update inspect from voila review

aiidalab_sssp/inspect/__init__.py

@@ -0,0 +1,2 @@
+# -*- coding: utf-8 -*-
+__version__ = "0.0.1"

aiidalab_sssp/plot.py → aiidalab_sssp/inspect/plot_utils.py

@@ -1,13 +1,42 @@
+import random
+
 import matplotlib.pyplot as plt
 import numpy as np
 
 
+def cmap(label: str) -> str:
+    """Return RGB string of color for given standard psp label"""
+    _, pp_family, pp_z, pp_type, pp_version = label.split("/")
+
+    if pp_family == "sg15" and pp_version == "v1.0":
+        return "#000000"
+
+    if pp_family == "sg15" and pp_version == "v1.2":
+        return "#708090"
+
+    if pp_family == "gbrv" and pp_version == "v1":
+        return "#4682B4"
+
+    if pp_family == "psl" and pp_version == "v1.0.0" and pp_type == "us":
+        return "#F50E02"
+
+    if pp_family == "psl" and pp_version == "v1.0.0" and pp_type == "paw":
+        return "#2D8B00F"
+
+    if pp_family == "dojo" and pp_version == "v04":
+        return "#F9A501"
+
+    # TODO: more mapping
+    # if a unknow type generate random color based on ascii sum
+    ascn = sum([ord(c) for c in label])
+    random.seed(ascn)
+    return "#%06x" % random.randint(0, 0xFFFFFF)
+
+
 def delta_measure_hist(pseudos: dict, measure_type):
 
     px = 1 / plt.rcParams["figure.dpi"]  # pixel in inches
     fig, ax = plt.subplots(1, 1, figsize=(1024 * px, 360 * px))
-    cmap = plt.get_cmap("tab20")
-    NUM_COLOR = 20
     structures = ["X", "XO", "X2O", "XO3", "X2O", "X2O3", "X2O5"]
     num_structures = len(structures)
 
@@ -23,7 +52,7 @@ def delta_measure_hist(pseudos: dict, measure_type):
         ylabel = "Δ -factor"
     elif measure_type == "nv_delta":
         keyname = "rel_errors_vec_length"
-        ylabel = "νΔ -factor"
+        ylabel = "ν -factor"
 
     for i, (label, output) in enumerate(pseudos.items()):
         N = len(structures)
@@ -44,7 +73,13 @@ def delta_measure_hist(pseudos: dict, measure_type):
         out_label = f"{pp_z}/{pp_type}({pp_family}-{pp_version})"
 
         ax.bar(
-            idx + width * i, y_delta, width, color=cmap(i / NUM_COLOR), label=out_label
+            idx + width * i,
+            y_delta,
+            width,
+            color=cmap(label),
+            edgecolor="black",
+            linewidth=1,
+            label=out_label,
         )
         ax.legend()
         ax.set_title(f"X={element}")
@@ -65,10 +100,8 @@ def convergence(pseudos: dict, wf_name, measure_name, ylabel, threshold=None):
     fig, (ax1, ax2) = plt.subplots(
         1, 2, gridspec_kw={"width_ratios": [2, 1]}, figsize=(1024 * px, 360 * px)
     )
-    cmap = plt.get_cmap("tab20")
-    NUM_COLOR = 20
 
-    for i, (label, output) in enumerate(pseudos.items()):
+    for label, output in pseudos.items():
         # Calculate the avg delta measure value
         structures = ["X", "XO", "X2O", "XO3", "X2O", "X2O3", "X2O5"]
         lst = []
@@ -94,18 +127,14 @@ def convergence(pseudos: dict, wf_name, measure_name, ylabel, threshold=None):
             wfc_cutoff = res["final_output_parameters"]["wfc_cutoff"]
 
             _, pp_family, pp_z, pp_type, pp_version = label.split("/")
-            out_label = (
-                f"{pp_z}/{pp_type}(νΔ={avg_delta:.2f})({pp_family}-{pp_version})"
-            )
+            out_label = f"{pp_z}/{pp_type}(ν={avg_delta:.2f})({pp_family}-{pp_version})"
 
