sciveo 0.1.27__tar.gz → 0.1.29__tar.gz

{sciveo-0.1.27 → sciveo-0.1.29}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sciveo
-Version: 0.1.27
+Version: 0.1.29
 Description-Content-Type: text/markdown
 Provides-Extra: mon
 Provides-Extra: net

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/cli.py

@@ -21,7 +21,8 @@ def main():
   config = GlobalConfiguration.get()
 
   parser = argparse.ArgumentParser(description='sciveo CLI')
-  parser.add_argument('command', choices=['init', 'monitor', 'scan', 'media-server'], help='Command to execute')
+  parser.add_argument('command', choices=['init', 'monitor', 'scan', 'media-server', 'media-run'], help='Command to execute')
+
   parser.add_argument('--period', type=int, default=120, help='Period in seconds')
   parser.add_argument('--block', type=bool, default=True, help='Block flag')
   parser.add_argument('--auth', type=str, default=config['secret_access_key'], help='Auth secret access key')
@@ -29,6 +30,13 @@ def main():
   parser.add_argument('--net', type=str, default=None, help='Network like 192.168.10.0/24')
   parser.add_argument('--port', type=int, default=22, help='Host port number, used for network ops')
   parser.add_argument('--localhost', type=bool, default=False, help='Add localhost to list of hosts')
+  parser.add_argument('--input-path', type=str, help='Input Path')
+  parser.add_argument('--output-path', type=str, help='Output Path')
+  parser.add_argument('--width', type=str, default=None, help='width')
+  parser.add_argument('--height', type=str, default=None, help='height')
+  parser.add_argument('--rate', type=int, help='Rate number')
+  parser.add_argument('--processor', type=str, help='Processor name')
+
   args = parser.parse_args()
 
   if args.command == 'monitor':
@@ -55,6 +63,10 @@ def main():
   elif args.command == 'media-server':
     from sciveo.media.pipelines.server import __START_SCIVEO_MEDIA_SERVER__
     __START_SCIVEO_MEDIA_SERVER__()
+  elif args.command == 'media-run':
+    if args.processor == "audio-plot":
+      from sciveo.media.pipelines.processors.audio.audio_extractor_process import plot_audio
+      plot_audio(args.width, args.height, args.rate, args.input_path, args.output_path)
   else:
     warning(args.command, "not implemented")
 

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/media/pipelines/processors/audio/audio.py

@@ -50,11 +50,11 @@ class AudioExtract(BaseProcessor):
       exception(e, aac_audio_local_path)
 
   def plot_new_process(self, w, h, sampling_rate, aac_audio_local_path, image_audio_local_path):
-    cmd = "python pipelines/processors/audio/audio_extractor_process.py "
+    cmd = "sciveo media-run --processor audio-plot "
     cmd += f"--width {w} --height {h} "
-    cmd += f"--sampling_rate {sampling_rate} "
-    cmd += f"--aac_audio_local_path '{aac_audio_local_path}' "
-    cmd += f"--image_audio_local_path '{image_audio_local_path}' "
+    cmd += f"--rate {sampling_rate} "
+    cmd += f"--input-path '{aac_audio_local_path}' "
+    cmd += f"--output-path '{image_audio_local_path}' "
     cmd += " 1>audio.log 2>audio-error.log"
     debug("plot_new_process cmd", cmd)
     os.system(cmd)

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/media/pipelines/processors/audio/audio_extractor_process.py

@@ -9,15 +9,13 @@
 # 2024
 #
 
-import cv2
-import argparse
+import numpy as np
 from scipy.io import wavfile
 from scipy.signal import butter, lfilter
 import matplotlib.pyplot as plt
 
 from sciveo.tools.logger import *
 from sciveo.tools.common import *
-from sciveo.media.pipelines.processors.base import *
 
 
 class AudioFFT:
@@ -136,16 +134,3 @@ def plot_audio(width, height, sampling_rate, aac_audio_local_path, image_audio_l
     plt.savefig(image_audio_local_path, dpi=dpi, bbox_inches='tight', pad_inches=0)
   except Exception as e:
     exception(e, aac_audio_local_path)
-
-
-parser = argparse.ArgumentParser(description='Audio Extractor')
-
-parser.add_argument('--width', type=str, help='width')
-parser.add_argument('--height', type=str, help='height')
-parser.add_argument('--sampling_rate', type=str, help='aduio sampling rate')
-parser.add_argument('--aac_audio_local_path', type=str, help='aac_audio_local_path')
-parser.add_argument('--image_audio_local_path', type=str, help='image_audio_local_path')
-
-args = parser.parse_args()
-
-plot_audio(args.width, args.height, args.sampling_rate, args.aac_audio_local_path, args.image_audio_local_path)

