Update inference files

indrapaul824 · Mar 29, 2022 · 88bb805 · 88bb805
1 parent e283f47
commit 88bb805
Show file tree

Hide file tree

Showing 8 changed files with 212 additions and 14 deletions.
diff --git a/requirements/req.in b/requirements/req.in
@@ -12,4 +12,5 @@ lxml
 cython
 tensorflowjs
 imutils
-opencv-contrib-python
+opencv-contrib-python
+termcolor
diff --git a/requirements/req.txt b/requirements/req.txt
@@ -218,7 +218,9 @@ tensorflow-io-gcs-filesystem==0.24.0
 tensorflowjs==3.15.0
     # via -r requirements/req.in
 termcolor==1.1.0
-    # via tensorflow
+    # via
+    #   -r requirements/req.in
+    #   tensorflow
 tf-estimator-nightly==2.8.0.dev2021122109
     # via tensorflow
 threadpoolctl==3.1.0

diff --git a/vortex/hybridPred.py b/vortex/hybridPred.py
@@ -0,0 +1,121 @@
+# import the necessary packages
+from tracemalloc import start
+from tensorflow.keras.preprocessing.image import img_to_array
+from tensorflow.keras.preprocessing.image import load_img
+from tensorflow.keras.models import load_model
+import numpy as np
+import matplotlib.pyplot as plt
+import mimetypes
+import argparse
+import imutils
+import cv2
+import os
+
+os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
+
+# font
+font = cv2.FONT_HERSHEY_SIMPLEX
+# Blue color in BGR
+color = (0,0,255)
+# Line thickness of 2 px
+thickness = 3
+# org
+org = (50, 50)
+# fontScale
+fontScale = 1
+
+# construct the argument parser and parse the arguments
+ap = argparse.ArgumentParser()
+ap.add_argument("-i", "--input", required=True,
+	help="path to input image/text file of image filenames")
+args = vars(ap.parse_args())
+
+def get_key(my_dict, val):
+    for key, value in my_dict.items():
+         if val == value:
+             return key
+
+    return "Key doesn't exist"
+
+# determine the input file type, but assume that we're working with
+# single input image
+filetype = mimetypes.guess_type(args["input"])[0]
+imagePaths = [args["input"]]
+# if the file type is a text file, then we need to process *multiple*
+# images
+if "text/plain" == filetype:
+	# load the filenames in our testing file and initialize our list
+	# of image paths
+	filenames = open(args["input"]).read().strip().split("\n")
+	imagePaths = []
+	# loop over the filenames
+	for f in filenames:
+		# construct the full path to the image filename and then
+		# update our image paths list
+		p = os.path.sep.join(["../data/processed/Obj_Det/images", f])
+		imagePaths.append(p)
+
+# load our trained bounding box regressor from disk
+print("[INFO] loading object detector...")
+model = load_model("./artifacts/border_box/detector.h5")
+print("[INFO] loaded object detector...")
+
+print("-"*50)
+
+print("[INFO] loading intensity estimator model...")
+in_model = load_model('./artifacts/in_estimator/in_estimator.h5')
+print("[INFO] loaded intensity estimator model...")
+
+# loop over the images that we'll be testing using our bounding box
+# regression model
+p = 1
+for imagePath in imagePaths:
+    # load the input image (in Keras format) from disk and preprocess
+    # it, scaling the pixel intensities to the range [0, 1]
+    image = load_img(imagePath, target_size=(224, 224))
+    image = img_to_array(image) / 255.0
+    image = np.expand_dims(image, axis=0)
+    # make bounding box predictions on the input image
+    preds = model.predict(image)[0]
+    (startX, startY, endX, endY) = preds
+    # load the input image (in OpenCV format), resize it such that it
+	# fits on our screen, and grab its dimensions
+    image = cv2.imread(imagePath)
+    image = imutils.resize(image, width=600)
+    (h, w) = image.shape[:2]
+	# scale the predicted bounding box coordinates based on the image
+	# dimensions
+    startX = int(startX * w)
+    startY = int(startY * h)
+    endX = int(endX * w)
+    endY = int(endY * h)
+
+    #crop the roi
+    roi = image[startY:endY,startX:endX]
+    roi = cv2.resize(roi,(100,100))
+    roi = img_to_array(roi) / 255.0
+    roi = np.expand_dims(roi, axis=0)
+    #pred intensity
+    classes = {'EXTREME':0, 'NORMAL':1, 'SEVERE':2, 'SUPER':3, 'VERY':4}
+    Y_pred = in_model.predict(roi)
+    Y_pred_classes = np.argmax(Y_pred,axis = 1) 
+
+    # draw the predicted bounding box on the image
+    cv2.rectangle(image, (startX, startY), (endX, endY),(0, 255, 0), 2)
+    #print(imagePath)
+    txt = str(get_key(classes, Y_pred_classes[0]))
+    image2 = cv2.putText(image.copy(),txt, (startX-2,startY-4), font,fontScale, color, thickness, cv2.LINE_AA)
+
+    print("The predicted class for the given image is: ", get_key(classes, Y_pred_classes[0]))
+    while True:
+        cv2.imshow("Detection",image)
+        cv2.imshow("Prediction",image2)
+        if cv2.waitKey(1) & 0xFF == ord('q'):
+            break
+    cv2.destroyAllWindows()
+    # cv2.imwrite('./result2/'+str(p)+'.jpg',image)
+    # p = p+1