-            ax1.plot(
-                x_wfc, y_wfc, marker="o", color=cmap(i / NUM_COLOR), label=out_label
-            )
+            ax1.plot(x_wfc, y_wfc, marker="o", color=cmap(label), label=out_label)
             ax2.plot(
                 x_rho,
                 y_rho,
                 marker="o",
-                color=cmap(i / NUM_COLOR),
+                color=cmap(label),
                 label=f"cutoff wfc = {wfc_cutoff} Ry",
             )
         except Exception:

aiidalab_sssp/inspect/subwidgets/periodic_table.py

@@ -0,0 +1,65 @@
+import ipywidgets as ipw
+from widget_periodictable import PTableWidget
+
+__all__ = ("PeriodicTable",)
+
+
+class PeriodicTable(ipw.VBox):
+    """Wrapper-widget for PTableWidget"""
+
+    def __init__(self, extended: bool = True, **kwargs):
+        self._disabled = kwargs.get("disabled", False)
+
+        self.select_any_all = ipw.Checkbox(
+            value=False,
+            description="Structures can include any chosen elements (instead of all)",
+            indent=False,
+            layout={"width": "auto"},
+            disabled=self.disabled,
+        )
+        self.ptable = PTableWidget(**kwargs)
+        self.ptable_container = ipw.VBox(
+            children=(self.select_any_all, self.ptable),
+            layout={
+                "width": "auto",
+                "height": "auto" if extended else "0px",
+                "visibility": "visible" if extended else "hidden",
+            },
+        )
+
+        super().__init__(
+            children=(self.ptable_container,),
+            layout=kwargs.get("layout", {}),
+        )
+
+    @property
+    def value(self) -> dict:
+        """Return value for wrapped PTableWidget"""
+
+        return not self.select_any_all.value, self.ptable.selected_elements.copy()
+
+    @property
+    def disabled(self) -> None:
+        """Disable widget"""
+        return self._disabled
+
+    @disabled.setter
+    def disabled(self, value: bool) -> None:
+        """Disable widget"""
+        if not isinstance(value, bool):
+            raise TypeError("disabled must be a boolean")
+
+        self.select_any_all.disabled = self.ptable.disabled = value
+
+    def reset(self):
+        """Reset widget"""
+        self.select_any_all.value = False
+        self.ptable.selected_elements = {}
+
+    def freeze(self):
+        """Disable widget"""
+        self.disabled = True
+
+    def unfreeze(self):
+        """Activate widget (in its current state)"""
+        self.disabled = False

aiidalab_sssp/widgets_plot.py → aiidalab_sssp/inspect/subwidgets/plot.py

@@ -2,19 +2,19 @@
 import traitlets
 from IPython.display import clear_output, display
 
-from aiidalab_sssp.plot import convergence, delta_measure_hist
+from aiidalab_sssp.inspect.plot_utils import convergence, delta_measure_hist
 
 
 class PlotDeltaMeasureWidget(ipw.VBox):
 
-    selected_pseudos = traitlets.Dict()
+    selected_pseudos = traitlets.Dict(allow_none=True)
 
     def __init__(self):
         # measure button
         # self.measure_tab = ipw.Tab(title=['Δ-factor', 'νΔ-factor'])
         self.measure_tab = ipw.Tab()
-        self.measure_tab.set_title(0, "Δ-factor")
-        self.measure_tab.set_title(1, "νΔ-factor")
+        self.measure_tab.set_title(0, "ν-factor")
+        self.measure_tab.set_title(1, "Δ-factor")
 
         # Delta mesure
         self.output_delta_measure = ipw.Output()
@@ -28,23 +28,25 @@ def __init__(self):
 
     @traitlets.observe("selected_pseudos")
     def _on_pseudos_change(self, change):
-        out_delta = ipw.Output()
         out_nv_delta = ipw.Output()
-        with out_delta:
-            fig = delta_measure_hist(change["new"], "delta")
-            display(fig)
+        out_delta = ipw.Output()
+
+        if change["new"]:
+            with out_nv_delta:
+                fig = delta_measure_hist(change["new"], "nv_delta")
+                display(fig)
 
-        with out_nv_delta:
-            fig = delta_measure_hist(change["new"], "nv_delta")
-            display(fig)
+            with out_delta:
+                fig = delta_measure_hist(change["new"], "delta")
+                display(fig)
 
-        children = [out_delta, out_nv_delta]
+        children = [out_nv_delta, out_delta]
         self.measure_tab.children = children
 
 
 class _PlotConvergenBaseWidget(ipw.VBox):
 
-    selected_pseudos = traitlets.Dict()
+    selected_pseudos = traitlets.Dict(allow_none=True)
 
     _WF = "Not implement"
     _MEASURE = "Not implement"
@@ -66,14 +68,15 @@ def _on_pseudos_change(self, change):
         with self.output:
             clear_output(wait=True)
 
-            fig = convergence(
-                change["new"],
-                wf_name=self._WF,
-                measure_name=self._MEASURE,
-                ylabel=self._YLABEL,
-                threshold=self._THRESHOLD,
-            )
-            display(fig)
+            if change["new"]:
+                fig = convergence(
+                    change["new"],
+                    wf_name=self._WF,
+                    measure_name=self._MEASURE,
+                    ylabel=self._YLABEL,
+                    threshold=self._THRESHOLD,
+                )
+                display(fig)
 
 
 class PlotCohesiveEnergyConvergeWidget(_PlotConvergenBaseWidget):