remove commented code and utilization details added

.gitignore

@@ -1,2 +1,3 @@
 flagged
 path
+__pycache__

app.py

@@ -1,89 +1,3 @@
-# import gradio as gr
-# from transformers import pipeline, AutoModelForImageSegmentation 
-# from gradio_imageslider import ImageSlider
-# import torch
-# from torchvision import transforms
-# import spaces
-# from PIL import Image
-
-# import numpy as np
-# import time
-
-# birefnet = AutoModelForImageSegmentation.from_pretrained(
-#     "ZhengPeng7/BiRefNet", trust_remote_code=True
-# )
-# device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-# print("Using device:", device)
-
-# birefnet.to(device)
-# transform_image = transforms.Compose(
-#     [
-#         transforms.Resize((1024, 1024)),
-#         transforms.ToTensor(),
-#         transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
-#     ]
-# )
-
-# # @spaces.GPU
-# # def PreProcess(image):
-# #     size = image.size
-# #     image = transform_image(image).unsqueeze(0).to(device)
-
-# #     with torch.no_grad():
-# #         preds = birefnet(image)[-1].sigmoid().cpu()
-# #     pred = preds[0].squeeze()
-# #     pred = transforms.ToPILImage()(pred)
-# #     mask = pred.resize(size)
-# #     # image.putalpha(mask)
-# #     return image
-
-# @spaces.GPU
-# def PreProcess(image):
-#     size = image.size  # Save original size
-#     image_tensor = transform_image(image).unsqueeze(0).to(device)  # Transform the image into a tensor
-
-#     with torch.no_grad():
-#         preds = birefnet(image_tensor)[-1].sigmoid().cpu()  # Get predictions
-#     pred = preds[0].squeeze()
-
-#     # Convert the prediction tensor to a PIL image
-#     pred_pil = transforms.ToPILImage()(pred)
-
-#     # Resize the mask to match the original image size
-#     mask = pred_pil.resize(size)
-
-#     # Convert the original image (passed as input) to a PIL image
-#     image_pil = image.convert("RGBA")  # Ensure the image has an alpha channel
-
-#     # Apply the alpha mask to the image
-#     image_pil.putalpha(mask)
-
-#     return image_pil
-
-# def segment_image(image):
-#     start = time.time()
-#     image = Image.fromarray(image)
-#     image = image.convert("RGB")
-#     org = image.copy()
-#     image = PreProcess(image)
-#     time_taken = np.round((time.time() - start),2)
-#     return (image, org), time_taken
-
-# slider = ImageSlider(label='birefnet', type="pil")
-# image = gr.Image(label="Upload an Image")
-
-# butterfly = Image.open("butterfly.png")
-# Dog = Image.open('Dog.jpg')
-
-# time_taken = gr.Textbox(label="Time taken", type="text")
-
-# demo = gr.Interface(
-#     segment_image, inputs=image, outputs=[slider,time_taken], examples=[butterfly,Dog], api_name="BiRefNet")
-
-# if __name__ == '__main__' :
-#     demo.launch()
-
 import requests
 import gradio as gr
 import tempfile
@@ -91,6 +5,7 @@ import os
 from transformers import pipeline
 from huggingface_hub import InferenceClient
 import time
+import psutil
 # import torch
 # import numpy as np
 
@@ -104,6 +19,9 @@ pipe = pipeline("automatic-speech-recognition", model=model_id) #, device=device
 def transcribe(inputs, use_api):
     start = time.time()
     API_STATUS = ''
+    
+    memory_before = psutil.Process(os.getpid()).memory_info().rss
+
     if inputs is None:
         raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
 
@@ -120,15 +38,24 @@ def transcribe(inputs, use_api):
             res = pipe(inputs, chunk_length_s=30)["text"]
         
         end = time.time() - start
-        return res, API_STATUS + str(round(end, 2)) + ' seconds'
-        # return res, end
+
+        # Measure memory after running the transcription process
+        memory_after = psutil.Process(os.getpid()).memory_info().rss
+        
+        # Calculate the difference to see how much memory was used by the code
+        memory_used = memory_after - memory_before  # Memory used in bytes
+        memory_used_gb = round(memory_used / (1024 ** 3), 2)  # Convert memory used to GB
+        total_memory_gb = round(psutil.virtual_memory().total / (1024 ** 3), 2)  # Total RAM in GB
+
+        # Calculate the percentage of RAM used by this process
+        memory_used_percent = round((memory_used / psutil.virtual_memory().total) * 100, 2)
+        
+        return res, API_STATUS + str(round(end, 2)) + ' seconds', f"RAM Used by code: {memory_used_gb} GB ({memory_used_percent}%) Total RAM: {total_memory_gb}"
+    
     
     except Exception as e:
         return fr'Error: {str(e)}', None
 
-def calculate_time_taken(start_time):
-    return time.time() - start_time
-
 demo = gr.Blocks()
 
 mf_transcribe = gr.Interface(
@@ -138,7 +65,10 @@ mf_transcribe = gr.Interface(
                     # gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
                     gr.Checkbox(label="Use API", value=False)
                 ],
-                outputs=["text",gr.Textbox(label="Time taken", type="text")],  # Placeholder for transcribed text and time taken
+                outputs=[gr.Textbox(label="Transcribed Text", type="text"),
+                         gr.Textbox(label="Time taken", type="text"),
+                         gr.Textbox(label="Utilization", type="text") 
+                        ],  # Placeholder for transcribed text and time taken
                 title="Welcome to QuickTranscribe",
                 description=(
                     "Transcribe long-form microphone or audio inputs with the click of a button!"
@@ -153,7 +83,10 @@ file_transcribe = gr.Interface(
                     # gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
                     gr.Checkbox(label="Use API", value=False)  # Checkbox for API usage
                 ],
-                outputs=["text",gr.Textbox(label="Time taken", type="text")],  # Placeholder for transcribed text and time taken
+                outputs=[ gr.Textbox(label="Transcribed Text", type="text"),
+                         gr.Textbox(label="Time taken", type="text"),
+                         gr.Textbox(label="Utilization", type="text") 
+                        ],  # Placeholder for transcribed text and time taken
                 title="Welcome to QuickTranscribe",
                 description=(
                     "Transcribe long-form microphone or audio inputs with the click of a button!"

requirements.txt

@@ -4,4 +4,5 @@ requests
 huggingface_hub
 pytest
 gradio
-ffmpeg
+ffmpeg
+psutil