Skip to content

Instantly share code, notes, and snippets.

@idanori

idanori/my_stateFarm.ipynb

Last active November 9, 2017 12:03
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save idanori/3dadb3c1d66223170d2df4cd802da7cf to your computer and use it in GitHub Desktop.
Save idanori/3dadb3c1d66223170d2df4cd802da7cf to your computer and use it in GitHub Desktop.
Download ZIP
Score: 0.715 - State Farm Distracted Driver Detection
Display the source blob
Display the rendered blob
Raw
my_stateFarm.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Using gpu device 0: Tesla K80 (CNMeM is disabled, cuDNN 5103)\n",
"/home/ubuntu/anaconda2/lib/python2.7/site-packages/theano/sandbox/cuda/__init__.py:600: UserWarning: Your cuDNN version is more recent than the one Theano officially supports. If you see any problems, try updating Theano or downgrading cuDNN to version 5.\n",
" warnings.warn(warn)\n"
]
}
],
"source": [
"from theano.sandbox import cuda\n",
"cuda.use('gpu')"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Using Theano backend.\n"
]
}
],
"source": [
"%matplotlib inline\n",
"from __future__ import print_function, division\n",
"\n",
"import utils\n",
"from utils import *\n",
"from IPython.display import FileLink\n",
"from shutil import copyfile, move\n",
"import pandas as pd\n",
"import matplotlib.image as mpimg"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"path = \"/home/ubuntu/data/stateFarm/\"\n",
"#path = \"/home/ubuntu/data/stateFarm/sample/\"\n",
"homeDataPath = '/home/ubuntu/data/stateFarm/'"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"batch_size=64"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Save Array function"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import bcolz\n",
"def save_array(fname, arr): c=bcolz.carray(arr, rootdir=fname, mode='w'); c.flush()\n",
"def load_array(fname): return bcolz.open(fname)[:]\n",
"def load_carray(fname): return bcolz.open(fname)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def onehot(x): return np.array(OneHotEncoder().fit_transform(x.reshape(-1,1)).todense())"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def hotone(x):\n",
" ret = []\n",
" for i in range(len(x)):\n",
" for j in range(len(x[i])): \n",
" if x[i][j] == 1:\n",
" ret += [j]\n",
" return ret"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#cocate all array from batch dictonary\n",
"def concat_batches(batches):\n",
" return np.concatenate([batches.next() for i in range(int(batches.N/batches.batch_size+1))])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Creating a Validation set from Train set, for state farm comp"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"/home/ubuntu/data/stateFarm/train\n"
]
}
],
"source": [
"%cd /home/ubuntu/data/stateFarm/train/"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"rm: cannot remove '../valid': No such file or directory\r\n"
]
}
],
"source": [
"%rm -r ../valid #delete old"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"%mkdir ../valid"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"for i in glob('c?'):\n",
" os.mkdir('../valid/'+i)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"dil = pd.read_csv(path + 'driver_imgs_list.csv')"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['p051' 'p049' 'p035' 'p050' 'p022']\n",
"['c7', 'c8', 'c6', 'c0', 'c2', 'c4', 'c9', 'c1', 'c3', 'c5']\n"
]
}
],
"source": [
"# get all different drivers from file and select 5 randomely\n",
"# group by subject\n",
"allClasses=glob('c?')\n",
"allDrivers = dil['subject'].drop_duplicates().as_matrix()\n",
"randValidationDrivers = np.random.choice(allDrivers, 5)\n",
"print(randValidationDrivers)\n",
"print(allClasses)"
]
},
{
"cell_type": "code",
"execution_count": 198,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"#allPicOfValidationDriver.iloc[0].img #finally mather fucker!"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"totalt pic moved: 4802\n"
]
}
],
"source": [
"totalPicMoved=0\n",
"\n",
"for driveClass in allClasses:\n",
" perClassSeries = dil['classname'].isin([driveClass])\n",
" picPerClass = dil[perClassSeries]\n",
"\n",
" for driver in randValidationDrivers:\n",
" perDriverPic = picPerClass['subject'].isin([driver])\n",
" picPerDriverInClass = picPerClass[perDriverPic]\n",
" picArray = picPerDriverInClass['img'].as_matrix()\n",
" #select 20% random photos and move it\n",
" #numOfValidPic = int(len(picArray)*0.2)\n",
" \n",
" #print 'num of valid pic ' + str(numOfValidPic) + '/' + str(len(picArray))\n",
" #print ('num of valid pic '+str(numOfValidPic)+ '/'+ str(len(picArray)))\n",
" \n",
" #validationPicArr=np.random.choice(picArray,numOfValidPic)\n",
" \n",
" shuf = np.random.permutation(picArray)\n",
" for pic in shuf: move(driveClass +'/'+ pic, '../valid/'+ driveClass +'/'+pic)\n",
" \n",
" totalPicMoved += len(shuf)\n",
" \n",
"print ('totalt pic moved: '+ str(totalPicMoved))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Create sample set"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"/home/ubuntu/data/stateFarm/train\n"
]
}
],
"source": [
"%cd /home/ubuntu/data/stateFarm/train"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"%mkdir ../sample\n",
"%mkdir ../sample/train\n",
