Release/v0.1.4

deploy.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+
+set -euo pipefail
+
+source .env
+
+aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin ${ECR_AWS_ACCOUNT_ID}.dkr.ecr.${ECR_AWS_REGION}.amazonaws.com
+
+
+docker compose -f docker-compose.yml pull
+docker compose -f docker-compose.yml up -d
+docker compose -f docker-compose.yml logs -f

hydroshift/__init__.py

@@ -1 +1 @@
-__version__ = "0.1.3"
+__version__ = "0.1.4"

hydroshift/_pages/changepoint.py

@@ -1,5 +1,6 @@
 """A tool to identify changepoints in hydrologic timeseries."""
 
+from pydoc import doc
 import traceback
 import uuid
 from collections import defaultdict
@@ -10,6 +11,7 @@
 import streamlit as st
 from docx import Document
 from docx.shared import Inches
+from docx.enum.text import WD_PARAGRAPH_ALIGNMENT
 from plotly.graph_objects import Figure
 
 from hydroshift.consts import (
@@ -195,6 +197,12 @@ def references(self) -> str:
     def word_data(self) -> BytesIO:
         """Export text as MS word."""
         document = Document()
+        s = document.sections[0]
+        s.header_distance = Inches(0.2)
+        p = s.header.add_paragraph()
+        p.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        r = p.add_run()
+        r.add_picture('hydroshift/images/dewberry_full_logo.jpg', height=Inches(0.3))
         document.add_heading(self.title, level=1)
         document.add_heading("Summary", level=2)
         self.add_markdown_to_doc(document, self.summary_text)

hydroshift/_pages/summary.py

@@ -57,9 +57,9 @@ def section_lp3(gage: Gage):
             }[est_method]
         with opt_col_2:
             with st.container():
-                use_map = st.toggle("Use regional skew", value=False, disabled=not gage.has_regional_skew)
-                lp3 = LP3Analysis(gage.gage_id, gage.ams_vals, use_map, est_method, "")
-                st.badge(f"Using skew value of {round(lp3.parameters[2], 2)}", color="blue")
+                skew_mode = st.radio("Skew Source", options=["Station Skew", "Regional Skew", "Weighted Skew"], horizontal=True, disabled=not gage.has_regional_skew)
+                lp3 = LP3Analysis(gage.gage_id, gage.ams_vals, skew_mode, est_method, "")
+                st.badge(f"Using skew value of {str(round(lp3.parameters[2], 2))}", color="blue")
 
         # Analysis and display
         if gage.missing_dates_ams is not None and len(gage.missing_dates_ams) > 0:

hydroshift/consts.py

@@ -42,6 +42,7 @@
 PEAKFQ_URL = "https://www.usgs.gov/tools/peakfq"
 DEWBERRY_URL = "https://www.dewberry.com/"
 GITHUB_URL = "https://github.com/Dewberry/hydroshift"
+BUG_URL = "https://github.com/Dewberry/hydroshift/issues"
 ADMIN_EMAIL = "klawson@dewberry.com"
 
 ### Text snippets ###
@@ -73,4 +74,5 @@ def svg2text(path: str) -> str:
 
 GITHUB_SVG = svg2text("hydroshift/images/github_logo.svg")
 DEWBERRY_SVG = svg2text("hydroshift/images/dewberry_logo.svg")
+BUG_SVG = svg2text("hydroshift/images/bug.svg")
 MAIL_SVG = svg2text("hydroshift/images/mail_logo.svg")

hydroshift/templates/changepoint_description.md

@@ -10,7 +10,7 @@ These test statistics were utilized in two distinct ways in this analysis.  A st
 A streaming analysis was performed by treating the data as a stream of values and repeating the static analysis after each new value was added. If the test statistic exceeds a specified threshold at any point within the subseries, a changepoint is marked, and a new data series is initialized for all further test statistics. The resulting change points are shown as dashed red lines in the top panel of Figure 1.
 
  - The threshold for identifying changepoints in the streaming analysis is defined using an Average Run Length (ARL0) parameter. ARL0 reflects the frequency with which a false positive would be raised on a stationary timeseries (e.g., for an ARL0 of 1,000 a false changepoint would be identified on a stationary timeseries on average every 1,000 samples.).  For this analysis, an ARL0 of {{ arl0 }} was used.
- - A burn-in period of {{ burn_in }} years was selected to ignore singificant change points in the first {{ burn_in }} years of the record due to the influence of small sample sizes.
+ - A burn-in period of {{ burn_in }} years was selected to ignore significant change points in the first {{ burn_in }} years of the record due to the influence of small sample sizes.
 
 If the flood peak series contained any gaps, the data around the gap was joined as if there was no gap between the data points. For large data gaps, this assumption could lead to inaccurate results.
 

hydroshift/templates/data_sources_side_bar.html

@@ -50,9 +50,12 @@
       <a href="{{ GITHUB_URL }}" target="_blank" title="GitHub Repository">
         {{ GITHUB_SVG | safe }}
       </a>
-      <a href="mailto:{{ ADMIN_EMAIL }}" target="_blank" title="Report a bug">
+      <a href="mailto:{{ ADMIN_EMAIL }}" target="_blank" title="Contact Authors">
         {{ MAIL_SVG | safe }}
       </a>
+      <a href="{{ BUG_URL }}" target="_blank" title="Report a bug">
+        {{ BUG_SVG | safe }}
+      </a>
       <a href="{{ DEWBERRY_URL }}" target="_blank" title="Dewberry">
         {{ DEWBERRY_SVG | safe }}
       </a>

hydroshift/utils/ffa.py

@@ -15,7 +15,7 @@ class LP3Analysis:
 
     gage_id: str
     peaks: list
-    use_map_skew: bool = False
+    skew_mode: str = "Station Skew"
     est_method: str = "MLE"
     label: str = ""
     return_periods: List[str] = field(default_factory=lambda: [1.1, 2, 5, 10, 25, 50, 100, 500])
@@ -41,8 +41,10 @@ def parameters(self) -> tuple[float]:
         elif self.est_method == "LMOM":
             mean_log, std_log, l3 = l_moments(self.log_peaks)
             skew_log = l3 / (std_log * 0.7797)  # pseudo-stdev
-        if self.use_map_skew:
+        if self.skew_mode == "Weighted Skew":
             skew_log = self.weighted_skew
+        elif self.skew_mode == "Regional Skew":
+            skew_log = self.map_skew
         return mean_log, std_log, skew_log
 
     @property

hydroshift/utils/plots.py

@@ -230,10 +230,7 @@ def plot_lp3(data: LP3Analysis | list[LP3Analysis]):
 
     # Formatting
     return_periods = [int(i) if i.is_integer() else round(i, 1) for i in i.return_periods]
-    if i.use_map_skew:
-        skew_txt = ""
-    else:
-        skew_txt = " (No Regional Skew)"
+    skew_txt = f" ({i.skew_mode})"
     fig.update_layout(
         title=f"{i.gage_id} | Log-Pearson Type III Estimates{skew_txt}",
         xaxis=dict(