pyadps 0.2.0b0__py3-none-any.whl → 0.3.0__py3-none-any.whl

pyadps/pages/05_QC_Test.py

@@ -0,0 +1,493 @@
+import numpy as np
+import pandas as pd
+import plotly.express as px
+import plotly.graph_objects as go
+import streamlit as st
+from plotly.subplots import make_subplots
+from plotly_resampler import FigureResampler
+from streamlit.runtime.state import session_state
+from utils.signal_quality import (
+    ev_check,
+    false_target,
+    pg_check,
+    echo_check,
+    correlation_check,
+)
+
+if "flead" not in st.session_state:
+    st.write(":red[Please Select Data!]")
+    st.stop()
+
+
+def reset_qctest():
+    # Reset Global Test
+    st.session_state.isQCTest = False
+    # Reset Local Tests
+    st.session_state.isQCCheck_QCT = False
+
+    # Reset Data
+    st.session_state.isBeamModified_QCT = False
+    st.session_state.beam_direction_QCT = st.session_state.beam_direction
+
+    # As QC test is not saved the Sensor Page Returns is set to False
+    st.session_state.isSensorPageReturn = False
+
+    # Reset Mask Data
+    # Copy the sensor mask if sensor mask completed.
+    # Else copy the default mask.
+    if st.session_state.isSensorTest:
+        st.session_state.qc_mask = np.copy(st.session_state.sensor_mask)
+        st.session_state.qc_mask_temp = np.copy(st.session_state.sensor_mask)
+    else:
+        st.session_state.qc_mask = np.copy(st.session_state.orig_mask)
+        st.session_state.qc_mask_temp = np.copy(st.session_state.orig_mask)
+
+
+def hard_reset(option):
+    # Reset Global Test
+    st.session_state.isQCTest = False
+    # Reset Local Tests
+    st.session_state.isQCCheck_QCT = False
+    st.session_state.isBeamModified_QCT = False
+    # Reset Data
+    st.session_state.beam_direction_QCT = st.session_state.beam_direction
+
+    st.session_state.isSensorPageReturn = False
+
+    # Reset Mask Data based on user options
+    if option == "Sensor Test":
+        st.session_state.qc_mask = np.copy(st.session_state.sensor_mask)
+        st.session_state.qc_mask_temp = np.copy(st.session_state.sensor_mask)
+
+    elif option == "Default":
+        st.session_state.qc_mask = np.copy(st.session_state.orig_mask)
+        st.session_state.qc_mask_temp = np.copy(st.session_state.orig_mask)
+
+
+def save_qctest():
+    st.session_state.qc_mask = np.copy(st.session_state.qc_mask_temp)
+    st.session_state.isQCTest = True
+    st.session_state.isProfileMask = False
+    st.session_state.isGridSave = False
+    st.session_state.isVelocityMask = False
+    # Indicate previous pages that Test has been carried out
+    st.session_state.isSensorPageReturn = True
+
+
+def qc_submit():
+    # st.write(st.session_state.newthresh)
+    st.session_state.isQCCheck_QCT = True
+
+    # First Quality check of the page
+    mask = np.copy(st.session_state.qc_mask_temp)
+    # if st.session_state.isSensorTest:
+    #     mask = np.copy(st.session_state.sensor_mask)
+    # else:
+    #     mask = np.copy(st.session_state.default_mask)
+
+    ds = st.session_state.ds
+    pgt = st.session_state.pgt_QCT
+    ct = st.session_state.ct_QCT
+    et = st.session_state.et_QCT
+    evt = st.session_state.evt_QCT
+    ft = st.session_state.ft_QCT
+    is3beam = st.session_state.is3beam_QCT
+    beam_ignore = st.session_state.beam_to_ignore
+    mask = pg_check(ds, mask, pgt, threebeam=is3beam)
+    mask = correlation_check(ds, mask, ct,is3beam,beam_ignore=beam_ignore)
+    mask = echo_check(ds, mask, et,is3beam,beam_ignore=beam_ignore)
+    mask = ev_check(ds, mask, evt)
+    mask = false_target(ds, mask, ft, threebeam=is3beam, beam_ignore=beam_ignore)
+    # Store the processed mask in a temporary mask
+    st.session_state.qc_mask_temp = mask
+
+
+if st.session_state.isSensorTest:
+    st.write(":grey[Working on a saved mask file ...]")
