Add styling and basic detectron functionality

app.py

@@ -1,173 +1,134 @@
-import detectron2
-import torch
-
-# Check this in logs
-try:
-    print(f"Is CUDA available: {torch.cuda.is_available()}")
-    # True
-    print(f"CUDA device: {torch.cuda.get_device_name(nn.cuda.current_device())}")
-except:
-    print('Couldnt find CUDA device')
-
-import faicons as fa
-import plotly.express as px
-
+# import detectron2
+# import torch
+#
+# # Check this in logs
+# try:
+#     print(f"Is CUDA available: {torch.cuda.is_available()}")
+#     # True
+#     print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
+# except:
+#     print('Couldnt find CUDA device')
+
+import base64
+import tempfile
+import cv2
+from io import BytesIO
+
+import pandas as pd
+from PIL import Image
+from shiny import App, ui, render, reactive, Session
+
+from python_utils import load_model
 # Load data and compute static values
-from shared import app_dir, tips
-from shinywidgets import render_plotly
-
-from shiny import reactive, render
-from shiny.express import input, ui
-
-bill_rng = (min(tips.total_bill), max(tips.total_bill))
+from shared import app_dir
+
+# Load the prediction model
+predictor = load_model()
+app_ui = ui.page_fluid(
+    ui.include_css("styles.css"),
+    ui.panel_title(ui.div("Orchid TZ Viability Analyzer", class_="navbar-title")),
+    ui.div(
+        ui.download_button("download", "Download Results", class_="btn-primary"),
+        style="position: absolute; top: 10px; right: 10px;"
+    ),
+    ui.layout_sidebar(
+        ui.sidebar(
+            ui.input_file("upload", "Upload Images",
+                         multiple=True,
+                         accept=[".png", ".jpg", ".jpeg"]),
+            ui.input_action_button("analyze", "Analyze", class_="btn-success"),
+            width =300
+        ),
+        ui.output_ui("results_container"),
+        border=False,
+        border_radius=False
+    )
+)
+
+
+
+def server(input, output, session: Session):
+    analysis_results = reactive.Value([])
+
+    @reactive.Effect
+    @reactive.event(input.analyze)
+    async def process_images():
+        files = input.upload()
+        if not files:
+            return
+
+        results = []
+        with tempfile.TemporaryDirectory() as temp_dir:
+            for idx, file in enumerate(files):
+                # Read image using OpenCV
+                im = cv2.imread(file["datapath"])
+
+                # Convert BGR to RGB for display
+                im_rgb = cv2.cvtColor(im, cv2.COLOR_BGR2RGB)
+                pil_img = Image.fromarray(im_rgb)
+
+                # Convert to base64 for HTML display
+                buffered = BytesIO()
+                pil_img.save(buffered, format="PNG")
+                img_base64 = base64.b64encode(buffered.getvalue()).decode()
+
+                # Run prediction with original BGR image
+                prediction = predictor(im)
+
+                results.append({
+                    "filename": file["name"],
+                    "image": img_base64,
+                    **prediction
+                })
+
+        # Update reactive value
+        analysis_results.set(results)
+
+    @output
+    @render.ui
+    def results_container():
+        results = analysis_results.get()
+        if not results:
+            return ui.div("No results yet. Upload images and click 'Analyze'.",
+                          class_="text-muted")
+
+        return ui.div(
+            [ui.div(
+                ui.row(
+                    ui.column(4, ui.img(src=f"data:image/png;base64,{r['image']}")),
+                    ui.column(4, ui.img(src=f"data:image/png;base64,{r['image']}")),
+                ),
+                ui.h5(r['filename'], style="margin-top: 15px;"),
+                ui.div(
+                    ui.span(f"Viable = {r.get('viable', '?')}"),
+                    ui.span(f"Nonviable = {r.get('nonviable', '?')}", style="margin: 0 15px;"),
+                    ui.span(f"Empty = {r.get('empty', '?')}"),
+                    class_="results-text"
+                ),
+                class_="card p-3"
+            ) for r in results]
+        )
+
+    @session.download()
+    def download():
+        results = analysis_results.get()
+        df = pd.DataFrame([{
+            "Filename": r["filename"],
+            "Viable": r.get("viable", ""),
+            "Nonviable": r.get("nonviable", ""),
+            "Empty": r.get("empty", "")
+        } for r in results])
+
+        # Create in-memory CSV file
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as tmp:
+            df.to_csv(tmp.name, index=False)
+            return tmp.name
+
+
+app = App(app_ui, server)
 
