[USGS-R#184] initial commit PP barplots

3_visualize/src/python_scripts/compare_baseline_multitask_reps.py

@@ -0,0 +1,117 @@
+# ---
+# jupyter:
+#   jupytext:
+#     text_representation:
+#       extension: .py
+#       format_name: light
+#       format_version: '1.5'
+#       jupytext_version: 1.14.4
+#   kernelspec:
+#     display_name: Python 3 (ipykernel)
+#     language: python
+#     name: python3
+# ---
+
+import pandas as pd
+import seaborn as sns
+import matplotlib.pyplot as plt
+
+validation_sites = ["01472104", "01473500", "01481500"]
+test_sites = ["01475530", "01475548"]
+
+models = ["0_baseline_LSTM", "2_multitask_dense"]
+
+
+def read_and_combine_dfs(model_ids, metric_type, partition, parent_dir):
+    f_name = "{p}{model}/exp_{metric_type}_metrics.csv"
+    dfs = []
+    for m in model_ids:
+        df = pd.read_csv(f_name.format(p=parent_dir, model=m, metric_type=metric_type), dtype={'site_id':str})
+        df['model'] = m
+        dfs.append(df)
+
+    df_comb = pd.concat(dfs)
+    df_comb = df_comb[df_comb['partition'] == partition]
+    df_comb = df_comb[df_comb['variable'].str.startswith('do')]
+    df_comb = df_comb[df_comb['rmse'].notna()]
+    return df_comb
+
+
+df_comb_reach_new = read_and_combine_dfs(models, 'reach', 'val', "../../")
+df_comb_reach_new['type'] = 'new inputs'
+
+df_comb_reach_old = read_and_combine_dfs(models, "reach", 'val', "archive_221215/")
+df_comb_reach_old['type'] = 'old inputs'
+
+df_comb_reach = pd.concat([df_comb_reach_new, df_comb_reach_old])
+
+# +
+g = sns.catplot(x='site_id', y='rmse', row='model', col='variable', data=df_comb_reach, hue='type', kind='bar', legend=False, ci='sd')
+g.set_xticklabels(rotation=90)
+for ax in g.axes.flatten():
+    ax.grid()
+    ax.set_axisbelow(True)
+
+plt.legend(bbox_to_anchor=(1.05, .55))
+plt.tight_layout()
+plt.savefig("figs/val_results_new_inputs.png")
+
+# +
+g = sns.catplot(x='site_id', y='rmse', col='variable', data=df_comb_reach_new, hue='model', kind='bar', legend=False, ci='sd')
+g.set_xticklabels(rotation=90)
+for ax in g.axes.flatten():
+    ax.grid()
+    ax.set_axisbelow(True)
+
+plt.legend(bbox_to_anchor=(1.05, .55))
+plt.tight_layout()
+plt.savefig("figs/val_results_new_inputs.png")
+# -
+
+g=sns.catplot(x='site_id', y='rmse', hue='model', col='variable', col_wrap=3, data=df_comb_reach, dodge=True, legend=False)
+g.set_xticklabels(rotation=90)
+for ax in g.axes.flatten():
+    ax.grid()
+plt.legend(bbox_to_anchor=(1.05, .55))
+plt.tight_layout()
+plt.savefig('val_results_strip.png')
+
+# +
+g=sns.catplot(x='variable', y='rmse', hue='model', col='site_id', col_wrap=3, data=df_comb_reach, dodge=True, legend=False)
+for site_id, ax in g.axes_dict.items():
+    ax.grid()
+    if site_id in validation_sites:
+        ax.text(1, 3.2, "**Validation Site**", ha='center')
+
+
+plt.legend(bbox_to_anchor=(1.5, 1.15))
+plt.tight_layout()
+plt.savefig('val_results_strip.png')
+# -
+
+df_comb = read_and_combine_dfs(models, 'overall', 'val', "archive_221215/")
+
+df_comb_new = read_and_combine_dfs(models, 'overall', 'val', "./")
+df_comb_new['type'] = 'new inputs'
+
+df_comb_old = read_and_combine_dfs(models, "overall", 'val', "archive_221215/")
+df_comb_old['type'] = 'old inputs'
+
+df_comb = pd.concat([df_comb_new, df_comb_old])
+
+g = sns.catplot(x='variable', y='rmse', data=df_comb, hue='type', col='model', kind="bar")
+for ax in g.axes.flatten():
+    ax.bar_label(ax.containers[0], label_type="center")
+    ax.bar_label(ax.containers[1], label_type="center")
+plt.savefig("figs/val_results_overall_new_inputs.png")
+
+# +
+ax = sns.barplot(x='variable', y='rmse', data=df_comb_new, hue='model')
+ax.bar_label(ax.containers[0], label_type="center")
+ax.bar_label(ax.containers[1], label_type="center")
+
+# plt.tight_layout()
+plt.savefig("figs/val_results_overall.png")
+# -
+
+