add 2023-01-26-model-run-times-pie.svelte to /models page (from scripts/compile_metrics.py)

remove SEM shaded areas from rolling MAE plot
update each-scatter-models.webp
remove mdsvex Layout for /paper and Img.svelte component

Author: janosh

.github/workflows/gh-pages.yml

@@ -24,6 +24,8 @@ jobs:
         uses: actions/setup-python@v4
         with:
           python-version: "3.11"
+          cache: pip
+          cache-dependency-path: setup.py
 
       - name: Set up node
         uses: actions/setup-node@v3
@@ -34,7 +36,7 @@ jobs:
         run: |
           pip install lazydocs  # used in package.json script
           # lazydocs needs package deps to be installed
-          pip install .
+          pip install -e .
           cd site
           npm install
 

matbench_discovery/__init__.py

@@ -22,7 +22,8 @@
 timestamp = f"{datetime.now():%Y-%m-%d@%H-%M-%S}"
 today = timestamp.split("@")[0]
 
-# load URLs from package.json
+# load docs, repo, package URLs from package.json
+print(f"{ROOT=}")
 
 with open(f"{ROOT}/site/package.json") as file:
     pkg = json.load(file)

matbench_discovery/data.py

@@ -139,7 +139,7 @@ def load_train_test(
 
 
 PRED_FILENAMES = {
-    "CGCNN": "cgcnn/2022-11-23-test-cgcnn-wbm-IS2RE/cgcnn-ensemble-preds.csv",
+    "CGCNN": "cgcnn/2023-01-26-test-cgcnn-wbm-IS2RE/cgcnn-ensemble-preds.csv",
     "Voronoi Random Forest": "voronoi/2022-11-27-train-test/e-form-preds-IS2RE.csv",
     "Wrenformer": "wrenformer/2022-11-15-wrenformer-IS2RE-preds.csv",
     "MEGNet": "megnet/2022-11-18-megnet-wbm-IS2RE/megnet-e-form-preds.csv",

matbench_discovery/plots.py

@@ -280,6 +280,7 @@ def rolling_mae_vs_hull_dist(
     backend: Backend = "plotly",
     y_label: str = "rolling MAE (eV/atom)",
     just_plot_lines: bool = False,
+    with_sem: bool = True,
     **kwargs: Any,
 ) -> plt.Axes | go.Figure:
     """Rolling mean absolute error as the energy to the convex hull is varied. A scale
@@ -307,6 +308,8 @@ def rolling_mae_vs_hull_dist(
         just_plot_line (bool, optional): If True, plot only the rolling MAE, no shapes
             and annotations. Also won't plot the standard error in the mean. Defaults
             to False.
+        with_sem (bool, optional): If True, plot the standard error of the mean as
+            shaded area around the rolling MAE. Defaults to True.
 
     Returns:
         tuple[plt.Axes | go.Figure, pd.DataFrame, pd.DataFrame]: matplotlib Axes or
@@ -358,7 +361,7 @@ def rolling_mae_vs_hull_dist(
     if backend == "matplotlib":
         # assert df_rolling_err.isna().sum().sum() == 0, "NaNs in df_rolling_err"
         # assert df_err_std.isna().sum().sum() == 0, "NaNs in df_err_std"
-        # for model in df_rolling_err:
+        # for model in df_rolling_err if with_sem else []:
         #     ax.fill_between(
         #         bins,
         #         df_rolling_err[model] + df_err_std[model],
@@ -413,7 +416,7 @@ def rolling_mae_vs_hull_dist(
         ax.set(xlim=x_lim, ylim=y_lim)
 
     elif backend == "plotly":
-        for idx, model in enumerate(df_rolling_err):
+        for idx, model in enumerate(df_rolling_err if with_sem else []):
             # set legendgroup to model name so SEM shading toggles with model curve
             ax.data[idx].legendgroup = model
             # set SEM area to same color as model curve
@@ -466,7 +469,7 @@ def rolling_mae_vs_hull_dist(
         ax.add_annotation(
             x=-dft_acc,
             y=dft_acc,
-            text=f"<a {href=}>Corrected GGA Accuracy</a> "
+            text=f"<a {href=}>Corrected GGA Accuracy<br>for rel. Energy</a> "
             "[<a href='#hautier_accuracy_2012' target='_self'>ref</a>]",
             showarrow=True,
             xshift=-10,

scripts/compile_metrics.py

@@ -7,11 +7,13 @@
 import requests
 import wandb
 import wandb.apis.public
+from pymatviz.utils import save_fig
 from sklearn.metrics import f1_score, r2_score
 from tqdm import tqdm
 
 from matbench_discovery import FIGS, MODELS, WANDB_PATH, today
 from matbench_discovery.data import PRED_FILENAMES, load_df_wbm_preds
+from matbench_discovery.plots import px
 