+    if st.session_state.isQCPageReturn:
+        st.write(
+            ":orange[Warning: QC test already completed. Reset the mask file to change settings.]"
+        )
+        reset_selectbox = st.selectbox(
+            "Choose reset option",
+            ("Sensor Test", "Default"),
+            index=None,
+            placeholder="Reset mask to ...",
+        )
+        if reset_selectbox is not None:
+            hard_reset(reset_selectbox)
+    elif st.session_state.isFirstQCVisit:
+        reset_qctest()
+        st.session_state.isFirstQCVisit = False
+else:
+    if st.session_state.isFirstQCVisit:
+        # This will rest to the default mask file
+        reset_qctest()
+        st.session_state.isFirstQCVisit = False
+    st.write(":orange[Creating a new mask file ...]")
+
+
+# Load data
+ds = st.session_state.ds
+flobj = st.session_state.flead
+vlobj = st.session_state.vlead
+velocity = st.session_state.velocity
+echo = st.session_state.echo
+correlation = st.session_state.correlation
+pgood = st.session_state.pgood
+ensembles = st.session_state.head.ensembles
+cells = flobj.field()["Cells"]
+fdata = flobj.fleader
+vdata = vlobj.vleader
+x = np.arange(0, ensembles, 1)
+y = np.arange(0, cells, 1)
+
+
+@st.cache_data
+def fillplot_plotly(data, colorscale="balance"):
+    fig = FigureResampler(go.Figure())
+    data1 = np.where(data == -32768, np.nan, data)
+    fig.add_trace(
+        go.Heatmap(z=data1[:, 0:-1], x=x, y=y, colorscale=colorscale, hoverongaps=False)
+    )
+    st.plotly_chart(fig)
+
+
+@st.cache_data
+def lineplot(data, title, slope=None, xaxis="time"):
+    if xaxis == "time":
+        xdata = st.session_state.date
+    else:
+        xdata = st.session_state.ensemble_axis
+    scatter_trace = go.Scatter(
+        x=xdata,
+        y=data,
+        mode="markers",
+        name=title,
+        marker=dict(color="blue", size=10),  # Customize marker color and size
+    )
+    # Create the slope line trace
+    if slope is not None:
+        line_trace = go.Scatter(
+            x=xdata,
+            y=slope,
+            mode="lines",
+            name="Slope Line",
+            line=dict(color="red", width=2, dash="dash"),  # Dashed red line
+        )
+        fig = go.Figure(data=[scatter_trace, line_trace])
+    else:
+        fig = go.Figure(data=[scatter_trace])
+
+    st.plotly_chart(fig)
+
+
+@st.cache_data
+def plot_noise(dep=0, rec=-1):
+    n = dep
+    m = rec
+    colorleft = [
+        "rgb(240, 255, 255)",
+        "rgb(115, 147, 179)",
+        "rgb(100, 149, 237)",
+        "rgb(15, 82, 186)",
+    ]
+    colorright = [
+        "rgb(250, 200, 152)",
+        "rgb(255, 165, 0)",
+        "rgb(255, 95, 31)",
+        "rgb(139, 64, 0)",
+    ]
+    fig = make_subplots(
+        rows=1,
+        cols=2,
+        subplot_titles=[
+            f"Deployment Ensemble ({x[n]+1})",
+            f"Recovery Ensemble ({x[m]+1})",
+        ],
+    )
+    for i in range(4):
+        fig.add_trace(
+            go.Scatter(
+                x=echo[i, :, n],
+                y=y,
+                name=f"Beam (D) {i+1}",
+                line=dict(color=colorleft[i]),
+            ),
+            row=1,
+            col=1,
+        )
+    for i in range(4):
+        fig.add_trace(
+            go.Scatter(
+                x=echo[i, :, m],
+                y=y,
+                name=f"Beam (R)  {i+1}",
+                line=dict(color=colorright[i]),
+            ),
+            row=1,
+            col=2,
+        )
+
+    fig.update_layout(height=600, width=800, title_text="Echo Intensity")
+    fig.update_xaxes(title="Echo (count)")
+    fig.update_yaxes(title="Cells")
+    st.plotly_chart(fig)
+
+
+tab1, tab2, tab3, tab4, tab5 = st.tabs(
+    [
+        "Noise Floor Identification",
+        "QC Tests",
+        "Display Mask",
+        "Fix Orientation",
+        "Save Data",
+    ]
+)
+#########  NOISE FLOOR IDENTIFICATION ##############
+with tab1:
+    dn = rn = 1
+    st.header("Noise Floor Identification", divider="blue")
+    st.write(
+        """
+        If the ADCP has collected data from the air either 
+        before deployment or after recovery, this data can 
+        be used to estimate the echo intensity threshold. 