"%mkdir ../sample/valid"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"for d in glob('c?'):\n",
" os.mkdir('../sample/train/'+d)\n",
" os.mkdir('../sample/valid/'+d)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"g = glob('c?/*.jpg')\n",
"shuf = np.random.permutation(g)\n",
"for i in range(1500): copyfile(shuf[i], '../sample/train/' + shuf[i])"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"/home/ubuntu/data/stateFarm/valid\n"
]
}
],
"source": [
"%cd /home/ubuntu/data/stateFarm/valid"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"g = glob('c?/*.jpg')\n",
"shuf = np.random.permutation(g)\n",
"for i in range(1000): copyfile(shuf[i], '../sample/valid/' + shuf[i])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## create result dir"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"/home/ubuntu/data/stateFarm\n"
]
}
],
"source": [
"%cd /home/ubuntu/data/stateFarm"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"%mkdir results\n",
"%mkdir sample/test"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## create batches"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 1500 images belonging to 10 classes.\n",
"Found 1000 images belonging to 10 classes.\n"
]
}
],
"source": [
"batches = get_batches(path+'train', batch_size=batch_size, target_size=(224,224))\n",
"val_batches = get_batches(path+'valid', batch_size=batch_size, shuffle=False, target_size=(224,224))"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(64, 3, 224, 224)\n",
"(64, 10)\n"
]
}
],
"source": [
"imgsBatch, labelBatch = val_batches.next()\n",
"print(imgsBatch.shape)\n",
"print(labelBatch.shape)\n"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false,
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 1500 images belonging to 10 classes.\n",
"Found 1000 images belonging to 10 classes.\n",
"Found 0 images belonging to 0 classes.\n"
]
}
],
"source": [
"(val_classes, trn_classes, val_labels, trn_labels, val_filenames, filenames,\n",
" test_filename) = get_classes(path)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Add conv layer"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### My banch-mark: I reached over 50% with out pre-trained model, no dropout, and only a sample set. "
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"path = \"/home/ubuntu/data/stateFarm/sample/\""
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 1500 images belonging to 10 classes.\n"
]
}
],
"source": [
"#generate a new train batch\n",
"imgGen = image.ImageDataGenerator(rotation_range=15, height_shift_range=0.05, \n",
" shear_range=0.1, channel_shift_range=20, width_shift_range=0.1)\n",
"\n",
"augmentBatch = imgGen.flow_from_directory(directory=path+'train', batch_size=64, target_size=(224,224), class_mode='categorical')"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 1000 images belonging to 10 classes.\n"
]
}
],
"source": [
"val_batches = get_batches(path+'valid', batch_size=batch_size, shuffle=False, target_size=(224,224))"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def convModelTrain (lr1=0.01):\n",
" baseModel = Sequential()\n",
" baseModel.add(BatchNormalization(axis=1, input_shape=(3,224,224)))\n",
" baseModel.add(Convolution2D(32, 3, 3, activation='relu'))\n",
" baseModel.add(BatchNormalization(axis=1))\n",
" baseModel.add(MaxPooling2D(pool_size=(2,2)))\n",
" baseModel.add(Convolution2D(64, 3, 3, activation='relu'))\n",
" baseModel.add(BatchNormalization(axis=1))\n",
" baseModel.add(MaxPooling2D(pool_size=(2,2)))\n",
" baseModel.add(Flatten())\n",
" baseModel.add(Dense(200,activation='relu'))\n",
" baseModel.add(BatchNormalization())\n",
" baseModel.add(Dense(10,activation='softmax'))\n",
" \n",
" #start low learning rate\n",
" baseModel.compile(Adam(lr=lr1), \n",
" loss='categorical_crossentropy',\n",
" metrics=['accuracy'])\n",
" \n",
" return baseModel"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"model = convModelTrain()"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"1500/1500 [==============================] - 37s - loss: 2.6393 - acc: 0.1867 - val_loss: 14.8380 - val_acc: 0.0650\n",
"Epoch 2/10\n",
"1500/1500 [==============================] - 30s - loss: 1.9383 - acc: 0.3073 - val_loss: 8.1190 - val_acc: 0.1480\n",
"Epoch 3/10\n",
"1500/1500 [==============================] - 30s - loss: 1.7712 - acc: 0.3733 - val_loss: 5.4116 - val_acc: 0.2380\n",
"Epoch 4/10\n",
"1500/1500 [==============================] - 30s - loss: 1.7091 - acc: 0.4100 - val_loss: 5.2409 - val_acc: 0.1920\n",
"Epoch 5/10\n",
"1500/1500 [==============================] - 30s - loss: 1.4909 - acc: 0.4793 - val_loss: 3.6305 - val_acc: 0.2760\n",
"Epoch 6/10\n",
"1500/1500 [==============================] - 30s - loss: 1.3505 - acc: 0.5220 - val_loss: 4.0475 - val_acc: 0.2320\n",
"Epoch 7/10\n",
"1500/1500 [==============================] - 30s - loss: 1.2873 - acc: 0.5353 - val_loss: 3.1316 - val_acc: 0.3000\n",
"Epoch 8/10\n",
"1500/1500 [==============================] - 30s - loss: 1.1416 - acc: 0.5980 - val_loss: 3.2008 - val_acc: 0.3500\n",
"Epoch 9/10\n",
"1500/1500 [==============================] - 30s - loss: 1.1070 - acc: 0.6267 - val_loss: 2.5659 - val_acc: 0.3600\n",
"Epoch 10/10\n",
