Update README.md

README.md

@@ -35,18 +35,13 @@ RIA_Example/
 │   └── example_model.onnx               # Exported ONNX model (Screens / Application Packager input)
 │
 ├── applications/
-│   └── example_application.json         # Application Composer output (Application Packager)
+│   └── example_application.json         # Application Composer output, can be built in RIA Screens (Application Packager)
 │
-├── screens-apps/
+
-│   └── zone-fingerprinting/
-│       ├── manifest.json                 # Screens app manifest
-│       ├── zone_fingerprint.onnx         # ONNX model for this app
-│       └── zone-fingerprinting.tar.gz    # Packaged app (upload directly to Screens)
 │
 ├── workflows/
 │   ├── train.yaml                        # Example Model Trainer workflow (committed to .riahub/workflows/)
-│   ├── hpo.yaml                          # Example HPO workflow
+
-│   └── compression.yaml                 # Example Compression workflow
 │
 └── curator-configs/
     └── example_curator_config.json       # Example curation configuration for the Curator tool
@@ -237,38 +232,16 @@ The application JSON format is documented in `schemas/application/ria_applicatio
 | `native-x86` | Standard x86 Linux deployment |
 | `native-arm64` | ARM edge devices |
 | `nvidia-x86` | GPU-accelerated inference on x86 |
+| `RIA Screens` | Web based runtime environment to monitor app |
+
 
 ---
 
 ### Screens
 
-Screens deploys a packaged RF inference application to a live pipeline. You upload a `.tar.gz` app package, configure a data source (live SDR, file playback, or synthetic), and start the pipeline. Results stream back to the browser in real time.
+Screens deploys a packaged RF inference application to a live pipeline. You build an app from Application Composer, configure a data source (live SDR, file playback, or synthetic), and start the pipeline. Results stream back to the browser in real time.
 
-**Example files:** `screens-apps/zone-fingerprinting/zone-fingerprinting.tar.gz`
 
-#### Uploading and running the Zone Fingerprinting demo
-
-The Zone Fingerprinting app classifies RF devices in real time into five device classes (three authorized, two unauthorized) using an ONNX model and a 128-feature IQ preprocessor.
-
-**Steps:**
-1. Go to **Screens** and click **New App**.
-2. Give it a name (e.g. `Zone Fingerprinting Demo`) and click **Create**.
-3. On the app page, click **Upload Package** and upload `zone-fingerprinting.tar.gz`.
-4. The app is now configured. To run it with a synthetic signal (no hardware needed):
-   - The default `manifest.json` uses `dataSource.type: synthetic` — no changes required.
-5. Click **Start**. The inference pipeline starts and begins streaming results.
-6. The dashboard shows:
-   - Live classification scores per device class
-   - Confidence threshold control
-   - Spectrogram panel
-   - Preprocessor feature monitor
-   - Event log
-
-**To run with a real SDR (PlutoSDR):**
-1. Ensure your SDR device is connected and detected.
-2. Edit the app configuration and change `dataSource.type` to `sdr` with your device identifier.
-3. Set `center_frequency`, `sample_rate`, and `gain` to match your signal of interest.
-4. Click **Restart**.
 
 #### App package format
 
@@ -276,7 +249,6 @@ A Screens app package is a `.tar.gz` containing:
 - `manifest.json` — describes the app (models, GUI layout, data source, preprocessor)
 - ONNX model file(s) at the path(s) listed in `manifest.models[].path`
 
-See `screens-apps/zone-fingerprinting/manifest.json` for a complete annotated example. The full manifest schema is at `schemas/screens/app_manifest.schema.json`.
 
 **Data source types:**
 
@@ -325,41 +297,6 @@ dataset.h5
 └── attrs           # Dataset-level attributes: name, version, radio_task, backend
 ```
 
-### Screens App Manifest (`manifest.json`)
-
-```json
-{
-  "name": "My App",
-  "version": "1.0.0",
-  "description": "Brief description",
-  "author": "Your Name",
-  "models": [
-    {
-      "name": "classifier",
-      "path": "model.onnx",
-      "backend": "onnxrt-cpu"
-    }
-  ],
-  "preprocess": "magnitude_phase_window_stats",
-  "dataSource": {
-    "type": "synthetic",
-    "params": {
-      "sample_rate": 1000000,
-      "center_frequency": 915000000,
-      "buffer_size": 1024
-    }
-  },
-  "components": [
-    {
-      "component": "BackendInferenceOutput",
-      "props": { "modelName": "classifier", "topK": 3 }
-    }
-  ],
-  "config": {
-    "confidence_threshold": 0.75
-  }
-}
-```
 
 ### Application JSON (`application.json`)