+        The plots below show the echo intensity from the first 
+        and last ensembles. The noise level is typically around 
+        30-40 counts throughout the entire profile.
+    """
+    )
+    dn = st.number_input("Deployment Ensemble", x[0] + 1, x[-1] + 1, x[0] + 1)
+    # r = st.number_input("Recovery Ensemble", -1 * (x[-1] + 1), -1 * (x[0] + 1), -1)
+    rn = st.number_input("Recovery Ensemble", x[0] + 1, x[-1] + 1, x[-1] + 1)
+    dn = dn - 1
+    rn = rn - 1
+
+    plot_noise(dep=dn, rec=rn)
+
+
+################## QC Test ###################
+with tab2:
+    st.header("Quality Control Tests", divider="blue")
+    st.write("")
+
+    left, right = st.columns([1, 1])
+    with left:
+        st.write(""" Teledyne RDI recommends these quality control tests, 
+                    some of which can be configured before deployment. 
+                    The pre-deployment values configured for the ADCP are listed 
+                    in the table below. The noise-floor identification graph above 
+                    can assist in determining the echo intensity threshold. 
+                    For more information about these tests, 
+                    refer to *Acoustic Doppler Current Profiler Principles of 
+                    Operation: A Practical Primer* by Teledyne RDI.""")
+        fdata = st.session_state.flead.field()
+        st.divider()
+        st.write(":blue-background[Additional Information:]")
+        st.write(f"Number of Pings per Ensemble: `{fdata["Pings"]}`")
+        st.write(f"Number of Beams: `{fdata["Beams"]}`")
+        st.divider()
+        st.write(":red-background[Thresholds used during deployment:]")
+        thresh = pd.DataFrame(
+            [
+                ["Correlation", fdata["Correlation Thresh"]],
+                ["Error Velocity", fdata["Error Velocity Thresh"]],
+                ["Echo Intensity", 0],
+                ["False Target", fdata["False Target Thresh"]],
+                ["Percentage Good", fdata["Percent Good Min"]],
+            ],
+            columns=["Threshold", "Values"],
+        )
+
+        st.write(thresh)
+
+    with right:
+        # with st.form(key="my_form"):
+        st.write("Would you like to apply new threshold?")
+
+        st.session_state.ct_QCT = st.number_input(
+            "Select Correlation Threshold",
+            0,
+            255,
+            fdata["Correlation Thresh"],
+        )
+
+        st.session_state.evt_QCT = st.number_input(
+            "Select Error Velocity Threshold",
+            0,
+            9999,
+            fdata["Error Velocity Thresh"],
+        )
+
+        st.session_state.et_QCT = st.number_input(
+            "Select Echo Intensity Threshold",
+            0,
+            255,
+            0,
+        )
+
+        st.session_state.ft_QCT = st.number_input(
+            "Select False Target Threshold",
+            0,
+            255,
+            fdata["False Target Thresh"],
+        )
+
+        st.session_state.is3beam_QCT = st.selectbox(
+            "Would you like to use a three-beam solution?", (True, False)
+        )
+
+        if st.session_state.is3beam_QCT:
+            beam_label_to_value = {
+                "None": None,
+                "Beam 1": 0,
+                "Beam 2": 1,
+                "Beam 3": 2,
+                "Beam 4": 3
+            }
+
+            selected_beam = st.selectbox(
+                "Select Beam to Ignore",
+                options=list(beam_label_to_value.keys()),
+                index=0  # Default is "None"
+            )
+            st.session_state.beam_to_ignore =  beam_label_to_value[selected_beam]
+
+        st.session_state.pgt_QCT = st.number_input(
+            "Select Percent Good Threshold",
+            0,
+            101,
+            fdata["Percent Good Min"],
+        )
+        submit_button = st.button(label="Submit", on_click=qc_submit)
+
+    # mask = st.session_state.qc_mask_temp
+    with left:
+        if submit_button:
+            st.session_state.isQCCheck_QCT = True
+            st.session_state.newthresh = pd.DataFrame(
+                [
+                    ["Correlation", str(st.session_state.ct_QCT)],
+                    ["Error Velocity", str(st.session_state.evt_QCT)],
+                    ["Echo Intensity", str(st.session_state.et_QCT)],
+                    ["False Target", str(st.session_state.ft_QCT)],
+                    ["Three Beam", str(st.session_state.is3beam_QCT)],
+                    ["Percentage Good", str(st.session_state.pgt_QCT)],
+                ],
+                columns=["Threshold", "Values"],
+            )
+
+        if st.session_state.isQCCheck_QCT:
+            st.write(":green-background[Current Thresholds]")
+            st.write(st.session_state.newthresh)
+
+
+with tab3:
+    st.header("Mask File", divider="blue")
+    st.write(
+        """
+    Display the mask file. 