 __author__ = "Janosh Riebesell"
 __date__ = "2022-11-28"
@@ -76,15 +78,14 @@
 # test set.
 
 for model in (pbar := tqdm(models)):
-    model_dict = models[model]
-    n_runs, filters = (model_dict.get(x) for x in ("n_runs", "filters"))
+    n_runs, filters = (models[model].get(x) for x in ("n_runs", "filters"))
     if n_runs == 0 or model in model_stats:
         continue
     pbar.set_description(model)
-    if "runs" in models:
-        runs: wandb.apis.public.Runs = models["runs"]
+    if "runs" in models[model]:
+        runs: wandb.apis.public.Runs = models[model]["runs"]
     else:
-        models["runs"] = runs = wandb.Api().runs(WANDB_PATH, filters=filters)
+        models[model]["runs"] = runs = wandb.Api().runs(WANDB_PATH, filters=filters)
 
     assert len(runs) == n_runs, f"found {len(runs)=} for {model}, expected {n_runs}"
 
@@ -96,8 +97,7 @@
 
     n_gpu, n_cpu = metadata.get("gpu_count", 0), metadata.get("cpu_count", 0)
     model_stats[model] = {
-        "run_time": run_time_total,
-        "run_time_h": f"{run_time_total / 3600:.1f} h",
+        "run_time_h": run_time_total / 3600,
         "GPU": n_gpu,
         "CPU": n_cpu,
         "slurm_jobs": n_runs,
@@ -109,7 +109,7 @@
     title=f"Run time distribution for {model}", xlabel="Run time [h]", ylabel="Count"
 )
 
-
+df_metrics = pd.DataFrame(model_stats).T
 # on 2022-11-28:
 # run_times = {'Voronoi Random Forest': 739608,
 #  'Wrenformer': 208399,
@@ -119,14 +119,12 @@
 
 
 # %%
-df_wbm = load_df_wbm_preds(models=list(models))
+df_wbm = load_df_wbm_preds(list(models))
 e_form_col = "e_form_per_atom_mp2020_corrected"
 each_col = "e_above_hull_mp2020_corrected_ppd_mp"
 
 
 # %%
-df_metrics = pd.DataFrame(model_stats).T
-
 for model in models:
     dct = {}
     e_above_hull_pred = df_wbm[model] - df_wbm[e_form_col]
@@ -154,7 +152,7 @@
 )
 
 
-# %%
+# %% export model metrics as styled HTML table
 styles = {
     "": "font-family: sans-serif; border-collapse: collapse;",
     "td, th": "border: 1px solid #ddd; text-align: left; padding: 8px;",
@@ -165,10 +163,38 @@
 # df_styled.to_html(html_path)
 
 
-# %%
-df = load_df_wbm_preds(list(models))
+# %% write model metrics to json for use by the website
+df_metrics["missing_preds"] = df_wbm[list(models)].isna().sum()
+df_metrics["missing_percent"] = [
+    f"{x / len(df_wbm):.2%}" for x in df_metrics.missing_preds
+]
 
-df_metrics["missing_preds"] = df[list(models)].isna().sum()
-df_metrics["missing_percent"] = [f"{x / len(df):.2%}" for x in df_metrics.missing_preds]
+df_metrics.attrs["total_run_time"] = df_metrics.run_time.sum()
 