"1500/1500 [==============================] - 30s - loss: 1.0664 - acc: 0.6333 - val_loss: 2.5508 - val_acc: 0.3610\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7fc0ee77bc90>"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.fit_generator(augmentBatch, augmentBatch.nb_sample, nb_epoch=10, verbose=1, \n",
" validation_data=val_batches, nb_val_samples=val_batches.nb_sample)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### stable around 0.35? reduce lr - divide by 10, and train again (still alot of over fitting)"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"model.optimizer.lr=0.001"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"1500/1500 [==============================] - 33s - loss: 0.9739 - acc: 0.6620 - val_loss: 3.5652 - val_acc: 0.2970\n",
"Epoch 2/10\n",
"1500/1500 [==============================] - 29s - loss: 0.8777 - acc: 0.7020 - val_loss: 3.2465 - val_acc: 0.3100\n",
"Epoch 3/10\n",
"1500/1500 [==============================] - 30s - loss: 0.8432 - acc: 0.7187 - val_loss: 2.6794 - val_acc: 0.3400\n",
"Epoch 4/10\n",
"1500/1500 [==============================] - 30s - loss: 0.7952 - acc: 0.7307 - val_loss: 2.5238 - val_acc: 0.4110\n",
"Epoch 5/10\n",
"1500/1500 [==============================] - 30s - loss: 0.7085 - acc: 0.7553 - val_loss: 2.3278 - val_acc: 0.4510\n",
"Epoch 6/10\n",
"1500/1500 [==============================] - 30s - loss: 0.6804 - acc: 0.7560 - val_loss: 4.8475 - val_acc: 0.2900\n",
"Epoch 7/10\n",
"1500/1500 [==============================] - 29s - loss: 0.7576 - acc: 0.7433 - val_loss: 4.5103 - val_acc: 0.2750\n",
"Epoch 8/10\n",
"1500/1500 [==============================] - 30s - loss: 0.6566 - acc: 0.7747 - val_loss: 3.9851 - val_acc: 0.3890\n",
"Epoch 9/10\n",
"1500/1500 [==============================] - 29s - loss: 0.6108 - acc: 0.7887 - val_loss: 3.5767 - val_acc: 0.4150\n",
"Epoch 10/10\n",
"1500/1500 [==============================] - 29s - loss: 0.6328 - acc: 0.7893 - val_loss: 2.8997 - val_acc: 0.5060\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7fc106eb4610>"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.fit_generator(augmentBatch, augmentBatch.nb_sample, nb_epoch=10, verbose=1, \n",
" validation_data=val_batches, nb_val_samples=val_batches.nb_sample)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"1500/1500 [==============================] - 33s - loss: 0.5757 - acc: 0.8020 - val_loss: 2.6913 - val_acc: 0.4940\n",
"Epoch 2/10\n",
"1500/1500 [==============================] - 29s - loss: 0.5323 - acc: 0.8173 - val_loss: 3.5743 - val_acc: 0.4380\n",
"Epoch 3/10\n",
"1500/1500 [==============================] - 30s - loss: 0.4798 - acc: 0.8293 - val_loss: 4.5922 - val_acc: 0.4530\n",
"Epoch 4/10\n",
"1500/1500 [==============================] - 30s - loss: 0.4664 - acc: 0.8393 - val_loss: 2.0947 - val_acc: 0.4830\n",
"Epoch 5/10\n",
"1500/1500 [==============================] - 30s - loss: 0.4332 - acc: 0.8547 - val_loss: 2.4715 - val_acc: 0.5310\n",
"Epoch 6/10\n",
"1500/1500 [==============================] - 30s - loss: 0.4017 - acc: 0.8693 - val_loss: 3.1665 - val_acc: 0.4410\n",
"Epoch 7/10\n",
"1500/1500 [==============================] - 30s - loss: 0.3888 - acc: 0.8687 - val_loss: 2.4856 - val_acc: 0.5430\n",
"Epoch 8/10\n",
"1500/1500 [==============================] - 30s - loss: 0.3996 - acc: 0.8633 - val_loss: 2.1204 - val_acc: 0.5150\n",
"Epoch 9/10\n",
"1500/1500 [==============================] - 30s - loss: 0.4210 - acc: 0.8520 - val_loss: 2.7792 - val_acc: 0.4900\n",
"Epoch 10/10\n",
"1500/1500 [==============================] - 30s - loss: 0.3383 - acc: 0.8880 - val_loss: 1.9513 - val_acc: 0.5620\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7fc106e3ae10>"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.fit_generator(augmentBatch, augmentBatch.nb_sample, nb_epoch=10, verbose=1, \n",
" validation_data=val_batches, nb_val_samples=val_batches.nb_sample)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Viewing model prediction examples (like in leasson 2) : TBD"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## conv layer from VGG\n",
"\n",
"I will use the weigth of VGG and remove all the non conv layers"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"import vgg16"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"vggModel=Vgg16()"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"____________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"====================================================================================================\n",