+    Ensure to save any necessary changes or apply additional thresholds if needed.
+    """
+    )
+
+    leftplot, rightplot = st.columns([1, 1])
+    if st.button("Display mask file"):
+        with leftplot:
+            st.subheader("Default Mask File")
+            st.write(
+                """
+                    CAPTION:
+                    ADCP assigns missing values based on thresholds
+                    set before deployment. These values cannot be
+                    recovered and are part of default mask file.
+                """
+            )
+            fillplot_plotly(st.session_state.orig_mask, colorscale="greys")
+        with rightplot:
+            st.subheader("Updated Mask File")
+            # values, counts = np.unique(mask, return_counts=True)
+            st.write(
+                """
+                    CAPTION:
+                    Updated mask displayed after applying threshold.
+                    If thresholds are not saved, default mask
+                    is displayed. 
+                """
+            )
+            fillplot_plotly(st.session_state.qc_mask_temp, colorscale="greys")
+
+with tab4:
+    ################## Fix Orientation ###################
+    st.subheader("Fix Orientation", divider="orange")
+
+    if st.session_state.beam_direction == "Up":
+        beamalt = "Down"
+    else:
+        beamalt = "Up"
+    st.write(
+        f"The current orientation of ADCP is `{st.session_state.beam_direction}`. Use the below option to correct the orientation."
+    )
+
+    beamdir_select = st.radio(f"Change orientation to {beamalt}", ["No", "Yes"])
+    if beamdir_select == "Yes":
+        st.session_state.beam_direction_QCT = beamalt
+        st.session_state.isBeamModified_QCT = True
+        st.write(f"The orientation changed to `{st.session_state.beam_direction_QCT}`")
+
+with tab5:
+    ################## Save Button #############
+    st.header("Save Data", divider="blue")
+    col1, col2 = st.columns([1, 1])
+    with col1:
+        save_mask_button = st.button(label="Save Mask Data", on_click=save_qctest)
+
+        if save_mask_button:
+            # st.session_state.qc_mask_temp = mask
+            st.success("Mask file saved")
+            # Table summarizing changes
+            changes_summary = pd.DataFrame(
+                [
+                    [
+                        "Quality Control Tests",
+                        "True" if st.session_state.isQCCheck_QCT else "False",
+                    ],
+                    ["Fix Orientation", st.session_state.beam_direction_QCT],
+                ],
+                columns=["Test", "Status"],
+            )
+
+            # Define a mapping function for styling
+            def status_color_map(value):
+                if value == "True":
+                    return "background-color: green; color: white"
+                elif value == "False":
+                    return "background-color: red; color: white"
+                elif value == "Up":
+                    return "background-color: blue; color: white"
+                elif value == "Down":
+                    return "background-color: orange; color: white"
+                else:
+                    return ""
+
+            # Apply styles using Styler.apply
+            styled_table = changes_summary.style.set_properties(
+                **{"text-align": "center"}
+            )
+            styled_table = styled_table.map(status_color_map, subset=["Status"])
+
+            # Display the styled table
+            st.write(styled_table.to_html(), unsafe_allow_html=True)
+        else:
+            st.warning("Mask data not saved")
+    with col2:
+        reset_mask_button = st.button("Reset mask Data", on_click=reset_qctest)
+        if reset_mask_button:
+            # st.session_state.qc_mask_temp = np.copy(st.session_state.orig_mask)
+            # st.session_state.isQCCheck_QCT = False
+            # st.session_state.isQCTest = False
+            # st.session_state.isGrid = False
+            # st.session_state.isProfileMask = False
+            # st.session_state.isVelocityMask = False
+            st.success("Mask data is reset to default")