 df_metrics.round(2).to_json(f"{MODELS}/{today}-model-stats.json", orient="index")
+
+
+# %% plot model run times as pie chart
+fig = px.pie(
+    df_metrics, values="run_time", names=df_metrics.index, hole=0.5
+).update_traces(
+    textinfo="percent+label",
+    textfont_size=14,
+    marker=dict(line=dict(color="#000000", width=2)),
+    hoverinfo="label+percent+name",
+    texttemplate="%{label}<br>%{percent:.1%}",
+    hovertemplate="%{label} %{percent:.1%} (%{value:.1f} h)",
+    rotation=90,
+    showlegend=False,
+)
+fig.add_annotation(
+    # add title in the middle saying "Total CPU+GPU time used"
+    text=f"Total CPU+GPU<br>time used:<br>{df_metrics.run_time.sum():.1f} h",
+    font=dict(size=18),
+    x=0.5,
+    y=0.5,
+    showarrow=False,
+)
+fig.update_layout(margin=dict(l=0, r=0, t=0, b=0))
+
+save_fig(fig, f"{FIGS}/{today}-model-run-times-pie.svelte")

scripts/make_api_docs.py

@@ -1,31 +1,34 @@
-import json
 import os
 from glob import glob
-from subprocess import run
 
-# Update auto-generated API docs. Also tweak lazydocs's markdown output for
-# - prettier badges linking to source code on GitHub
-# - remove bold tags since they break inline code
+from lazydocs import generate_docs
+
+from matbench_discovery import ROOT, URLs
+
+# Update auto-generated API docs.
 
-pkg = json.load(open("site/package.json"))
-route = "site/src/routes/api"
+out_path = f"{ROOT}/site/src/routes/api"
 
-for path in glob(f"{route}/*.md"):
+for path in glob(f"{out_path}/*.md"):
     os.remove(path)
 
-run(
-    f"lazydocs matbench_discovery --output-path {route} "
-    f"--no-watermark --src-base-url {pkg['repository']}/blob/main",
-    shell=True,
+generate_docs(
+    ["matbench_discovery"],
+    output_path=out_path,
+    watermark=False,
+    src_base_url=f"{URLs['Repo']}/blob/-",
 )
 
-for path in glob(f"{route}/*.md"):
-    markdown = open(path).read()
-    # remove <b> tags from generated markdown as they break inline code
-    markdown = markdown.replace("<b>", "").replace("</b>", "")
-    # improve style of badges linking to source code on GitHub
-    markdown = markdown.replace(
+# Tweak lazydocs's markdown output:
+# - remove bold tags since they break inline code
+# - make badges linking to GitHub source code blue with flat style, add alt text
+for path in glob(f"{out_path}/*.md"):
+    text = open(path).read()
+
+    text = text.replace("<b>", "").replace("</b>", "")
+
+    text = text.replace(
         'src="https://img.shields.io/badge/-source-cccccc?style=flat-square"',
         'src="https://img.shields.io/badge/source-blue?style=flat" alt="source link"',
     )
-    open(path, "w").write(markdown)
+    open(path, "w").write(text)

scripts/rolling_mae_vs_hull_dist_all_models.py

@@ -38,6 +38,7 @@
     e_above_hull_true=df_wbm[e_above_hull_col],
     e_above_hull_errors=df_wbm.filter(like=" MAE="),
     backend=backend,
+    with_sem=False,
     # template="plotly_white",
 )
 
@@ -51,13 +52,16 @@
     for line in fig.lines:
         line._linewidth *= 2
 else:
-    # select only every 10th point from each trace
+    # keep every 10th point from each trace to reduce plot size for website
     for trace in fig.data:
         if trace.name and trace.name.startswith("MAE") and len(trace.x) < 100:
             continue  # skip the MAE < DFT error area traces
         trace.x = trace.x[::10]
         trace.y = trace.y[::10]
 
+    # increase line width
+    fig.update_traces(line=dict(width=3))
+
     # increase legend handle size and reverse order
     fig.update_layout(legend=dict(itemsizing="constant"), legend_traceorder="reversed")
 

scripts/scatter_e_above_hull_models.py

@@ -122,8 +122,8 @@ def _add_metrics_to_legend(fig: go.Figure) -> None:
     facet_row_spacing=0.15,
     hover_data=hover_cols,
     hover_name=id_col,
-    color="clf",
-    color_discrete_map=dict(zip(classes, ("green", "yellow", "red", "blue"))),
+    # color="clf",
+    # color_discrete_map=dict(zip(classes, px.colors.qualitative.Pastel)),
     # opacity=0.4,
     range_x=[-2, 2],
     range_y=[-2, 2],
@@ -181,7 +181,39 @@ def _add_metrics_to_legend(fig: go.Figure) -> None:
     **axis_titles,
 )
 
