate lag_7_case_rate lag_14_case_rate lag_0_death_rate
199-
#> 0.0011425 0.0012481 0.0003041 0.1351769
200-
#> lag_7_death_rate lag_14_death_rate
201-
#> 0.1471127 0.1062473
202-
203-
#>
204-
205-
#> ── Postprocessor ───────────────────────────────────────────────────────────────
206-
207-
#>
208-
209-
#> 5 Frosting layers.
210-
211-
#> 1. layer_predict()
212-
213-
#> 2. layer_residual_quantiles()
214-
215-
#> 3. layer_add_forecast_date()
216-
217-
#> 4. layer_add_target_date()
218-
219-
#> 5. layer_threshold()
220-
221-
#>
222-
223178
The fitted model here involved preprocessing the data to appropriately
224179
generate lagged predictors, estimating a linear model with `stats::lm()`
225180
and then postprocessing the results to be meaningful for epidemiological
226181
tasks. We can also examine the predictions.
227182

228183
``` r
229184
two_week_ahead$predictions
185+
#> # A tibble: 56 × 5
186+
#> geo_value .pred .pred_distn forecast_date target_date
187+
#> <chr> <dbl> <dist> <date> <date>
188+
#> 1 ak 0.449 quantiles(0.45)[2] 2021-12-31 2022-01-14
189+
#> 2 al 0.574 quantiles(0.57)[2] 2021-12-31 2022-01-14
190+
#> 3 ar 0.673 quantiles(0.67)[2] 2021-12-31 2022-01-14
191+
#> 4 as 0 quantiles(0.12)[2] 2021-12-31 2022-01-14
192+
#> 5 az 0.679 quantiles(0.68)[2] 2021-12-31 2022-01-14
193+
#> 6 ca 0.575 quantiles(0.57)[2] 2021-12-31 2022-01-14
194+
#> 7 co 0.862 quantiles(0.86)[2] 2021-12-31 2022-01-14
195+
#> 8 ct 1.07 quantiles(1.07)[2] 2021-12-31 2022-01-14
196+
#> 9 dc 2.12 quantiles(2.12)[2] 2021-12-31 2022-01-14
197+
#> 10 de 1.09 quantiles(1.09)[2] 2021-12-31 2022-01-14
198+
#> # ℹ 46 more rows
230199
```
231200

232-
#> # A tibble: 56 × 5
233-
#> geo_value .pred .pred_distn forecast_date target_date
234-
#> <chr> <dbl> <dist> <date> <date>
235-
#> 1 ak 0.449 quantiles(0.45)[2] 2021-12-31 2022-01-14
236-
#> 2 al 0.574 quantiles(0.57)[2] 2021-12-31 2022-01-14
237-
#> 3 ar 0.673 quantiles(0.67)[2] 2021-12-31 2022-01-14
238-
#> 4 as 0 quantiles(0.12)[2] 2021-12-31 2022-01-14
239-
#> 5 az 0.679 quantiles(0.68)[2] 2021-12-31 2022-01-14
240-
#> 6 ca 0.575 quantiles(0.57)[2] 2021-12-31 2022-01-14
241-
#> 7 co 0.862 quantiles(0.86)[2] 2021-12-31 2022-01-14
242-
#> 8 ct 1.07 quantiles(1.07)[2] 2021-12-31 2022-01-14
243-
#> 9 dc 2.12 quantiles(2.12)[2] 2021-12-31 2022-01-14
244-
#> 10 de 1.09 quantiles(1.09)[2] 2021-12-31 2022-01-14
245-
#> # ℹ 46 more rows
246-
247201
The results above show a distributional forecast produced using data
248202
through the end of 2021 for the 14th of January 2022. A prediction for
249203
the death rate per 100K inhabitants is available for every state

0 commit comments

Comments
 (0)