model card renders

Author: isabelizimm

.gitignore

@@ -1,5 +1,6 @@
 /docs/site/*
-
+docs/source/reference
+vetiver/templates/*.html
 # Mac specific
 .DS_Store
 

vetiver/data/__init__.py

@@ -3,6 +3,7 @@
 sources = {
     "mtcars": _files("vetiver") / "data/mtcars.csv",
     "chicago": _files("vetiver") / "data/chicago.csv",
+    "sacramento": _files("vetiver") / "data/sacramento.csv",
 }
 
 
@@ -16,5 +17,6 @@ def __getattr__(k):
     f_path = sources.get("mtcars")
     if k == "chicago":
         f_path = sources.get("chicago")
-
+    elif k == "sacramento":
+        f_path = sources.get("sacramento")
     return pd.read_csv(f_path)

vetiver/data/sacramento.csv

@@ -0,0 +1,173 @@
+"price","type","sqft","beds","baths"
+68212,"Residential",1167,3,1
+89921,"Condo",1122,3,1
+98937,"Residential",1146,3,2
+122682,"Residential",1329,4,2
+123000,"Residential",1240,4,2
+124100,"Residential",1601,3,2
+126640,"Residential",1088,3,2
+131200,"Residential",1380,4,2
+141000,"Condo",1039,2,2
+146250,"Residential",1418,3,2
+149593,"Residential",1146,4,2
+152000,"Condo",1304,2,2
+164000,"Residential",1120,4,1
+173000,"Residential",1590,4,2
+181000,"Residential",1172,3,2
+189000,"Residential",1420,3,2
+195000,"Multi_Family",2162,4,4
+231477,"Residential",1478,3,2
+234697,"Residential",1287,3,2
+242638,"Residential",2163,3,2
+244500,"Residential",1473,3,3
+254200,"Residential",1921,4,2
+263500,"Residential",1713,4,2
+265000,"Condo",1672,3,2
+285000,"Residential",2056,4,2
+292024,"Residential",1126,2,1
+334150,"Residential",1690,2,1
+370500,"Residential",2734,4,3
+372000,"Residential",2110,3,2
+430000,"Residential",3615,5,3
+510000,"Residential",3508,5,4
+539000,"Residential",2462,4,2
+98000,"Residential",1220,2,1
+99000,"Residential",1643,3,1
+100000,"Condo",722,1,1
+111000,"Residential",1039,3,1
+123750,"Residential",1110,3,2
+126000,"Residential",1080,3,2
+135500,"Residential",1264,3,2
+145000,"Residential",1448,4,2
+146000,"Residential",1183,4,2
+150000,"Residential",1364,3,2
+161653,"Residential",1156,3,2
+175000,"Residential",1671,3,2
+180400,"Residential",1716,3,2
+205000,"Residential",1953,4,3
+230000,"Residential",1567,3,2
+236250,"Residential",1291,3,1
+256054,"Multi_Family",2085,5,4
+261000,"Residential",1567,3,2
+270000,"Residential",1531,3,2
+280000,"Residential",2494,4,3
+330000,"Residential",1258,3,2
+350000,"Residential",2295,4,3
+460000,"Residential",2129,3,2
+504000,"Residential",2791,3,3
+613401,"Residential",3838,4,3
+614000,"Residential",3059,4,3.5
+48000,"Condo",484,1,1
+100000,"Condo",625,1,1
+101000,"Residential",888,2,1
+114750,"Residential",779,2,1
+116100,"Residential",1174,3,1
+120000,"Multi_Family",1995,4,4
+123000,"Residential",1104,3,2
+130000,"Residential",1209,3,2
+133105,"Residential",1139,2,1
+136500,"Residential",1690,2,2
+145000,"Residential",1120,3,2
+147000,"Residential",1590,4,2
+149600,"Residential",1407,3,2
+150000,"Condo",1057,2,2
+150000,"Residential",1646,3,2
+165000,"Residential",1112,2,2
+168000,"Residential",1039,3,1
+188325,"Residential",1152,3,1
+215000,"Residential",1856,4,2
+220000,"Residential",950,2,1
+232000,"Residential",1375,3,2
+232500,"Condo",1250,2,2
+233641,"Residential",1058,3,1
+236000,"Multi_Family",2382,5,4
+240000,"Residential",1816,4,2
+252155,"Residential",1574,4,2
+254172,"Residential",1830,3,2
+261000,"Residential",1904,4,2
+270000,"Residential",1457,3,2
+285000,"Multi_Family",960,2,2
+296000,"Residential",2109,4,2
+304000,"Residential",2489,5,3
+305000,"Residential",1851,4,3
+325000,"Residential",2347,5,3
+351000,"Residential",2789,4,2
+355000,"Residential",3499,5,3.5
+360552,"Residential",1871,4,2
+362305,"Residential",1800,3,2
+395100,"Residential",3281,5,3
+408431,"Residential",2085,4,3
+445000,"Residential",2002,4,2
+500000,"Residential",1905,3,2
+677048,"Residential",2025,3,2
+95625,"Residential",1089,3,2
+115000,"Residential",911,2,1
+115000,"Condo",846,1,1
+115620,"Residential",1410,3,2
+131750,"Residential",1118,2,2
+137760,"Residential",1260,3,2
+176095,"Residential",1200,3,2
+176250,"Residential",1170,3,2
+178000,"Residential",1199,3,2
+190000,"Residential",796,2,1
+193000,"Residential",1452,3,2
+193500,"Residential",1513,3,2
+204918,"Residential",2306,4,2
+210944,"Condo",984,1,1
+212500,"Condo",1462,3,2.5
+220000,"Residential",1566,4,2
+220000,"Residential",1419,4,2
+220000,"Residential",1261,3,2
+232500,"Residential",2258,5,3
+233000,"Residential",1074,3,2
+242000,"Residential",1406,2,2
+243500,"Residential",1876,4,2
+247000,"Residential",1676,3,2
+249000,"Residential",1899,4,3
+260000,"Residential",1196,3,2
+270000,"Residential",2750,2,2
+271000,"Residential",1910,3,2.5
+280000,"Residential",1443,3,2
+285000,"Residential",1857,3,2
+289000,"Residential",2096,4,3
+290000,"Residential",1720,3,2
+294000,"Residential",1282,3,2
+299000,"Residential",1144,2,1
+300000,"Residential",1457,3,2
+300567,"Residential",1577,3,2
+310000,"Residential",1800,4,3
+315000,"Residential",3134,5,3
+326951,"Residential",1744,3,2
+331200,"Residential",2484,4,2
+335000,"Residential",1616,3,2
+350000,"Residential",2960,3,2
+375000,"Residential",2616,5,3
+380000,"Residential",1727,3,1
+400000,"Residential",3566,5,4
+455000,"Residential",2379,4,3
+600000,"Residential",3901,5,5
+61000,"Residential",876,2,1
+70000,"Residential",1011,2,1
+80000,"Residential",1158,4,2
+85500,"Residential",1092,3,2
+92000,"Condo",956,2,1
+112500,"Residential",1249,3,2
+114800,"Residential",1161,3,2
+128687,"Residential",962,3,1
+130000,"Residential",1127,3,2
+132000,"Residential",970,2,1
+145846,"Residential",1543,3,2
+150000,"Residential",1392,4,2
+170000,"Residential",1232,3,2
+171750,"Residential",1177,2,2
+172000,"Residential",1582,3,2
+176850,"Residential",1477,4,2
+189000,"Residential",1039,3,1
+191675,"Residential",1638,3,2
+209347,"Residential",1262,3,2
+216000,"Residential",1416,3,2
+216021,"Residential",888,3,1
+220000,"Residential",1770,4,2
+224252,"Residential",960,3,1
+228000,"Residential",1450,3,2
+235301,"Residential",1685,4,2

