diff --git a/notebooks/Block_5/Jupyter Notebook Block 5 - Object Detection and Segmentation.ipynb b/notebooks/Block_5/Jupyter Notebook Block 5 - Object Detection and Segmentation.ipynb index 6c4199128c9b258b1be706818cfb5b69bb6f7ae4..632a2ee81cfaa5c0863d4c8c33a6ccbe62d4ed7a 100644 --- a/notebooks/Block_5/Jupyter Notebook Block 5 - Object Detection and Segmentation.ipynb +++ b/notebooks/Block_5/Jupyter Notebook Block 5 - Object Detection and Segmentation.ipynb @@ -143,31 +143,20 @@ "name": "stdout", "output_type": "stream", "text": [ - "Found 34 images belonging to 8 classes.\n" + "Found 0 images belonging to 8 classes.\n" ] }, { - "data": { - "image/png": "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\n", - "text/plain": [ - "<Figure size 432x288 with 1 Axes>" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - }, - { - "data": { - "text/plain": [ - "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 1., 1., 1., 1., 1., 1.,\n", - " 1., 1., 1., 2., 2., 3., 3., 4.], dtype=float32)" - ] - }, - "execution_count": 4, - "metadata": {}, - "output_type": "execute_result" + "ename": "ValueError", + "evalue": "Asked to retrieve element 0, but the Sequence has length 0", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-4-a6b4e6ab5ab7>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 24\u001b[0m batch_size=25, class_mode='sparse', shuffle=False)\n\u001b[1;32m 25\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 26\u001b[0;31m \u001b[0mplot_img\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdir_iter\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m...\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 27\u001b[0m \u001b[0mdir_iter\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/keras_preprocessing/image/iterator.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, idx)\u001b[0m\n\u001b[1;32m 55\u001b[0m \u001b[0;34m'but the Sequence '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 56\u001b[0m 'has length {length}'.format(idx=idx,\n\u001b[0;32m---> 57\u001b[0;31m length=len(self)))\n\u001b[0m\u001b[1;32m 58\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseed\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseed\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtotal_batches_seen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mValueError\u001b[0m: Asked to retrieve element 0, but the Sequence has length 0" + ] } ], "source": [ @@ -383,7 +372,7 @@ " ['...']\n", "Train for 24 steps, validate for 4 steps\n", "Epoch 1/20\n", - " 8/24 [=========>....................] - ETA: 8s - loss: 2.1694 - accuracy: 0.1250 " + " 2/24 [=>............................] - ETA: 16s - loss: 2.4319 - accuracy: 0.1000" ] }, { @@ -398,45 +387,45 @@ "name": "stdout", "output_type": "stream", "text": [ - "24/24 [==============================] - 11s 451ms/step - loss: 2.1161 - accuracy: 0.1333 - val_loss: 2.0744 - val_accuracy: 0.1875\n", + "24/24 [==============================] - 10s 398ms/step - loss: 2.1400 - accuracy: 0.1125 - val_loss: 2.0787 - val_accuracy: 0.2000\n", "Epoch 2/20\n", - "24/24 [==============================] - 8s 351ms/step - loss: 2.0759 - accuracy: 0.1500 - val_loss: 2.0721 - val_accuracy: 0.2000\n", + "24/24 [==============================] - 9s 363ms/step - loss: 2.0715 - accuracy: 0.1688 - val_loss: 2.0529 - val_accuracy: 0.2250\n", "Epoch 3/20\n", - "24/24 [==============================] - 8s 348ms/step - loss: 2.0754 - accuracy: 0.1437 - val_loss: 2.0638 - val_accuracy: 0.2000\n", + "24/24 [==============================] - 8s 347ms/step - loss: 2.0450 - accuracy: 0.1708 - val_loss: 2.0063 - val_accuracy: 0.2250\n", "Epoch 4/20\n", - "24/24 [==============================] - 8s 346ms/step - loss: 2.0343 - accuracy: 0.2146 - val_loss: 2.0168 - val_accuracy: 0.2000\n", + "24/24 [==============================] - 8s 353ms/step - loss: 1.9824 - accuracy: 0.2104 - val_loss: 1.9108 - val_accuracy: 0.2250\n", "Epoch 5/20\n", - "24/24 [==============================] - 8s 342ms/step - loss: 1.9712 - accuracy: 0.2146 - val_loss: 1.9564 - val_accuracy: 0.2250\n", + "24/24 [==============================] - 8s 342ms/step - loss: 1.9082 - accuracy: 0.2438 - val_loss: 1.8576 - val_accuracy: 0.3250\n", "Epoch 6/20\n", - "24/24 [==============================] - 8s 351ms/step - loss: 1.9165 - accuracy: 0.2521 - val_loss: 1.7011 - val_accuracy: 0.3500\n", + "24/24 [==============================] - 8s 347ms/step - loss: 1.8140 - accuracy: 0.3167 - val_loss: 1.7524 - val_accuracy: 0.3875\n", "Epoch 7/20\n", - "24/24 [==============================] - 8s 349ms/step - loss: 1.9125 - accuracy: 0.2208 - val_loss: 1.8084 - val_accuracy: 0.3125\n", + "24/24 [==============================] - 8s 346ms/step - loss: 1.7485 - accuracy: 0.3271 - val_loss: 1.6948 - val_accuracy: 0.3875\n", "Epoch 8/20\n", - "24/24 [==============================] - 8s 350ms/step - loss: 1.8092 - accuracy: 0.2708 - val_loss: 1.6824 - val_accuracy: 0.4000\n", + "24/24 [==============================] - 9s 355ms/step - loss: 1.6554 - accuracy: 0.3812 - val_loss: 1.5996 - val_accuracy: 0.4500\n", "Epoch 9/20\n", - "24/24 [==============================] - 8s 345ms/step - loss: 1.7669 - accuracy: 0.3083 - val_loss: 1.6866 - val_accuracy: 0.4000\n", + "24/24 [==============================] - 9s 355ms/step - loss: 1.5621 - accuracy: 0.4083 - val_loss: 1.5193 - val_accuracy: 0.5000\n", "Epoch 10/20\n", - "24/24 [==============================] - 8s 339ms/step - loss: 1.7030 - accuracy: 0.3292 - val_loss: 1.5380 - val_accuracy: 0.4625\n", + "24/24 [==============================] - 8s 353ms/step - loss: 1.5562 - accuracy: 0.3812 - val_loss: 1.5248 - val_accuracy: 0.4500\n", "Epoch 11/20\n", - "24/24 [==============================] - 8s 348ms/step - loss: 1.6934 - accuracy: 0.3396 - val_loss: 1.4929 - val_accuracy: 0.4125\n", + "24/24 [==============================] - 8s 335ms/step - loss: 1.5109 - accuracy: 0.4125 - val_loss: 1.5640 - val_accuracy: 0.4375\n", "Epoch 12/20\n", - "24/24 [==============================] - 9s 358ms/step - loss: 1.6249 - accuracy: 0.3771 - val_loss: 1.5017 - val_accuracy: 0.4125\n", + "24/24 [==============================] - 8s 340ms/step - loss: 1.4337 - accuracy: 0.4583 - val_loss: 1.5276 - val_accuracy: 0.4625\n", "Epoch 13/20\n", - "24/24 [==============================] - 8s 350ms/step - loss: 1.5701 - accuracy: 0.4021 - val_loss: 1.4963 - val_accuracy: 0.4750\n", + "24/24 [==============================] - 8s 353ms/step - loss: 1.3907 - accuracy: 0.4875 - val_loss: 1.5706 - val_accuracy: 0.4125\n", "Epoch 14/20\n", - "24/24 [==============================] - 8s 346ms/step - loss: 1.5946 - accuracy: 0.3875 - val_loss: 1.4775 - val_accuracy: 0.5125\n", + "24/24 [==============================] - 8s 330ms/step - loss: 1.3740 - accuracy: 0.4563 - val_loss: 1.5009 - val_accuracy: 0.4500\n", "Epoch 15/20\n", - "24/24 [==============================] - 8s 343ms/step - loss: 1.4456 - accuracy: 0.4292 - val_loss: 1.4721 - val_accuracy: 0.4500\n", + "24/24 [==============================] - 8s 350ms/step - loss: 1.3238 - accuracy: 0.4979 - val_loss: 1.4842 - val_accuracy: 0.4250\n", "Epoch 16/20\n", - "24/24 [==============================] - 8s 339ms/step - loss: 1.4155 - accuracy: 0.4437 - val_loss: 1.4771 - val_accuracy: 0.5125\n", + "24/24 [==============================] - 8s 339ms/step - loss: 1.3219 - accuracy: 0.4917 - val_loss: 1.4689 - val_accuracy: 0.4750\n", "Epoch 17/20\n", - "24/24 [==============================] - 8s 350ms/step - loss: 1.4227 - accuracy: 0.4500 - val_loss: 1.4232 - val_accuracy: 0.5125\n", + "24/24 [==============================] - 8s 336ms/step - loss: 1.2339 - accuracy: 0.5188 - val_loss: 