+    # print("pred image saved")
+
+
+
diff --git a/vortex/hybridPredict.ipynb b/vortex/hybridPredict.ipynb
@@ -0,0 +1,73 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Hybrid Prediction - Cyclone Detection and Intensity Estimation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[INFO] loading object detector...\n",
+      "2022-03-29 11:59:21.081233: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
+      "2022-03-29 11:59:21.081325: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: indra-pop-os\n",
+      "2022-03-29 11:59:21.081351: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: indra-pop-os\n",
+      "2022-03-29 11:59:21.081544: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.54.0\n",
+      "2022-03-29 11:59:21.081621: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 510.54.0\n",
+      "2022-03-29 11:59:21.081646: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 510.54.0\n",
+      "2022-03-29 11:59:21.082183: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
+      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "[INFO] loaded object detector...\n",
+      "--------------------------------------------------\n",
+      "[INFO] loading intensity estimator model...\n",
+      "[INFO] loaded intensity estimator model...\n",
+      "The predicted class for the given image is:  NORMAL\n"
+     ]
+    }
+   ],
+   "source": [
+    "!python hybridPred.py --input ../data/processed/threshZero_data/NORMAL/3DIMG_01DEC2020_0430_L1B_STD_IR2.jpg"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "interpreter": {
+   "hash": "92f8549edb95926e67d9af130c51c844e0526e8912c849c6708914c947287dc8"
+  },
+  "kernelspec": {
+   "display_name": "Python 3.9.7 ('vortex3.9')",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.11"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/vortex/notebooks/Intensity Classifier .ipynb b/vortex/notebooks/Intensity Classifier .ipynb
@@ -2767,23 +2767,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 27,
    "id": "198823a4",
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "2022-03-29 11:05:39.167162: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
-      "2022-03-29 11:05:39.167194: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: indra-pop-os\n",
-      "2022-03-29 11:05:39.167201: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: indra-pop-os\n",
-      "2022-03-29 11:05:39.167311: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.54.0\n",
-      "2022-03-29 11:05:39.167355: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 510.54.0\n",
-      "2022-03-29 11:05:39.167364: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 510.54.0\n",
-      "2022-03-29 11:05:39.167639: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
+      "2022-03-29 11:14:41.297217: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n",
+      "2022-03-29 11:14:41.297259: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: indra-pop-os\n",
+      "2022-03-29 11:14:41.297270: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: indra-pop-os\n",
+      "2022-03-29 11:14:41.297368: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.54.0\n",
+      "2022-03-29 11:14:41.297403: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 510.54.0\n",
+      "2022-03-29 11:14:41.297413: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:310] kernel version seems to match DSO: 510.54.0\n",
+      "2022-03-29 11:14:41.297702: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
       "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
-      "The predicted class for the given image is:  NORMAL\n"
+      "\u001b[32mThe predicted class for the given image is: \u001b[0m \u001b[31mNORMAL\u001b[0m\n"
      ]
     }
    ],

diff --git a/vortex/notebooks/__init__.py b/vortex/notebooks/__init__.py
diff --git a/vortex/notebooks/__pycache__/__init__.cpython-39.pyc b/vortex/notebooks/__pycache__/__init__.cpython-39.pyc
diff --git a/vortex/notebooks/in_estimator/predict.py b/vortex/notebooks/in_estimator/predict.py
@@ -1,9 +1,10 @@
 import tensorflow as tf
-from tensorflow.keras.preprocessing.image import img_to_array, load_img
-from tensorflow.keras.models import load_model
+from keras.preprocessing.image import img_to_array, load_img
+from keras.models import load_model
 import numpy as np
 import mimetypes
 import argparse
+from termcolor import colored
 import os
 
 os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
@@ -55,4 +56,4 @@ def get_key(my_dict, val):
     # Convert predictions classes to one hot vectors 
     Y_pred_classes = np.argmax(Y_pred,axis = 1) 
 
-    print("The predicted class for the given image is: ", get_key(classes, Y_pred_classes[0]))
+    print(colored("The predicted class for the given image is: ", "green"), colored(get_key(classes, Y_pred_classes[0]), "red"))