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/media/pipelines/processors/image/album_in_image.py

@@ -11,6 +11,7 @@
 
 import cv2
 import math
+import numpy as np
 
 from sciveo.tools.logger import *
 from sciveo.tools.common import *

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/media/pipelines/processors/image/depth_esimation.py

@@ -13,6 +13,7 @@ import os
 import cv2
 import gc
 from PIL import Image
+import numpy as np
 
 import torch
 from transformers import AutoImageProcessor, AutoModelForDepthEstimation

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo/media/pipelines/processors/video/video_album.py

@@ -10,6 +10,7 @@
 #
 
 import cv2
+import numpy as np
 
 from sciveo.tools.common import *
 from sciveo.media.pipelines.processors.base import *

sciveo-0.1.29/sciveo/ml/evaluation/object_detection.py

@@ -0,0 +1,163 @@
+#
+# Pavlin Georgiev, Softel Labs
+#
+# This is a proprietary file and may not be copied,
+# distributed, or modified without express permission
+# from the owner. For licensing inquiries, please
+# contact pavlin@softel.bg.
+#
+# 2024
+#
+
+import time
+import math
+import numpy as np
+
+from sciveo.tools.logger import *
+from sciveo.ml.images.object_detection import *
+
+
+"""
+
+Object Detection Evaluation
+
+"""
+class EvalObjectDetection:
+  def __init__(self, predictions, labels, class_names):
+    self.predictions = predictions
+    self.labels = labels
+    self.class_names = class_names
+
+  # Convert from yolo predictions
+  def from_yolo(self):
+    self.converted_predictions = []
+    for predicted in predictions:
+      boxes = predicted.boxes
+      class_names = predicted.names
+
+      converted_prediction = {}
+      for i in range(len(boxes)):
+        class_id = int(boxes.cls[i].item())
+        label_class = class_names[class_id]
+        if label_class not in self.class_names:
+          continue
+        box = boxes.xyxyn[i].tolist()
+        confidence = boxes.conf[i].item()
+        box_confidence = box + [confidence]
+
+        converted_prediction.setdefault(label_class, [])
+        converted_prediction[label_class].append(box_confidence)
+      self.converted_predictions.append(converted_prediction)
+
+    # TODO: currently labels of type [ {"class 1": [ ['x1':x1,'y1':y1,'x2':x2,'y2':y2] ... ], "class 2": [ ['x1':x1,'y1':y1,'x2':x2,'y2':y2] ... ]} ... ]
+    self.converted_labels = []
+    for label in self.labels:
+      converted_label = {}
+      for class_name, class_boxes in label.items():
+        converted_label[class_name] = []
+        for box in class_boxes:
+          converted_label[class_name].append([box['x1'], box['y1'], box['x2'], box['y2']])
+      self.converted_labels.append(converted_label)
+
+  def compute_iou(self, box1, box2):
+    """
+    box1, box2: [x1, y1, x2, y2] format.
+    Returns the IoU between the two boxes.
+    """
+    x1_inter = max(box1[0], box2[0])
+    y1_inter = max(box1[1], box2[1])
+    x2_inter = min(box1[2], box2[2])
+    y2_inter = min(box1[3], box2[3])
+
+    inter_area = max(0, x2_inter - x1_inter) * max(0, y2_inter - y1_inter)
+    box1_area = (box1[2] - box1[0]) * (box1[3] - box1[1])
+    box2_area = (box2[2] - box2[0]) * (box2[3] - box2[1])
+    union_area = box1_area + box2_area - inter_area
+
+    if union_area == 0:
+      return 0.0
+
+    return inter_area / union_area
+
+  # Compute the Average Precision (AP)
+  def compute_ap(self, class_name, iou_threshold=0.0):
+    """
+    Calculate the Average Precision based on IoU and confidence scores.
+    Returns the AP for the current class.
+    """
+
+    count_labels = 0
+    true_positives = []
+    false_positives = []
+    false_negatives = []
+    detected = []
+
+    for i, prediction in enumerate(self.converted_predictions):
+      current_predictions = prediction.get(class_name, [])
+      current_labels = self.converted_labels[i].get(class_name, [])
+
+      count_labels += len(current_labels)
+      TP = 0
+
+      for prediction_box in current_predictions:
+        max_iou = 0
+        gt_match = None
+
+        for label_box in current_labels:
+          iou = self.compute_iou(prediction_box[:4], label_box)
+          if iou > 0.5 and iou >= max_iou:
+            max_iou = iou
+            gt_match = label_box
+
+        if max_iou >= iou_threshold and gt_match not in detected:
+          true_positives.append(1)
+          false_positives.append(0)
+          detected.append(gt_match)
+          TP += 1
+        else:
+          true_positives.append(0)
+          false_positives.append(1)
+
+      FN = len(current_labels) - TP
+      false_negatives.append(FN)
+
+    # Convert to cumulative sums
+    tp_cumsum = np.cumsum(true_positives)
+    fp_cumsum = np.cumsum(false_positives)
+
+    # Compute recall and precision
+    recall = tp_cumsum / count_labels
+    precision = tp_cumsum / (tp_cumsum + fp_cumsum)
+
+    # Compute AP using the trapezoidal rule (integrating precision over recall)
+    ap = np.trapz(precision, recall)
+    return ap, np.sum(false_negatives) / count_labels
+
+  def threshold(self, class_name, max_false_negative):
+    list_ap = []
+    list_fn = []
+    list_thresholds = np.linspace(0.0, 1.0, 101).tolist()
+    for i, iou_threshold in enumerate(list_thresholds):
+      ap, FN = self.compute_ap(class_name, iou_threshold)
+      #debug("AP", ap, "FN", FN)
+      if FN > max_false_negative:
+        ap = 0.0
+      list_ap.append(ap)
+      list_fn.append(FN)
+    idx = list_ap.index(max(list_ap))
+    debug("Threshold", class_name, list_thresholds[idx], "AP", list_ap[idx], "FN", list_fn[idx])
+    return list_thresholds[idx], list_ap[idx], list_fn[idx]
+
+  def evaluate(self, iou_thresholds={"default": 0.0}):
+    """
+    Calculate metrics like mAP based on IoU.
+    """
+    aps = []
+    class_ap = {}
+    for class_name in self.class_names:
+      ap, false_negatives = self.compute_ap(class_name, iou_thresholds.get(class_name, iou_thresholds["default"]))
+      aps.append(ap)
+      class_ap[class_name] = ap
+
+    mAP = np.mean(aps)
+    return {'mAP': mAP, 'AP per class': class_ap}