vetiver/model_card.py

@@ -15,7 +15,6 @@ def model_card(path="."):
     """
     src_path = _files("vetiver") / "templates/model_card.qmd"
 
-    if _log is not None:
-        _log.info("Writing model card template...")
+    logging.info("Writing model card template:")
 
     return shutil.copy(src=src_path, dst=path)

vetiver/templates/model_card.qmd

@@ -1,34 +1,64 @@
 ---
 title: "Model Card: your model title"
-date: '`python date.today()`'
 output:
   html_document
-params:
-    board: !python pins.board_rsconnect()
-    name: isabel.zimmerman/sacramento_rf
-    version: NULL
 ---
+```{python}
+#| echo: false
+from IPython.display import display, Markdown
+import datetime
+display(Markdown(f"""
+###### {datetime.date.today()}
+"""))
+```
+
+```{python}
+#| tags: [parameters]
+#| include: false
+import pins
+
+raw = "https://colorado.rstudio.com/rsc/"
+paths = {"v": "sacramento/"}
+board = pins.board_urls(raw, paths, allow_pickle_read=True)
+```
 
-```{python setup, include=FALSE}
-from datetime import date
+```{python}
+#| include: false
+import datetime
+import pandas as pd
+import plotly.express as px
 from sklearn import metrics
 import vetiver
-import pins
 