-# Add page title and sidebar
-ui.page_opts(title="Restaurant tipping", fillable=True)
 
-with ui.sidebar(open="desktop"):
-    ui.input_slider(
-        "total_bill",
-        "Bill amount",
-        min=bill_rng[0],
-        max=bill_rng[1],
-        value=bill_rng,
-        pre="$",
-    )
-    ui.input_checkbox_group(
-        "time",
-        "Food service",
-        ["Lunch", "Dinner"],
-        selected=["Lunch", "Dinner"],
-        inline=True,
-    )
-    ui.input_action_button("reset", "Reset filter")
-
-# Add main content
-ICONS = {
-    "user": fa.icon_svg("user", "regular"),
-    "wallet": fa.icon_svg("wallet"),
-    "currency-dollar": fa.icon_svg("dollar-sign"),
-    "ellipsis": fa.icon_svg("ellipsis"),
-}
-
-with ui.layout_columns(fill=False):
-    with ui.value_box(showcase=ICONS["user"]):
-        "Total tippers"
-
-        @render.express
-        def total_tippers():
-            tips_data().shape[0]
-
-    with ui.value_box(showcase=ICONS["wallet"]):
-        "Average tip"
-
-        @render.express
-        def average_tip():
-            d = tips_data()
-            if d.shape[0] > 0:
-                perc = d.tip / d.total_bill
-                f"{perc.mean():.1%}"
-
-    with ui.value_box(showcase=ICONS["currency-dollar"]):
-        "Average bill"
-
-        @render.express
-        def average_bill():
-            d = tips_data()
-            if d.shape[0] > 0:
-                bill = d.total_bill.mean()
-                f"${bill:.2f}"
-
-
-with ui.layout_columns(col_widths=[6, 6, 12]):
-    with ui.card(full_screen=True):
-        ui.card_header("Tips data")
-
-        @render.data_frame
-        def table():
-            return render.DataGrid(tips_data())
-
-    with ui.card(full_screen=True):
-        with ui.card_header(class_="d-flex justify-content-between align-items-center"):
-            "Total bill vs tip"
-            with ui.popover(title="Add a color variable", placement="top"):
-                ICONS["ellipsis"]
-                ui.input_radio_buttons(
-                    "scatter_color",
-                    None,
-                    ["none", "sex", "smoker", "day", "time"],
-                    inline=True,
-                )
-
-        @render_plotly
-        def scatterplot():
-            color = input.scatter_color()
-            return px.scatter(
-                tips_data(),
-                x="total_bill",
-                y="tip",
-                color=None if color == "none" else color,
-                trendline="lowess",
-            )
-
-    with ui.card(full_screen=True):
-        with ui.card_header(class_="d-flex justify-content-between align-items-center"):
-            "Tip percentages"
-            with ui.popover(title="Add a color variable"):
-                ICONS["ellipsis"]
-                ui.input_radio_buttons(
-                    "tip_perc_y",
-                    "Split by:",
-                    ["sex", "smoker", "day", "time"],
-                    selected="day",
-                    inline=True,
-                )
-
-        @render_plotly
-        def tip_perc():
-            from ridgeplot import ridgeplot
-
-            dat = tips_data()
-            dat["percent"] = dat.tip / dat.total_bill
-            yvar = input.tip_perc_y()
-            uvals = dat[yvar].unique()
-
-            samples = [[dat.percent[dat[yvar] == val]] for val in uvals]
-
-            plt = ridgeplot(
-                samples=samples,
-                labels=uvals,
-                bandwidth=0.01,
-                colorscale="viridis",
-                colormode="row-index",
-            )
-
-            plt.update_layout(
-                legend=dict(
-                    orientation="h", yanchor="bottom", y=1.02, xanchor="center", x=0.5
-                )
-            )
-
-            return plt
-
-
-ui.include_css(app_dir / "styles.css")
 