"lambda_1 (Lambda) (None, 3, 224, 224) 0 lambda_input_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_1 (ZeroPadding2D) (None, 3, 226, 226) 0 lambda_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_1 (Convolution2D) (None, 64, 224, 224) 1792 zeropadding2d_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_2 (ZeroPadding2D) (None, 64, 226, 226) 0 convolution2d_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_2 (Convolution2D) (None, 64, 224, 224) 36928 zeropadding2d_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_1 (MaxPooling2D) (None, 64, 112, 112) 0 convolution2d_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_3 (ZeroPadding2D) (None, 64, 114, 114) 0 maxpooling2d_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_3 (Convolution2D) (None, 128, 112, 112) 73856 zeropadding2d_3[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_4 (ZeroPadding2D) (None, 128, 114, 114) 0 convolution2d_3[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_4 (Convolution2D) (None, 128, 112, 112) 147584 zeropadding2d_4[0][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_2 (MaxPooling2D) (None, 128, 56, 56) 0 convolution2d_4[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_5 (ZeroPadding2D) (None, 128, 58, 58) 0 maxpooling2d_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_5 (Convolution2D) (None, 256, 56, 56) 295168 zeropadding2d_5[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_6 (ZeroPadding2D) (None, 256, 58, 58) 0 convolution2d_5[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_6 (Convolution2D) (None, 256, 56, 56) 590080 zeropadding2d_6[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_7 (ZeroPadding2D) (None, 256, 58, 58) 0 convolution2d_6[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_7 (Convolution2D) (None, 256, 56, 56) 590080 zeropadding2d_7[0][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_3 (MaxPooling2D) (None, 256, 28, 28) 0 convolution2d_7[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_8 (ZeroPadding2D) (None, 256, 30, 30) 0 maxpooling2d_3[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_8 (Convolution2D) (None, 512, 28, 28) 1180160 zeropadding2d_8[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_9 (ZeroPadding2D) (None, 512, 30, 30) 0 convolution2d_8[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_9 (Convolution2D) (None, 512, 28, 28) 2359808 zeropadding2d_9[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_10 (ZeroPadding2D) (None, 512, 30, 30) 0 convolution2d_9[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_10 (Convolution2D) (None, 512, 28, 28) 2359808 zeropadding2d_10[0][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_4 (MaxPooling2D) (None, 512, 14, 14) 0 convolution2d_10[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_11 (ZeroPadding2D) (None, 512, 16, 16) 0 maxpooling2d_4[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_11 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_11[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_12 (ZeroPadding2D) (None, 512, 16, 16) 0 convolution2d_11[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_12 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_12[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_13 (ZeroPadding2D) (None, 512, 16, 16) 0 convolution2d_12[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_13 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_13[0][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_5 (MaxPooling2D) (None, 512, 7, 7) 0 convolution2d_13[0][0] \n",
"____________________________________________________________________________________________________\n",
"flatten_1 (Flatten) (None, 25088) 0 maxpooling2d_5[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_1 (Dense) (None, 4096) 102764544 flatten_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"dropout_1 (Dropout) (None, 4096) 0 dense_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_2 (Dense) (None, 4096) 16781312 dropout_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"dropout_2 (Dropout) (None, 4096) 0 dense_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_3 (Dense) (None, 1000) 4097000 dropout_2[0][0] \n",
"====================================================================================================\n",
"Total params: 138357544\n",
"____________________________________________________________________________________________________\n"
]
}
],
"source": [
"vggModel.model.summary()"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"last conv layer index 37\n"
]
}
],
"source": [
"## get index of last conv layer\n",
"lastConvIndex = -1\n",
"for i in range(len(vggModel.model.layers)):\n",
" if (type(vggModel.model.layers[i]) == Convolution2D):\n",
" lastConvIndex=i\n",
" \n",
"print ('last conv layer index ' + str(i) )"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"convLayers= vggModel.model.layers[:lastConvIndex+1]"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"vggConvLayersModel = Sequential(convLayers)"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"____________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"====================================================================================================\n",
"lambda_1 (Lambda) (None, 3, 224, 224) 0 lambda_input_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_1 (ZeroPadding2D) (None, 3, 226, 226) 0 lambda_1[0][0] \n",
" lambda_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_1 (Convolution2D) (None, 64, 224, 224) 1792 zeropadding2d_1[0][0] \n",
" zeropadding2d_1[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_2 (ZeroPadding2D) (None, 64, 226, 226) 0 convolution2d_1[0][0] \n",
" convolution2d_1[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_2 (Convolution2D) (None, 64, 224, 224) 36928 zeropadding2d_2[0][0] \n",
