Added templates for vignettes to test GitHub Actions and fix previous…

… issue.

_pkgdown.yml

@@ -17,3 +17,9 @@ navbar:
     - text: "Articles"
       menu:
         - text: "Anomaly Detection"
+        - href: articles/anomaly-detection.html
+        - text: "Cross validation"
+        - href: articles/cross-validation.html
+        - text: "Historical Forecast"
+        - href: articles/historical-forecast.html
+

vignettes/anomaly-detection.Rmd

@@ -2,7 +2,7 @@
 title: "Anomaly Detection"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{nixtlar}
+  %\VignetteIndexEntry{Anomaly Detection}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---
@@ -34,15 +34,15 @@ head(df)
 ```
 
 ```{r}
-timegpt_anomalies_df <- timegpt_anomaly_detection(df, id_col = "unique_id", level = c(95))
-head(timegpt_anomalies_df)
+timegpt_anomalies <- timegpt_anomaly_detection(df, id_col = "unique_id") 
+head(timegpt_anomalies)
 ```
 
-## 3. Plot TimeGPT forecast 
+## 2. Plot TimeGPT forecast 
 `nixtlar` includes a function to plot the historical data and any output from `timegpt_forecast`, `timegpt_historic`, `timegpt_anomaly_detection` and `timegpt_cross_validation`. If you have long series, you can use `max_insample_length` to only plot the last N historical values (the forecast will always be plotted in full). 
 
 ```{r}
-timegpt_plot(df, timegpt_anomalies_df, id_col = "unique_id", plot_anomalies = TRUE)
+timegpt_plot(df, timegpt_anomalies, id_col = "unique_id", plot_anomalies = TRUE)
 ```
 
 ```{r, include=FALSE}

vignettes/get-started.Rmd

@@ -2,7 +2,7 @@
 title: "Get Started"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{nixtlar}
+  %\VignetteIndexEntry{Get Started}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---
@@ -78,15 +78,15 @@ head(df)
 To generate a forecast for this dataset, use `timegpt_forecast`. Default names for the time and the target columns are `ds` and `y`. If your time and target columns have different names, specify them with `time_col` and `target_col`. Since it has multiple ids (one for every electricity market), you'll need to specify the name of the column that contains the ids, which in this case is `unique_id`. To do this, simply use `id_col="unique_id"`. You can also choose confidence levels (0-100) for prediction intervals with `level`. 
 
 ```{r}
-timegpt_fcst_df <- timegpt_forecast(df, h = 8, id_col = "unique_id")
-head(timegpt_fcst_df)
+timegpt_fcst <- timegpt_forecast(df, h = 8, id_col = "unique_id", level = c(80,95))
+head(timegpt_fcst)
 ```
 
 ## 3. Plot TimeGPT forecast 
 `nixtlar` includes a function to plot the historical data and any output from `timegpt_forecast`, `timegpt_historic`, `timegpt_anomaly_detection` and `timegpt_cross_validation`. If you have long series, you can use `max_insample_length` to only plot the last N historical values (the forecast will always be plotted in full). 
 
 ```{r}
-timegpt_plot(df, timegpt_fcst_df, id_col = "unique_id", max_insample_length = 200)
+timegpt_plot(df, timegpt_fcst, id_col = "unique_id", max_insample_length = 200)
 ```
 
 ```{r, include=FALSE}

vignettes/historical-forecast.Rmd

@@ -0,0 +1,53 @@
+---
+title: "Historical Forecast"
+output: rmarkdown::html_vignette
+vignette: >
+  %\VignetteIndexEntry{Historical Forecast}
+  %\VignetteEngine{knitr::rmarkdown}
+  %\VignetteEncoding{UTF-8}
+---
+```{r setup, include=FALSE}
+library(httptest2)
+start_vignette("mocks")
+
+options("TIMEGPT_TOKEN"="dummy_token")
+options(digits=7)
+
+knitr::opts_chunk$set(
+  collapse = TRUE,
+  comment = "#>", 
+  fig.width = 7, 
+  fig.height = 4
+)
+```
+
+```{r}
+library(nixtlar)
+```
+
+## 1. TimeGPT Historical Forecast
+
+```{r}
+df <- nixtlar::electricity
+head(df)
+```
+
+```{r}
+timegpt_fitted_values <- timegpt_historic(df, id_col = "unique_id", level = c(80,95))
+head(timegpt_fitted_values)
+```
+
+## 2. Plot TimeGPT forecast 
+`nixtlar` includes a function to plot the historical data and any output from `timegpt_forecast`, `timegpt_historic`, `timegpt_anomaly_detection` and `timegpt_cross_validation`. If you have long series, you can use `max_insample_length` to only plot the last N historical values (the forecast will always be plotted in full). 
+
+```{r}
+timegpt_plot(df, timegpt_fitted_values, id_col = "unique_id", max_insample_length = 200)
+```
+
+```{r, include=FALSE}
+end_vignette()
+```
+
+
+
+