-# library(yardstick)
-# knitr::opts_chunk$set(echo = FALSE)
-v = vetiver.VetiverModel.from_pin(pins.board_rsconnect(server_url = "https://colorado.rstudio.com/rsc/"), "isabel.zimmerman/cars_mpg")
+v = vetiver.VetiverModel.from_pin(board, "v")
+v_meta = board.pin_meta("v")
 ```
 
+
 A [model card](https://doi.org/10.1145/3287560.3287596) provides brief, transparent, responsible reporting for a trained machine learning model.
 
 ## Model details
 
 - Developed by PERSON AND/OR TEAM
-- `python print(f"{v.description} using {v.ptype} features")`
+```{python}
+#| echo: false
+display(Markdown(f"""
+- A {v.description} using {len(v.ptype.construct().dict())} features.
+"""))
+```
 - More details about how model was developed and what it is predicting
 - More details on feature engineering and/or data preprocessing for model
-- Version `python v.metadata.version` of this model was published at `r v_meta.created`
+```{python}
+#| echo: false
+display(Markdown(f"""
+- Version {v.metadata.get("version")} was created at {
+    datetime.datetime.strptime(
+    v_meta.created, '%Y%m%dT%H%M%SZ')
+    .strftime('%Y-%m-%d %H:%M:%S')
+    }
+"""))
+```
+
 - Citation and/or license details for the model
 - If you have questions about this model, please contact PERSON@ORG.ORG
 
@@ -53,70 +83,64 @@ A [model card](https://doi.org/10.1145/3287560.3287596) provides brief, transpar
 
 - The training dataset for this model was ...
 - The training dataset for this model has the "prototype" or signature:
-
 ```{python}
-v.ptype
+#| echo: false
+v.ptype.construct().schema().get("properties")
 ```
 
 - The evaluation dataset used in this model card is ...
 - We chose this evaluation data because ...
 
+
+## Quantitative analyses
+:::{.panel-tabset}
 ```{python}
-## EVALUATION DATA:
-from vetiver.data import mtcars
+#| include: false
 ## consider using a package like skimr or DataExplorer for automated
-## presentation of evaluation data characteristics
-```
 
-
-## Quantitative analyses {.tabset}
-
-```{python}
-## compute predictions for your evaluation data
-## `handler_startup` is designed to get the R process ready to make predictions
-#suppressPackageStartupMessages(handler_startup(v))
-new_data["preds"] = v.model.predict(mtcars.drop(columns="mpg"))
 ```
-
-
-### Overall model performance
+## Overall model performance
 
 ```{python}
+#| echo: false
+## compute predictions for your evaluation data
+from vetiver.data import sacramento
 
-metric_set = [metrics.mean_absolute_error, metrics.mean_squared_error]
+sacramento["preds"] = v.model.predict(sacramento.drop(columns=["price"]))
+metric_set = [metrics.mean_absolute_error, metrics.mean_squared_error, metrics.r2_score]
 
+for metric in metric_set:
+    print(metric.__qualname__, ": ", metric(y_pred=sacramento["preds"], y_true=sacramento["price"]))
 ```
 
-### Disaggregated model performance
+## Disaggregated model performance
 
 ```{python}
+#| echo: false
 
-preds %>%
-    group_by(type) %>%
-    metrics(price, .pred)
+dmp = pd.DataFrame()
+for metric in metric_set:
+    dmp[metric.__qualname__] = sacramento.groupby("type")\
+        .apply(lambda x: metric(y_pred=x["preds"], y_true=x["price"]))
+dmp
 ```
 
-### Visualize model performance
+## Visualize model performance
 
-```{python, fig.height=3}
-preds %>%
-    ggplot(aes(price, .pred, color = type)) +
-    geom_abline(slope = 1, lty = 2, color = "gray60", size = 1.2) +
-    geom_point(alpha = 0.5, show.legend = FALSE) +
-    facet_wrap(vars(type))
+```{python}
+performance = px.scatter(sacramento, x="price", y = "preds", facet_row="type", trendline="ols")
+performance.update_yaxes(matches=None)
+performance.show()
 ```
 
-### Make a custom plot
+## Make a custom plot
 
-```{r}
-preds %>%
-    mutate(.resid = price - .pred) %>%
-    ggplot(aes(longitude, latitude, color = .resid)) +
-    geom_point(alpha = 0.8) +
-    scale_color_gradient2() +
-    coord_fixed()
+```{python}
+custom = px.violin(sacramento, x="beds", y = "price")
+custom.show()
 ```
 
+:::
 
 ## Ethical considerations
 

vetiver/vetiver_model.py

@@ -87,7 +87,7 @@ def from_pin(cls, board, name: str, version: str = None):
             description=meta.description,
             metadata=_model_meta(
                 user=meta.user,
-                version=meta.version.version.abc,
+                version=meta.version.version,
                 url=meta.user.get("url"),  # None all the time, besides Connect
                 required_pkgs=meta.user.get("required_pkgs"),
             ),