1.4910 - val_accuracy: 0.5000\n", "Epoch 18/20\n", - "24/24 [==============================] - 9s 355ms/step - loss: 1.3675 - accuracy: 0.4833 - val_loss: 1.5720 - val_accuracy: 0.4500\n", + "24/24 [==============================] - 8s 343ms/step - loss: 1.3074 - accuracy: 0.5000 - val_loss: 1.4635 - val_accuracy: 0.4625\n", "Epoch 19/20\n", - "24/24 [==============================] - 9s 355ms/step - loss: 1.3681 - accuracy: 0.4812 - val_loss: 1.5468 - val_accuracy: 0.4750\n", + "24/24 [==============================] - 8s 349ms/step - loss: 1.2370 - accuracy: 0.5167 - val_loss: 1.4229 - val_accuracy: 0.4625\n", "Epoch 20/20\n", - "24/24 [==============================] - 8s 341ms/step - loss: 1.2912 - accuracy: 0.4917 - val_loss: 1.5665 - val_accuracy: 0.4875\n" + "24/24 [==============================] - 8s 340ms/step - loss: 1.1461 - accuracy: 0.5521 - val_loss: 1.4321 - val_accuracy: 0.4875\n" ] } ], @@ -3614,21 +3603,17 @@ "id": "9zKyjQ7nGhN4" }, "source": [ - "## Suggestions for Your Project in DLV\n", + "# Suggestions for Your Project in DLV\n", "\n", "1. Get familiar with different ConvNet architectures such as _EfficientNets_, _MobileNet_, etc. and apply transfer learning to your own dataset. Discuss the resulting confusion matrices and record test set accuracies, F1-scores, etc.\n", "\n", - "2. Get acquainted with the [Coconut Annotator](https://github.com/jsbroks/coco-annotator) or [labelImg](https://github.com/tzutalin/labelImg): download your own \n", - "image dataset and annotate images. Use Transfer Learning for classification.\n", - "\n", - "3. Discover which parts of an image are relevant for image classification. Apply GradCam and get familiar with [Layer-Wise Relevance Propagation](https://towardsdatascience.com/indepth-layer-wise-relevance-propagation-340f95deb1ea). Use LRP with Keras (https://pypi.org/project/keras-explain/) to your image classification task.\n", + "2. Scrape your own image dataset and label objects in your images by means of e.g. [labelImg](https://github.com/tzutalin/labelImg). Use [YOLO](https://github.com/Ma-Dan/keras-yolo4) and [SSD](https://github.com/pierluigiferrari/ssd_keras) to detect objects in your dataset. Compare your results with respect to speed and Intersection of Union (IoU) or Mean Average Precision (MAP) (see lecture notes).\n", "\n", - "4. Use [YOLO](https://github.com/Ma-Dan/keras-yolo4) to detect objects in your dataset.\n", + "3. Get acquainted with the [Coconut Annotator](https://github.com/jsbroks/coco-annotator) to annotate and segment objects in your images. Use Transfer Learning for object detection and classification. See [Mask RCNN for Object Detection and Segmentation](https://github.com/matterport/Mask_RCNN)\n", "\n", - "5. Get acquainted with the [Coconut Annotator](https://github.com/jsbroks/coco-annotator) or [labelImg](https://github.com/tzutalin/labelImg): download your own \n", - "image dataset and segment objects in images. Use Mask R-CNN to detect and segment objects in validation dataset.\n", + "4. Discover which parts of an image are relevant for image classification. Apply GradCam and get familiar with [Layer-Wise Relevance Propagation](https://towardsdatascience.com/indepth-layer-wise-relevance-propagation-340f95deb1ea). Use LRP with Keras (https://pypi.org/project/keras-explain/) to your image classification task.\n", "\n", - "6. Compare speed and accuracy of object detection by Mask R-CNN and Yolo. Consider as well testing [SSD](https://github.com/pierluigiferrari/ssd_keras) " + "5. Label joints of animals in your images by means of [DeepLabCut](http://www.mackenziemathislab.org/deeplabcut). Classify animals or poses of animals by means of (relative) joint coordinates. See as well [Real Time Pose Estimation](https://github.com/michalfaber/keras_Realtime_Multi-Person_Pose_Estimation)" ] } ],