 # --------------------------------------------------------
 # Reactive calculations and effects
 # --------------------------------------------------------
-
-
-@reactive.calc
-def tips_data():
-    bill = input.total_bill()
-    idx1 = tips.total_bill.between(bill[0], bill[1])
-    idx2 = tips.time.isin(input.time())
-    return tips[idx1 & idx2]
-
-
-@reactive.effect
-@reactive.event(input.reset)
-def _():
-    ui.update_slider("total_bill", value=bill_rng)
-    ui.update_checkbox_group("time", selected=["Lunch", "Dinner"])

python_utils/__init__.py

@@ -0,0 +1 @@
+from .get_model import *

python_utils/get_model.py

@@ -0,0 +1,83 @@
+def get_set_up():
+    import torch
+    TORCH_VERSION = ".".join(torch.__version__.split(".")[:2])
+    CUDA_VERSION = torch.__version__.split("+")[-1]
+    print("torch: ", TORCH_VERSION, "; cuda: ", CUDA_VERSION)
+    print(f'GPU available: {torch.cuda.is_available()}')
+    print(torch.cuda.get_device_capability())
+
+    # print("detectron2:", detectron2.__version__)
+
+
+def load_model():
+    # def predictor(img):
+    #     return {}
+    # return predictor
+    # import some common detectron2 utilities
+    import torch
+    from detectron2 import model_zoo
+    from detectron2.engine import DefaultPredictor
+    from detectron2.config import get_cfg
+    from detectron2.data.datasets import register_coco_instances
+
+    import os
+    import numpy as np
+
+    ## define relevant parameters
+    cfg = get_cfg()
+    cfg.merge_from_file(
+        model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x.yaml"))
+    cfg.MODEL.ROI_HEADS.NUM_CLASSES = 4  # should be 3 after renaming 'Seed' class
+    if not torch.cuda.is_available():
+        cfg.MODEL.DEVICE = "cpu"
+    else:
+        cfg.MODEL.DEVICE = 'cuda'
+
+    register_coco_instances(
+        'seeds', {"thing_classes": ['Seed', 'Viable', 'Non-Viable', 'Empty'],
+                  "thing_colors": [(0, 0, 0), (0, 255, 0), (255, 0, 0), (0, 0, 255)]},
+        'dataset1/train/annotations_train.json', 'dataset1/train/')
+    cfg.DATASETS.TRAIN = ("seeds",)
+
+    mean = [0.5, 0.2, 0.1]
+    std = [0.5, 0.1, 0.1]  # mean_and_std("dataset1/Part1_COCO/images/train/")
+    cfg.MODEL.PIXEL_MEAN = np.array(mean, dtype=float).tolist()
+    cfg.MODEL.PIXEL_STD = np.array(std, dtype=float).tolist()
+
+    cfg.MODEL.ROI_HEADS.NAME = "CascadeROIHeads"
+    cfg.MODEL.ROI_BOX_HEAD.CLS_AGNOSTIC_BBOX_REG = True
+    cfg.MODEL.ROI_MASK_HEAD.CLS_AGNOSTIC_MASK = True
+    cfg.MODEL.ANCHOR_GENERATOR.SIZES = [[32, 64, 128, 256, 512, 1024]]
+    cfg.MODEL.ANCHOR_GENERATOR.ASPECT_RATIOS = [[0.125, 0.25, 0.5, 1.0, 2.0, 4.0, 8.0]]
+    cfg.MODEL.FPN.IN_FEATURES = ["res2", "res3", "res4", "res5"]
+    cfg.MODEL.RPN.IN_FEATURES = ["p2", "p3", "p4", "p5"]
+    cfg.MODEL.ROI_HEADS.IN_FEATURES = ["p2", "p3", "p4", "p5"]
+    cfg.MODEL.FPN.NORM = "GN"
+    cfg.MODEL.ROI_BOX_HEAD.NORM = "GN"
+    cfg.MODEL.ROI_MASK_HEAD.NORM = "GN"
+    cfg.MODEL.RESNETS.NORM = "GN"
+    cfg.SOLVER.CLIP_GRADIENTS.ENABLED = True
+    cfg.SOLVER.CLIP_GRADIENTS.CLIP_TYPE = "norm"
+
+    cfg.MODEL.RPN.NMS_THRESH = 0.3
+    cfg.MODEL.RPN.PRE_NMS_TOPK_TEST = 12000
+    cfg.MODEL.RPN.POST_NMS_TOPK_TEST = 8000
+    # threshold for confidence
+    cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.05
+    # removing overlapping bounding boxes of the same class
+    cfg.MODEL.ROI_HEADS.NMS_THRESH_TEST = 0.5
+    # max number of instances per image
+    cfg.TEST.DETECTIONS_PER_IMAGE = 1200
+
+    ## Load trained model
+    ## Local files
+    # cfg.OUTPUT_DIR = "../YOLO/outputs/output20"
+    # cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth")
+    ## Or hugging face model
+    cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_X_101_32x8d_FPN_3x.yaml")
+    predictor = DefaultPredictor(cfg)
+    return predictor
+
+if __name__ == '__main__':
+    # get_set_up()
+    load_model()