sciveo-0.1.29/sciveo/version.py

@@ -0,0 +1,2 @@
+
+__version__ = '0.1.29'

{sciveo-0.1.27 → sciveo-0.1.29}/sciveo.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sciveo
-Version: 0.1.27
+Version: 0.1.29
 Description-Content-Type: text/markdown
 Provides-Extra: mon
 Provides-Extra: net

sciveo-0.1.29/sciveo.egg-info/requires.txt

@@ -0,0 +1,79 @@
+numpy>=0.0.0
+requests>=0.0.0
+
+[all]
+psutil>=0.0.0
+netifaces>=0.0.0
+scapy>=0.0.0
+
+[media]
+scikit-learn>=0.0.0
+scipy>=0.0.0
+scikit-video>=0.0.0
+scikit-image>=0.0.0
+pycryptodome>=0.0.0
+exifread>=0.0.0
+qrcode[pil]>=0.0.0
+boto3
+pandas>=0.0.0
+pika>=0.0.0
+regex>=0.0.0
+matplotlib>=0.0.0
+joblib>=0.0.0
+tqdm>=0.0.0
+ffmpeg-python>=0.0.0
+opencv-python-headless>=0.0.0
+opencv-contrib-python-headless>=0.0.0
+
+[media-all]
+psutil>=0.0.0
+netifaces>=0.0.0
+scapy>=0.0.0
+scikit-learn>=0.0.0
+scipy>=0.0.0
+scikit-video>=0.0.0
+scikit-image>=0.0.0
+pycryptodome>=0.0.0
+exifread>=0.0.0
+qrcode[pil]>=0.0.0
+boto3
+pandas>=0.0.0
+pika>=0.0.0
+regex>=0.0.0
+matplotlib>=0.0.0
+joblib>=0.0.0
+tqdm>=0.0.0
+ffmpeg-python>=0.0.0
+opencv-python-headless>=0.0.0
+opencv-contrib-python-headless>=0.0.0
+fastapi
+uvicorn[standard]
+tensorflow>=0.0.0
+keras>=0.0.0
+torch>=0.0.0
+torchvision>=0.0.0
+diffusers>=0.0.0
+transformers>=0.0.0
+accelerate>=0.0.0
+annoy>=0.0.0
+
+[media-ml]
+tensorflow>=0.0.0
+keras>=0.0.0
+torch>=0.0.0
+torchvision>=0.0.0
+diffusers>=0.0.0
+transformers>=0.0.0
+accelerate>=0.0.0
+annoy>=0.0.0
+
+[media-server]
+fastapi
+uvicorn[standard]
+
+[mon]
+psutil>=0.0.0
+
+[net]
+netifaces>=0.0.0
+scapy>=0.0.0