" zeropadding2d_2[1][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_1 (MaxPooling2D) (None, 64, 112, 112) 0 convolution2d_2[0][0] \n",
" convolution2d_2[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_3 (ZeroPadding2D) (None, 64, 114, 114) 0 maxpooling2d_1[0][0] \n",
" maxpooling2d_1[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_3 (Convolution2D) (None, 128, 112, 112) 73856 zeropadding2d_3[0][0] \n",
" zeropadding2d_3[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_4 (ZeroPadding2D) (None, 128, 114, 114) 0 convolution2d_3[0][0] \n",
" convolution2d_3[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_4 (Convolution2D) (None, 128, 112, 112) 147584 zeropadding2d_4[0][0] \n",
" zeropadding2d_4[1][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_2 (MaxPooling2D) (None, 128, 56, 56) 0 convolution2d_4[0][0] \n",
" convolution2d_4[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_5 (ZeroPadding2D) (None, 128, 58, 58) 0 maxpooling2d_2[0][0] \n",
" maxpooling2d_2[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_5 (Convolution2D) (None, 256, 56, 56) 295168 zeropadding2d_5[0][0] \n",
" zeropadding2d_5[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_6 (ZeroPadding2D) (None, 256, 58, 58) 0 convolution2d_5[0][0] \n",
" convolution2d_5[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_6 (Convolution2D) (None, 256, 56, 56) 590080 zeropadding2d_6[0][0] \n",
" zeropadding2d_6[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_7 (ZeroPadding2D) (None, 256, 58, 58) 0 convolution2d_6[0][0] \n",
" convolution2d_6[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_7 (Convolution2D) (None, 256, 56, 56) 590080 zeropadding2d_7[0][0] \n",
" zeropadding2d_7[1][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_3 (MaxPooling2D) (None, 256, 28, 28) 0 convolution2d_7[0][0] \n",
" convolution2d_7[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_8 (ZeroPadding2D) (None, 256, 30, 30) 0 maxpooling2d_3[0][0] \n",
" maxpooling2d_3[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_8 (Convolution2D) (None, 512, 28, 28) 1180160 zeropadding2d_8[0][0] \n",
" zeropadding2d_8[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_9 (ZeroPadding2D) (None, 512, 30, 30) 0 convolution2d_8[0][0] \n",
" convolution2d_8[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_9 (Convolution2D) (None, 512, 28, 28) 2359808 zeropadding2d_9[0][0] \n",
" zeropadding2d_9[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_10 (ZeroPadding2D) (None, 512, 30, 30) 0 convolution2d_9[0][0] \n",
" convolution2d_9[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_10 (Convolution2D) (None, 512, 28, 28) 2359808 zeropadding2d_10[0][0] \n",
" zeropadding2d_10[1][0] \n",
"____________________________________________________________________________________________________\n",
"maxpooling2d_4 (MaxPooling2D) (None, 512, 14, 14) 0 convolution2d_10[0][0] \n",
" convolution2d_10[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_11 (ZeroPadding2D) (None, 512, 16, 16) 0 maxpooling2d_4[0][0] \n",
" maxpooling2d_4[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_11 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_11[0][0] \n",
" zeropadding2d_11[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_12 (ZeroPadding2D) (None, 512, 16, 16) 0 convolution2d_11[0][0] \n",
" convolution2d_11[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_12 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_12[0][0] \n",
" zeropadding2d_12[1][0] \n",
"____________________________________________________________________________________________________\n",
"zeropadding2d_13 (ZeroPadding2D) (None, 512, 16, 16) 0 convolution2d_12[0][0] \n",
" convolution2d_12[1][0] \n",
"____________________________________________________________________________________________________\n",
"convolution2d_13 (Convolution2D) (None, 512, 14, 14) 2359808 zeropadding2d_13[0][0] \n",
" zeropadding2d_13[1][0] \n",
"====================================================================================================\n",
"Total params: 14714688\n",
"____________________________________________________________________________________________________\n"
]
}
],
"source": [
"vggConvLayersModel.summary()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"### create an augmented training set\n",
"#### make sure not to use shuffle"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"path = \"/home/ubuntu/data/stateFarm/\"\n",
"#path = \"/home/ubuntu/data/stateFarm/sample/\""
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 17622 images belonging to 10 classes.\n"
]
}
],
"source": [
"#generate an augmented train batch\n",
"imgGen = image.ImageDataGenerator(rotation_range=15, height_shift_range=0.05, \n",
" shear_range=0.1, channel_shift_range=20, width_shift_range=0.1)\n",
"\n",
"augmentTrainBatch = imgGen.flow_from_directory(directory=path+'train', batch_size=64, target_size=(224,224), \n",
" shuffle=False, class_mode='categorical')"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"train_batches = augmentTrainBatch"
]
},
{
"cell_type": "code",
"execution_count": 69,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 4802 images belonging to 10 classes.\n"
]
}
],
"source": [
"#generate a new validation batch\n",