+# add transparent rectangle with TN, TP, FN, FP labels in each quadrant
+for idx, _model in enumerate(models, 1):
+    for s1, s2, color, label in (
+        (-1, -1, "green", "TP"),
+        (-1, 1, "yellow", "FN"),
+        (1, -1, "red", "FP"),
+        (1, 1, "blue", "TN"),
+    ):
+        fig.add_shape(
+            type="rect",
+            x0=0,
+            y0=0,
+            x1=s1 * 100,
+            y1=s2 * 100,
+            fillcolor=color,
+            opacity=0.3,
+            layer="below",
+            xref=f"x{idx if idx > 1 else ''}",
+            yref=f"y{idx if idx > 1 else ''}",
+        )
+        fig.add_annotation(
+            xref=f"x{idx if idx > 1 else ''}",
+            yref=f"y{idx if idx > 1 else ''}",
+            x=s1 * 1.7,
+            y=s2 * 1.7,
+            text=label,
+            showarrow=False,
+            font=dict(size=16),
+        )
 
 fig.show()
+
+
+# %%
 img_path = f"{STATIC}/{today}-each-scatter-models.webp"
 save_fig(fig, img_path, scale=4, width=1000, height=500)

site/src/figs/2023-01-26-model-run-times-pie.svelte

@@ -0,0 +1 @@
+<div {...$$props}>                            <div id="62ff26e4-7a3d-4d32-a0e7-5b50a3c0eed0" class="plotly-graph-div" style="height:100%; width:100%;"></div>            <script type="text/javascript">                                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById("62ff26e4-7a3d-4d32-a0e7-5b50a3c0eed0")) {                    Plotly.newPlot(                        "62ff26e4-7a3d-4d32-a0e7-5b50a3c0eed0",                        [{"domain":{"x":[0.0,1.0],"y":[0.0,1.0]},"hole":0.5,"hovertemplate":"%{label} %{percent:.1%} (%{value:.1f} h)","labels":["CGCNN","Voronoi Random Forest","Wrenformer","MEGNet","M3GNet","BOWSR MEGNet"],"legendgroup":"","name":"","showlegend":false,"values":[11.349722222222223,204.7,57.88861111111111,3.4433333333333334,83.64944444444444,2776.570277777778],"type":"pie","textfont":{"size":14},"marker":{"line":{"color":"#000000","width":2}},"hoverinfo":"label+percent+name","rotation":90,"textinfo":"percent+label","texttemplate":"%{label}<br>%{percent:.1%}"}],                        {"template":{"data":{"barpolar":[{"marker":{"line":{"color":"rgb(17,17,17)","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"barpolar"}],"bar":[{"error_x":{"color":"#f2f5fa"},"error_y":{"color":"#f2f5fa"},"marker":{"line":{"color":"rgb(17,17,17)","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"bar"}],"carpet":[{"aaxis":{"endlinecolor":"#A2B1C6","gridcolor":"#506784","linecolor":"#506784","minorgridcolor":"#506784","startlinecolor":"#A2B1C6"},"baxis":{"endlinecolor":"#A2B1C6","gridcolor":"#506784","linecolor":"#506784","minorgridcolor":"#506784","startlinecolor":"#A2B1C6"},"type":"carpet"}],"choropleth":[{"colorbar":{"outlinewidth":0,"ticks":""},"type":"choropleth"}],"contourcarpet":[{"colorbar":{"outlinewidth":0,"ticks":""},"type":"contourcarpet"}],"contour":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"contour"}],"heatmapgl":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"heatmapgl"}],"heatmap":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"heatmap"}],"histogram2dcontour":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"histogram2dcontour"}],"histogram2d":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"histogram2d"}],"histogram":[{"marker":{"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"histogram"}],"mesh3d":[{"colorbar":{"outlinewidth":0,"ticks":""},"type":"mesh3d"}],"parcoords":[{"line":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"parcoords"}],"pie":[{"automargin":true,"type":"pie"}],"scatter3d":[{"line":{"colorbar":{"outlinewidth":0,"ticks":""}},"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scatter3d"}],"scattercarpet":