requirements.txt

@@ -1,9 +1,6 @@
-faicons
 shiny
 shinywidgets
-plotly
 pandas
-ridgeplot
 opencv-python-headless
 pyyaml==5.1
 torch

styles.css

@@ -1,12 +1,122 @@
-:root {
-  --bslib-sidebar-main-bg: #f8f8f8;
+/* www/styles.css */
+/* Modern sleek theme with dark mode elements */
+
+body {
+    background-color: #f8f9fa;
+    font-family: 'Segoe UI', system-ui, -apple-system, sans-serif;
+}
+
+.container-fluid {
+    padding: 20px;
+    max-width: 1400px;
+    margin: 0 auto;
+}
+
+/* Header styling */
+.navbar-title {
+    color: #2c3e50 !important;
+    font-weight: 700;
+    font-size: 1.8rem;
+    padding: 15px 0;
+}
+
+/* Sidebar styling */
+.card.shiny-input-container {
+    background-color: #ffffff;
+    border-radius: 12px;
+    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.05);
+    padding: 20px;
+    margin-bottom: 20px;
+}
+
+/* Upload button styling */
+.btn-file {
+    background-color: #4a90e2;
+    color: white !important;
+    border-radius: 8px;
+    padding: 10px 20px;
+    transition: all 0.3s ease;
+}
+
+.btn-file:hover {
+    background-color: #357abd;
+    transform: translateY(-1px);
+}
+
+/* Analyze button styling */
+.btn-success {
+    background-color: #27ae60 !important;
+    border: none;
+    border-radius: 8px;
+    padding: 12px 25px;
+    font-weight: 600;
+    transition: all 0.3s ease;
 }
 
-.popover {
-  --bs-popover-header-bg: #222;
-  --bs-popover-header-color: #fff;
+.btn-success:hover {
+    background-color: #219653 !important;
+    transform: translateY(-1px);
 }
 
-.popover .btn-close {
-  filter: var(--bs-btn-close-white-filter);
+/* Image cards styling */
+.card {
+    background: white;
+    border: none;
+    border-radius: 15px;
+    box-shadow: 0 4px 12px rgba(0, 0, 0, 0.08);
+    margin-bottom: 25px;
+    overflow: hidden;
+    transition: transform 0.2s ease;
+}
+
+.card:hover {
+    transform: translateY(-3px);
+}
+
+/* Image display styling */
+img {
+    border-radius: 10px;
+    object-fit: cover;
+    max-height: 300px;
+    width: 100%;
+    margin: 10px 0;
+    box-shadow: 0 2px 6px rgba(0, 0, 0, 0.1);
+}
+
+/* Results text styling */
+.results-text {
+    color: #2c3e50;
+    font-family: 'Courier New', monospace;
+    font-size: 1.1rem;
+    margin: 15px 0;
+    padding: 12px;
+    background-color: #f8f9fa;
+    border-radius: 6px;
+    border-left: 4px solid #4a90e2;
+}
+
+/* Download button styling */
+.btn-primary {
+    background-color: #2ecc71 !important;
+    border: none;
+    border-radius: 8px;
+    padding: 10px 25px;
+    font-weight: 600;
+    box-shadow: 0 2px 4px rgba(0, 0, 0, 0.1);
+}
+
+.btn-primary:hover {
+    background-color: #27ae60 !important;
+}
+
+/* Responsive design */
+@media (max-width: 768px) {
+    .col-md-4 {
+        flex: 0 0 100%;
+        max-width: 100%;
+    }
+
+    img {
+        max-height: 200px;
+    }
 }