"val_batches = get_batches(path+'valid', batch_size=batch_size, shuffle=False, target_size=(224,224))"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 79726 images belonging to 1 classes.\n"
]
}
],
"source": [
"test_batch = get_batches(path+'test', batch_size=batch_size, shuffle=False, target_size=(224,224), class_mode='binary')"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA9YAAAELCAYAAAA81h5qAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f47b3195990>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"#display images\n",
"img,lbl = val_batches.next()\n",
"plots(img[:4], titles=hotone(lbl[:4]))\n"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"val_batches.reset()"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train_label_arr=onehot(train_batches.classes)"
]
},
{
"cell_type": "code",
"execution_count": 64,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"train_file_names_arr=train_batches.filenames"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"valid_label_arr=onehot(val_batches.classes)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"valid_file_names_arr=val_batches.filenames"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test_file_names_arr=test_batch.filenames"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Save arrays"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": true
},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'train_label_arr' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-10-40c516c3a3e0>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0msave_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"train_label_arr\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_label_arr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0msave_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"train_file_names_arr\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_file_names_arr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mNameError\u001b[0m: name 'train_label_arr' is not defined"
]
}
],
"source": [
"save_array(\"train_label_arr\", train_label_arr)\n",
"save_array(\"train_file_names_arr\", train_file_names_arr)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": true
},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'valid_label_arr' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-11-373613202196>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0msave_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"valid_label_arr\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalid_label_arr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0msave_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"valid_file_names_arr\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalid_file_names_arr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mNameError\u001b[0m: name 'valid_label_arr' is not defined"
]
}
],
"source": [
"save_array(\"valid_label_arr\", valid_label_arr)\n",
"save_array(\"valid_file_names_arr\", valid_file_names_arr)"
]
},
{
"cell_type": "code",
"execution_count": 67,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"save_array(\"test_file_names_arr\", test_file_names_arr)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load arrays"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train_label_arr = load_array(\"train_label_arr\")\n",
"train_file_names_arr = load_array(\"train_file_names_arr\")"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"valid_label_arr = load_array(\"valid_label_arr\")\n",
"valid_file_names_arr = load_array(\"valid_file_names_arr\")"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"test_file_names_arr = load_array(\"test_file_names_arr\")"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(4802, 10)"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"valid_label_arr.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### predict batches on pretrained model (base on conv layer of VGG16)"
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"predictTrainVggConv = vggConvLayersModel.predict_generator(train_batches, train_batches.N)"
]
},
{
"cell_type": "code",
"execution_count": 96,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predictValidationVggConv = vggConvLayersModel.predict_generator(val_batches, val_batches.N)"
]
},
{
"cell_type": "code",
"execution_count": 76,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predict_test_on_VggConv = vggConvLayersModel.predict_generator(test_batch, test_batch.N)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Save predication as carray"
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"save_array(\"predict_test_on_VggConv\", predict_test_on_VggConv)"
]
},
{
"cell_type": "code",
"execution_count": 101,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"save_array(\"predictTrainVggConv\", predictTrainVggConv)"
]
},
{
"cell_type": "code",
