[UPDATE] instructions in README -> MANAGED TO DEPLOY ON GCLOUD

cloud_analysis/README.md

@@ -4,26 +4,64 @@
 
 - If lost, there is a button `console` on the upper right corner that brings us back to the projects
 
+## Create the Docker image for Cloud analysis
+
+The Docker image used for the cloud analysis is slightly different:
+
+- The entrypoint uses `cloud_analysis/main.py`
+- Some libraries are added to the base `pyproject.toml` (i.e. `Flask`, `distlib`)
+- `snowmobile_cloud` also has a `.env` file containing the information for `Gmail` loggin
+
+Make sure you have the following `gmail_logs.env` in the folder `cloud_analysis`:
+
+```
+[email protected]
+GMAIL_PASS=XXX
+[email protected]
+```
+
+You may need to **Use an App Password** that you can set from your [Google Account](https://myaccount.google.com/) -> Security -> 2-step validation -> App Password (bottom of the page).
+
+Then run:
+
+```bash
+docker build -t snowmobile_cloud -f cloud_Dockerfile .
+```
+
+It is possible to test the image by sending a file to the endpoint:
+
+```
+docker run --rm -p 8080:8080 -v $PWD:/app snowmobile_cloud
+```
+
+And in another terminal:
+
+```bash
+cd cloud_analysis
+./test.sh
+```
+
+The sender should receive an email.
 
 ## Login stuff
 
 1- Generate the Google Cloud credentials:
 
-```
+```bash
 gcloud auth application-default login
 ```
 
 The credentials are stored in `~/.config/gcloud/`
 
 2- Login to your Google Cloud account:
 
-```
+```bash
 gcloud auth login
 ```
 
 3- Log in to the project
 
-```
+```bash
 gcloud config set project snokuter-akustikk
 ```
 
@@ -33,7 +71,7 @@ gcloud config set project snokuter-akustikk
 
 Enable the Cloud Functions, Cloud Run, and Cloud Build APIs:
 
-```
+```bash
 gcloud services enable cloudfunctions.googleapis.com
 gcloud services enable run.googleapis.com
 gcloud services enable cloudbuild.googleapis.com
@@ -43,22 +81,29 @@ gcloud services enable cloudbuild.googleapis.com
 
 The credentials are stored in `~/.docker/config.json`
 
-```
+```bash
 gcloud auth configure-docker
 ```
 
 3- Change the name of the image to push it on GCloud
 
 The Docker image will be saved in Google's container registry, which is directly supported by Cloud Run
 
-```
-docker tag ghcr.io/ninanor/snowmobile_analyzer:main gcr.io/snoskuter-akustikk/snowmobile_analyzer:main
-docker push gcr.io/snoskuter-akustikk/snowmobile_analyzer:main
+```bash
+docker tag snowmobile_cloud gcr.io/snoskuter-akustikk/snowmobile_cloud:main
+docker push gcr.io/snoskuter-akustikk/snowmobile_cloud:main
 
 ```
 
 4- Deploy the Docker image to Google Cloud run
 
+```bash
+gcloud run deploy \
+    model \
+    --image gcr.io/snoskuter-akustikk/snowmobile_cloud:main \
+    --memory 1Gi \
+    --region europe-north1 \
+    --platform managed
 ```
-gcloud run deploy model --image gcr.io/snoskuter-akustikk/snowmobile_analyzer:main
-```
+
+Be sure to allocate enough memory for the service to be able to process the files and run the model.

cloud_analysis/main.py

@@ -1,4 +1,3 @@
-
 #!/usr/env/bin python3
 
 from flask import Flask, request, jsonify
@@ -56,7 +55,6 @@ def send_email(subject, body):
         logging.error(f"Error sending email: {e}")
 
 
-
 def analyseAudioFile(
         audio_file_path, batch_size=1, num_workers=4, min_hr = 0.1, min_conf = 0.99
 ):
@@ -138,8 +136,8 @@ def process_audio_endpoint():
 
     detection_count = on_process_audio(audio_id, audio_rec, audio_file_path)
 
-    #if detection_count > 0:
-    send_email("Snowmobile Detection Alert", f"{detection_count} snowmobile detections were made in the audio file!")
+    if detection_count > 0:
+        send_email("Snowmobile Detection Alert", f"{detection_count} snowmobile detections were made in the audio file!")
 
     return jsonify({"message": "Audio processing completed!"})
 

cloud_analysis/test_cloud.sh

@@ -0,0 +1,10 @@
+ID_TOKEN=$(gcloud auth print-identity-token)
+
+curl -X POST \
+     -H "Content-Type: application/json" \
+     -H "Authorization: Bearer ${ID_TOKEN}" \
+     -d '{"audio_file_path": "/app/example/example_audio.mp3", "audio_id": "test-id", "audio_rec": {"location": {"latitude": 0, "longitude": 0}}}' \
+     https://model-4uhtnq5xla-lz.a.run.app/process-audio
+
+
+# /home/benjamin.cretois/data/snowmobile/example_audio.mp3

cloud_analysis/trigger_audio_analysis.py

@@ -0,0 +1,36 @@
+import requests
+import os
+from google.cloud import storage
+
+def trigger_audio_analysis(data, context):
+    """Background Cloud Function to be triggered by Cloud Storage.
+       This function is triggered when a file is uploaded to the GCS bucket.
+    """
+    file_name = data['name']
+    bucket_name = data['bucket']
+
+    # URL of your Cloud Run service
+    cloud_run_url = "https://model-4uhtnq5xla-lz.a.run.app/process-audio-gcs"
+
+    # ID Token for authentication
+    id_token = os.environ.get("ID_TOKEN")
+
+    # Payload to send to Cloud Run
+    payload = {
+        "bucket_name": bucket_name,
+        "blob_name": file_name,
+        "audio_id": "example-audio-id",
+        "audio_rec": {"location": {"latitude": 0, "longitude": 0}}
+    }
+
+    # Headers for the request
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {id_token}"
+    }
+
+    # Send a POST request to the Cloud Run service
+    response = requests.post(cloud_run_url, json=payload, headers=headers)
+
+    # Log the response
+    print(f"Response from Cloud Run: {response.text}")