did not do part 2. but must complete even if of no use.

Tooling/workload_generator/README.md

@@ -0,0 +1,54 @@
+# CSE546 workload_generator
+
+This repository contains code examples for you to use our workload generator.
+
+Usage:
+```
+usage: workload_generator.py [-h] [--num_request NUM_REQUEST] [--url URL] [--image_folder IMAGE_FOLDER] [--prediction_file PREDICTION_FILE]
+Upload images
+options:
+  -h, --help            show this help message and exit
+  --num_request NUM_REQUEST
+                        one image per request
+  --url URL             URL to the backend server, e.g. http://3.86.108.221:8000/
+  --image_folder IMAGE_FOLDER
+                        the path of the folder where images are saved
+  --prediction_file PREDICTION_FILE
+                        the path of the classification results file
+```
+
+Examples:
+
+The following command sends three requests to a php backend.
+```
+python3 workload_generator.py \
+ --num_request 3 \
+ --url 'http://your_host_ip/php_server.php' \
+ --image_folder "your_local_image_folder" \
+ --prediction_file "ground_truth_prediction_file"
+```
+
+The following command sends three requests to a node.js/Python backend.
+```
+python3 workload_generator.py \
+ --num_request 3 \
+ --url 'http://<server_ip>:<server_port>' \
+ --image_folder "your_local_image_folder" \
+ --prediction_file "ground_truth_prediction_file"
+```
+**Note**: The workload generator accepts the absolute path in the command line arguments. 
+
+Sample output:
+```
+test_411.jpg uploaded!
+Classification result: test_411:Emily
+Waiting ...
+.
++++++ Test Result Statistics +++++
+Total number of requests: 1000
+Total number of requests completed successfully: 1000
+Total number of failed requests: 0
+Total number of correct predictions: 1000
+Total number of wrong predictions: 0
+Total Test Duration: 2.104762077331543 (seconds)
+```

Tooling/workload_generator/workload_generator.py

@@ -0,0 +1,95 @@
+__copyright__   = "Copyright 2024, VISA Lab"
+__license__     = "MIT"
+
+import sys
+import os
+import time
+import _thread
+import argparse
+import requests
+import subprocess
+import numpy as np
+import pandas as pd
+import threading
+from concurrent.futures import ThreadPoolExecutor
+from concurrent.futures import as_completed
+
+parser = argparse.ArgumentParser(description='Upload images')
+parser.add_argument('--num_request', type=int, help='one image per request')
+parser.add_argument('--url', type=str, help='URL to the backend server, e.g. http://3.86.108.221:8000/')
+parser.add_argument('--image_folder', type=str, help='the path of the folder where images are saved')
+parser.add_argument('--prediction_file', type=str, help='the path of the classification results file')
+args = parser.parse_args()
+
+num_request     = args.num_request
+url             = args.url
+image_folder    = args.image_folder
+prediction_file = args.prediction_file
+prediction_df   = pd.read_csv(prediction_file)
+responses       = 0
+err_responses   = 0
+correct_predictions = 0
+wrong_predictions   = 0
+ex_requests         = []
+
+def send_one_request(image_path):
+    global prediction_df, responses, err_responses, correct_predictions, wrong_predictions, ex_requests
+    # Define http payload, "myfile" is the key of the http payload
+    file = {"inputFile": open(image_path,'rb')}
+    try:
+        response = requests.post(url, files=file)
+        # Print error message if failed
+        if response.status_code != 200:
+            print('sendErr: '+response.url)
+            err_responses +=1
+        else :
+            filename    = os.path.basename(image_path)
+            image_msg   = filename + ' uploaded!'
+            msg         = image_msg + '\n' + 'Classification result: ' + response.text
+            print(msg)
+            responses   +=1
+            correct_result = prediction_df.loc[prediction_df['Image'] == filename.split('.')[0], 'Results'].iloc[0]
+            if correct_result.strip() == response.text.split(':')[1].strip():
+                correct_predictions +=1
+            else:
+                wrong_predictions +=1
+    except requests.exceptions.RequestException as errex:
+        print("Exception:", errex)
+        ex_requests.append(image_path)
+
+num_max_workers = 100
+image_path_list = []
+test_start_time = time.time()
+
+for i, name in enumerate(os.listdir(image_folder)):
+    if i == num_request:
+        break
+    image_path_list.append(os.path.join(image_folder,name))
+
+with ThreadPoolExecutor(max_workers = num_max_workers) as executor:
+    executor.map(send_one_request, image_path_list)
+
+print(f"Attempt-1 {responses}/{num_request} requests successful.")
+
+# Retry requests until all requests are successful or ex_requests is empty
+retry_attempt=2
+while ex_requests:
+    retry_requests = ex_requests.copy()
+    ex_requests.clear()
+    with ThreadPoolExecutor(max_workers=num_max_workers) as executor:
+        executor.map(send_one_request, retry_requests)
+    print(f"Attempt-{retry_attempt} {responses}/{num_request} requests successful.")
+    retry_attempt += 1
+
+test_duration = time.time() - test_start_time
+print("All requests have been processed or retried.")
+
+if num_request == (responses + err_responses):
+   print (f"+++++ Test Result Statistics +++++")
+   print (f"Total number of requests: {num_request}")
+   print (f"Total number of requests completed successfully: {responses}")
+   print (f"Total number of failed requests: {err_responses}")
+   print (f"Total number of correct predictions : {correct_predictions}")
+   print (f"Total number of wrong predictions: {wrong_predictions}")
+   print (f"Total Test Duration: {test_duration} (seconds)")
+   print ("++++++++++++++++++++++++++++++++++++")