"execution_count": 102,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"save_array(\"predictValidationVggConv\", predictValidationVggConv)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Load prediction as carray"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predict_test_on_VggConv = load_carray(\"predict_test_on_VggConv\")"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predictTrainVggConv = load_carray(\"predictTrainVggConv\")"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predictValidationVggConv = load_carray(\"predictValidationVggConv\")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(79726, 512, 14, 14)"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predict_test_on_VggConv.shape"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(17622, 512, 14, 14)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predictTrainVggConv.shape"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(4802, 512, 14, 14)"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predictValidationVggConv.shape"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": true
},
"outputs": [
{
"data": {
"text/plain": [
"carray((79726, 512, 14, 14), float32)\n",
" nbytes: 29.80 GB; cbytes: 5.64 GB; ratio: 5.28\n",
" cparams := cparams(clevel=5, shuffle=1, cname='blosclz')\n",
" rootdir := 'predict_test_on_VggConv'\n",
" mode := 'a'\n"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predict_test_on_VggConv"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Create a dense model with input from vgg16 conv model I created before"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(512, 14, 14)"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#vggConvLayersModel is a pre-teained model with VGG16 conv layers only\n",
"#vggConvLayersModel prediction will be the input to this dense layer\n",
"vggConvLayersModel.layers[-1].output_shape[1:]"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def genDenseModel(layer):\n",
" baseModel = Sequential()\n",
" baseModel.add(MaxPooling2D(input_shape=layer.output_shape[1:]))\n",
" #baseModel.add(Convolution2D(32, 3, 3, activation='relu'))\n",
" #baseModel.add(BatchNormalization(axis=1))\n",
" #baseModel.add(MaxPooling2D(pool_size=(2,2)))\n",
" #baseModel.add(Convolution2D(64, 3, 3, activation='relu'))\n",
" #baseModel.add(BatchNormalization(axis=1))\n",
" #baseModel.add(MaxPooling2D(pool_size=(2,2)))\n",
" baseModel.add(Flatten())\n",
" baseModel.add(Dropout(0.5))\n",
" baseModel.add(Dense(256,activation='relu'))\n",
" baseModel.add(BatchNormalization())\n",
" baseModel.add(Dropout(0.5))\n",
" baseModel.add(Dense(256,activation='relu'))\n",
" baseModel.add(BatchNormalization())\n",
" baseModel.add(Dropout(0.8))\n",
" baseModel.add(Dense(10,activation='softmax'))\n",
" \n",
" baseModel.compile(Adam(lr=0.0001), \n",
" loss='categorical_crossentropy',\n",
" metrics=['accuracy'])\n",
" \n",
" return baseModel"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"denseModel = genDenseModel(vggConvLayersModel.layers[-1])\n"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"____________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"====================================================================================================\n",
"maxpooling2d_6 (MaxPooling2D) (None, 512, 7, 7) 0 maxpooling2d_input_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"flatten_2 (Flatten) (None, 25088) 0 maxpooling2d_6[0][0] \n",
"____________________________________________________________________________________________________\n",
"dropout_3 (Dropout) (None, 25088) 0 flatten_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_4 (Dense) (None, 256) 6422784 dropout_3[0][0] \n",
"____________________________________________________________________________________________________\n",
"batchnormalization_1 (BatchNormal(None, 256) 512 dense_4[0][0] \n",
"____________________________________________________________________________________________________\n",
"dropout_4 (Dropout) (None, 256) 0 batchnormalization_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_5 (Dense) (None, 256) 65792 dropout_4[0][0] \n",
"____________________________________________________________________________________________________\n",
"batchnormalization_2 (BatchNormal(None, 256) 512 dense_5[0][0] \n",
"____________________________________________________________________________________________________\n",
"dropout_5 (Dropout) (None, 256) 0 batchnormalization_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_6 (Dense) (None, 10) 2570 dropout_5[0][0] \n",
"====================================================================================================\n",
"Total params: 6492170\n",
"____________________________________________________________________________________________________\n"
]
}
],
"source": [
"denseModel.summary()"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"#start low learning rate\n",
"denseModel.compile(Adam(lr=0.01), \n",
" loss='categorical_crossentropy',\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 17622 samples, validate on 4802 samples\n",
"Epoch 1/6\n",
"17622/17622 [==============================] - 11s - loss: 1.6289 - acc: 0.5319 - val_loss: 0.5995 - val_acc: 0.8020\n",
"Epoch 2/6\n",
"17622/17622 [==============================] - 10s - loss: 0.6179 - acc: 0.8387 - val_loss: 0.5553 - val_acc: 0.8201\n",
"Epoch 3/6\n",
"17622/17622 [==============================] - 10s - loss: 0.4833 - acc: 0.8888 - val_loss: 0.6474 - val_acc: 0.7888\n",
"Epoch 4/6\n",
"17622/17622 [==============================] - 10s - loss: 0.4619 - acc: 0.8977 - val_loss: 0.5377 - val_acc: 0.8297\n",