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scattercarpet"}],"scattergeo":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scattergeo"}],"scattergl":[{"marker":{"line":{"color":"#283442"}},"type":"scattergl"}],"scattermapbox":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scattermapbox"}],"scatterpolargl":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scatterpolargl"}],"scatterpolar":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scatterpolar"}],"scatter":[{"marker":{"line":{"color":"#283442"}},"type":"scatter"}],"scatterternary":[{"marker":{"colorbar":{"outlinewidth":0,"ticks":""}},"type":"scatterternary"}],"surface":[{"colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"type":"surface"}],"table":[{"cells":{"fill":{"color":"#506784"},"line":{"color":"rgb(17,17,17)"}},"header":{"fill":{"color":"#2a3f5f"},"line":{"color":"rgb(17,17,17)"}},"type":"table"}]},"layout":{"annotationdefaults":{"arrowcolor":"#f2f5fa","arrowhead":0,"arrowwidth":1},"autotypenumbers":"strict","coloraxis":{"colorbar":{"outlinewidth":0,"ticks":""}},"colorscale":{"diverging":[[0,"#8e0152"],[0.1,"#c51b7d"],[0.2,"#de77ae"],[0.3,"#f1b6da"],[0.4,"#fde0ef"],[0.5,"#f7f7f7"],[0.6,"#e6f5d0"],[0.7,"#b8e186"],[0.8,"#7fbc41"],[0.9,"#4d9221"],[1,"#276419"]],"sequential":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"sequentialminus":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]},"colorway":["#636efa","#EF553B","#00cc96","#ab63fa","#FFA15A","#19d3f3","#FF6692","#B6E880","#FF97FF","#FECB52"],"font":{"color":"#f2f5fa","size":13},"geo":{"bgcolor":"rgb(17,17,17)","lakecolor":"rgb(17,17,17)","landcolor":"rgb(17,17,17)","showlakes":true,"showland":true,"subunitcolor":"#506784"},"hoverlabel":{"align":"left"},"hovermode":"closest","mapbox":{"style":"dark"},"paper_bgcolor":"rgba(0,0,0,0)","plot_bgcolor":"rgb(17,17,17)","polar":{"angularaxis":{"gridcolor":"#506784","linecolor":"#506784","ticks":""},"bgcolor":"rgb(17,17,17)","radialaxis":{"gridcolor":"#506784","linecolor":"#506784","ticks":""}},"scene":{"xaxis":{"backgroundcolor":"rgb(17,17,17)","gridcolor":"#506784","gridwidth":2,"linecolor":"#506784","showbackground":true,"ticks":"","zerolinecolor":"#C8D4E3"},"yaxis":{"backgroundcolor":"rgb(17,17,17)","gridcolor":"#506784","gridwidth":2,"linecolor":"#506784","showbackground":true,"ticks":"","zerolinecolor":"#C8D4E3"},"zaxis":{"backgroundcolor":"rgb(17,17,17)","gridcolor":"#506784","gridwidth":2,"linecolor":"#506784","showbackground":true,"ticks":"","zerolinecolor":"#C8D4E3"}},"shapedefaults":{"line":{"color":"#f2f5fa"}},"sliderdefaults":{"bgcolor":"#C8D4E3","bordercolor":"rgb(17,17,17)","borderwidth":1,"tickwidth":0},"ternary":{"aaxis":{"gridcolor":"#506784","linecolor":"#506784","ticks":""},"baxis":{"gridcolor":"#506784","linecolor":"#506784","ticks":""},"bgcolor":"rgb(17,17,17)","caxis":{"gridcolor":"#506784","linecolor":"#506784","ticks":""}},"title":{"x":0.05},"updatemenudefaults":{"bgcolor":"#506784","borderwidth":0},"xaxis":{"automargin":true,"gridcolor":"#283442","linecolor":"#506784","ticks":"","title":{"standoff":15},"zerolinecolor":"#283442","zerolinewidth":2},"yaxis":{"automargin":true,"gridcolor":"#283442","linecolor":"#506784","ticks":"","title":{"standoff":15},"zerolinecolor":"#283442","zerolinewidth":2},"margin":{"b":20,"l":30,"r":20,"t":60}}},"legend":{"tracegroupgap":0},"annotations":[{"font":{"size":18},"showarrow":false,"text":"Total CPU+GPU<br>time used:<br>3137.6 h","x":0.5,"y":0.5}],"margin":{"l":0,"r":0,"t":0,"b":0}},                        {"showTips": false, "modeBarButtonsToRemove": ["lasso2d", "select2d", "autoScale2d", "toImage"], "responsive": true, "displaylogo": false}                    )                };                            </script>        </div>