{sciveo-0.1.27 → sciveo-0.1.29}/setup.py

@@ -21,13 +21,15 @@ extras_require = {
         'scapy>=0.0.0',
       ],
       'media': [
-        'scikit-learn', 'scipy', 'scikit-video', 'scikit-image', 'pycryptodome', 'exifread', 'qrcode[pil]',
-        'boto3', 'pandas', 'pika', 'regex', 'matplotlib', 'joblib', 'tqdm',
-        'ffmpeg-python', 'opencv-python-headless', 'opencv-contrib-python-headless',
+        'scikit-learn>=0.0.0', 'scipy>=0.0.0', 'scikit-video>=0.0.0', 'scikit-image>=0.0.0', 'pycryptodome>=0.0.0', 'exifread>=0.0.0', 'qrcode[pil]>=0.0.0',
+        'boto3', 'pandas>=0.0.0', 'pika>=0.0.0', 'regex>=0.0.0', 'matplotlib>=0.0.0', 'joblib>=0.0.0', 'tqdm>=0.0.0',
+        'ffmpeg-python>=0.0.0', 'opencv-python-headless>=0.0.0', 'opencv-contrib-python-headless>=0.0.0',
       ],
       'media-server': ['fastapi', 'uvicorn[standard]'],
       'media-ml': [
-        'tensorflow', 'keras', 'torch', 'torchvision', 'diffusers', 'transformers', 'accelerate', 'annoy',
+        'tensorflow>=0.0.0', 'keras>=0.0.0',
+        'torch>=0.0.0', 'torchvision>=0.0.0',
+        'diffusers>=0.0.0', 'transformers>=0.0.0', 'accelerate>=0.0.0', 'annoy>=0.0.0',
       ]
 }
 

sciveo-0.1.27/sciveo/ml/evaluation/object_detection.py

@@ -1,33 +0,0 @@
-#
-# Pavlin Georgiev, Softel Labs
-#
-# This is a proprietary file and may not be copied,
-# distributed, or modified without express permission
-# from the owner. For licensing inquiries, please
-# contact pavlin@softel.bg.
-#
-# 2024
-#
-
-import math
-import numpy as np
-
-from sciveo.ml.images.object_detection import *
-
-
-"""
-
-Object Detection Evaluation
-
-"""
-
-
-class EvaluateObjectDetection:
-  def __init__(self, X, Y_true, Y_predicted):
-    self.X = X
-    self.Y_true = Y_true
-    self.Y_predicted = Y_predicted
-
-  def evaluate(self):
-    pass
-

sciveo-0.1.27/sciveo/version.py

@@ -1,2 +0,0 @@
-
-__version__ = '0.1.27'

sciveo-0.1.27/sciveo.egg-info/requires.txt

@@ -1,79 +0,0 @@
-numpy>=0.0.0
-requests>=0.0.0
-
-[all]
-psutil>=0.0.0
-netifaces>=0.0.0
-scapy>=0.0.0
-
-[media]
-scikit-learn
-scipy
-scikit-video
-scikit-image
-pycryptodome
-exifread
-qrcode[pil]
-boto3
-pandas
-pika
-regex
-matplotlib
-joblib
-tqdm
-ffmpeg-python
-opencv-python-headless
-opencv-contrib-python-headless
-
-[media-all]
-psutil>=0.0.0
-netifaces>=0.0.0
-scapy>=0.0.0
-scikit-learn
-scipy
-scikit-video
-scikit-image
-pycryptodome
-exifread
-qrcode[pil]
-boto3
-pandas
-pika
-regex
-matplotlib
-joblib
-tqdm
-ffmpeg-python
-opencv-python-headless
-opencv-contrib-python-headless
-fastapi
-uvicorn[standard]
-tensorflow
-keras
-torch
-torchvision
-diffusers
-transformers
-accelerate
-annoy
-
-[media-ml]
-tensorflow
-keras
-torch
-torchvision
-diffusers
-transformers
-accelerate
-annoy
-
-[media-server]
-fastapi
-uvicorn[standard]
-
-[mon]
-psutil>=0.0.0
-
-[net]
-netifaces>=0.0.0
-scapy>=0.0.0