change plotly template to plotly_dark

rename scripts/{precision_recall->cumulative_clf_metrics}.py
set transparent paper_bgcolor
set cum clf metrics config=dict(showTips=False, displayModeBar=False, scrollZoom=True)

Author: janosh

matbench_discovery/plots.py

@@ -55,11 +55,14 @@
     dft="DFT",
 )
 px.defaults.labels = quantity_labels | model_labels
-pastel_layout = dict(
-    colorway=px.colors.qualitative.Pastel, margin=dict(l=40, r=30, t=60, b=30)
+global_layout = dict(
+    # colorway=px.colors.qualitative.Pastel,
+    margin=dict(l=30, r=20, t=60, b=20),
+    paper_bgcolor="rgba(0,0,0,0)",
+    # plot_bgcolor="rgba(0,0,0,0)",
 )
-pio.templates["pastel"] = dict(layout=pastel_layout)
-pio.templates.default = "plotly_white+pastel"
+pio.templates["global"] = dict(layout=global_layout)
+pio.templates.default = "plotly_dark+global"
 
 # https://github.com/plotly/Kaleido/issues/122#issuecomment-994906924
 # when seeing MathJax "loading" message in exported PDFs, try:
@@ -607,7 +610,7 @@ def cumulative_precision_recall(
             )
         fig.for_each_annotation(lambda a: a.update(text=""))
         fig.update_layout(legend=dict(title=""))
-        fig.update_layout(showlegend=False)
+        # fig.update_layout(showlegend=False)
 
     return fig, df_cum
 

readme.md

@@ -13,9 +13,9 @@ Matbench Discovery
 
 </h4>
 
-Matbench Discovery is an [interactive leaderboard](https://matbench-discovery.janosh.dev/figures) and associated [PyPI package](https://pypi.org/project/matbench-discovery) for benchmarking ML energy models on a task designed to closely emulate a real-world computational materials discovery workflow in which these models would be used for a pre-triaging step to determine how to allocate limited compute budget on DFT structure relaxations.
+Matbench Discovery is an [interactive leaderboard](https://matbench-discovery.janosh.dev/figures) and associated [PyPI package](https://pypi.org/project/matbench-discovery) for benchmarking ML energy models on a task designed to closely emulate a real-world computational materials discovery workflow. In it, these models take on the role of a triaging step to prior to DFT to determine how to allocate limited compute budget for structure relaxations.
 
-We welcome contributions that add new models to the leaderboard through [GitHub PRs](https://github.com/janosh/matbench-discovery/pulls).
+We welcome contributions that add new models to the leaderboard through [GitHub PRs](https://github.com/janosh/matbench-discovery/pulls). See the [usage and contributing guide](site/src/routes/how-to-contribute/+page.md).
 
 Several new energy models specifically designed to handle unrelaxed structures were published in 2021/22
 

scripts/precision_recall.py scripts/cumulative_clf_metrics.py

@@ -36,7 +36,7 @@
     show_optimal=True,
 )
 
-title = f"{today} - Cumulative Precision and Recall for Stable Materials"
+title = f"{today} - Cumulative Precision, Recall and F1 Score for Stable Materials"
 # xlabel_cumulative = "Materials predicted stable sorted by hull distance"
 if backend == "matplotlib":
     fig.suptitle(title)
@@ -51,6 +51,11 @@
 img_path = f"{ROOT}/figures/{today}-precision-recall-curves"
 # if hasattr(fig, "write_image"):
 #     fig.write_image(f"{img_path}.pdf")
-#     fig.write_html(f"{img_path}.svelte", include_plotlyjs=False, full_html=False)
+#     fig.write_html(
+#         f"{img_path}.svelte",
+#         include_plotlyjs=False,
+#         full_html=False,
+#         config=dict(showTips=False, displayModeBar=False, scrollZoom=True),
+#     )
 # else:
 #     fig.savefig(f"{img_path}.pdf")

site/src/routes/+page.svelte

@@ -1,10 +1,23 @@
 <script lang="ts">
-  import Plot from '$root/figures/2022-12-25-precision-recall-curves.svelte'
+  import CumulativeMetricsPlot from '$root/figures/2022-12-25-precision-recall-curves.svelte'
   import Readme from '$root/readme.md'
+  import { onMount } from 'svelte'
+
+  onMount(() => {
+    // make links to site/src/routes/ files deployment-compatible
+    const site_links: NodeListOf<HTMLAnchorElement> = document.querySelectorAll(
+      `a[href^='site/src/routes']`
+    )
+    for (const link of site_links) {
+      link.href = link.href.replace(`/site/src/routes`, ``).split('/+page')[0]
+    }
+  })
 </script>
 
 <Readme />
 
 {#if typeof document !== `undefined`}
-  <Plot style="margin: 0 -5vw;" />
+  <CumulativeMetricsPlot
+    style="width: 90vw; transform: translateX(-45vw) translateX(min(25em, 45vw))"
+  />
 {/if}

site/src/routes/about-the-data/+page.svelte site/src/routes/about-the-test-set/+page.svelte

@@ -9,10 +9,8 @@
 
   onMount(() => {
     for (const img of document.querySelectorAll(`img`)) {
-      const src = img.getAttribute(`src`)
-      if (figs[`../data/wbm/${src}`]) {
-        img.src = figs[`../data/wbm/${src}`]
-      }
+      const src = `../data/wbm/${img.getAttribute(`src`)}`
+      if (figs[src]) img.src = figs[src]
     }
   })
 </script>

site/src/routes/how-to-use/+page.svx site/src/routes/how-to-contribute/+page.md

@@ -1,6 +1,6 @@
 ## Installation
 
-The recommended way to use this benchmark is through the [PyPI Python package](https://pypi.org/project/matbench-discovery):
+The recommended way to use this benchmark is through its Python package [available onPyPI](https://pypi.org/project/matbench-discovery):
 
 ```zsh
 pip install matbench-discovery
@@ -11,6 +11,7 @@ pip install matbench-discovery
 Here's an example script of how to download the training and test set files for training a new model, recording the results and submitting them via pull request to this benchmark:
 
 <!-- TODO remove notest meta key once repo is public and file can be downloaded without token -->
+
 ```py notest
 from matbench_discovery.data import load_train_test
 from matbench_discovery.data import df_wbm