"Epoch 5/6\n",
"17622/17622 [==============================] - 10s - loss: 0.4076 - acc: 0.9172 - val_loss: 0.6088 - val_acc: 0.8009\n",
"Epoch 6/6\n",
"17622/17622 [==============================] - 10s - loss: 0.4227 - acc: 0.9161 - val_loss: 0.6452 - val_acc: 0.7963\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7faec3d1eb10>"
]
},
"execution_count": 38,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(predictTrainVggConv, predict_train_label_arr, nb_epoch=6, verbose=1, \n",
" validation_data=(predictValidationVggConv,predict_valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 17622 samples, validate on 4802 samples\n",
"Epoch 1/6\n",
"17622/17622 [==============================] - 11s - loss: 0.3889 - acc: 0.9234 - val_loss: 0.4878 - val_acc: 0.8430\n",
"Epoch 2/6\n",
"17622/17622 [==============================] - 10s - loss: 0.3800 - acc: 0.9277 - val_loss: 0.6478 - val_acc: 0.8057\n",
"Epoch 3/6\n",
"17622/17622 [==============================] - 10s - loss: 0.3862 - acc: 0.9284 - val_loss: 0.6066 - val_acc: 0.8061\n",
"Epoch 4/6\n",
"17622/17622 [==============================] - 10s - loss: 0.3557 - acc: 0.9347 - val_loss: 0.5317 - val_acc: 0.8374\n",
"Epoch 5/6\n",
"17622/17622 [==============================] - 10s - loss: 0.3580 - acc: 0.9355 - val_loss: 0.7985 - val_acc: 0.7553\n",
"Epoch 6/6\n",
"17622/17622 [==============================] - 10s - loss: 0.3281 - acc: 0.9425 - val_loss: 0.6677 - val_acc: 0.8294\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7faec3d21710>"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.optimizer.lr= 0.0001\n",
"denseModel.fit(predictTrainVggConv, predict_train_label_arr, nb_epoch=6, verbose=1, \n",
" validation_data=(predictValidationVggConv,predict_valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"collapsed": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 17622 samples, validate on 4802 samples\n",
"Epoch 1/20\n",
"17622/17622 [==============================] - 11s - loss: 0.3481 - acc: 0.9398 - val_loss: 0.6624 - val_acc: 0.8057\n",
"Epoch 2/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3353 - acc: 0.9426 - val_loss: 0.5394 - val_acc: 0.8199\n",
"Epoch 3/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3218 - acc: 0.9449 - val_loss: 0.8696 - val_acc: 0.7676\n",
"Epoch 4/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3353 - acc: 0.9453 - val_loss: 0.5419 - val_acc: 0.8347\n",
"Epoch 5/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3014 - acc: 0.9498 - val_loss: 0.6754 - val_acc: 0.8215\n",
"Epoch 6/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2896 - acc: 0.9526 - val_loss: 0.5942 - val_acc: 0.8344\n",
"Epoch 7/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3038 - acc: 0.9535 - val_loss: 0.7759 - val_acc: 0.7791\n",
"Epoch 8/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2944 - acc: 0.9530 - val_loss: 0.6203 - val_acc: 0.8261\n",
"Epoch 9/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2791 - acc: 0.9566 - val_loss: 0.6107 - val_acc: 0.8415\n",
"Epoch 10/20\n",
"17622/17622 [==============================] - 10s - loss: 0.3153 - acc: 0.9501 - val_loss: 0.5159 - val_acc: 0.8488\n",
"Epoch 11/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2896 - acc: 0.9564 - val_loss: 0.5793 - val_acc: 0.8374\n",
"Epoch 12/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2599 - acc: 0.9622 - val_loss: 0.5487 - val_acc: 0.8438\n",
"Epoch 13/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2473 - acc: 0.9662 - val_loss: 0.6705 - val_acc: 0.8099\n",
"Epoch 14/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2763 - acc: 0.9604 - val_loss: 0.6171 - val_acc: 0.8455\n",
"Epoch 15/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2677 - acc: 0.9627 - val_loss: 0.7420 - val_acc: 0.8142\n",
"Epoch 16/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2643 - acc: 0.9637 - val_loss: 0.8399 - val_acc: 0.7861\n",
"Epoch 17/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2643 - acc: 0.9628 - val_loss: 0.7567 - val_acc: 0.8130\n",
"Epoch 18/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2769 - acc: 0.9597 - val_loss: 0.7878 - val_acc: 0.7943\n",
"Epoch 19/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2686 - acc: 0.9615 - val_loss: 0.7557 - val_acc: 0.7947\n",
"Epoch 20/20\n",
"17622/17622 [==============================] - 10s - loss: 0.2465 - acc: 0.9684 - val_loss: 0.6179 - val_acc: 0.8255\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7faec3d21450>"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(predictTrainVggConv, predict_train_label_arr, nb_epoch=20, verbose=1, \n",
" validation_data=(predictValidationVggConv,predict_valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 113,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.save_weights('State_Farm_Trained.h5')"
]
},
{
"cell_type": "code",
"execution_count": 86,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.load_weights('State_Farm_Trained.h5')"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"## 1st Submit- score 0.818"
]
},
{
"cell_type": "code",
"execution_count": 85,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def do_clip(arr, mx): return np.clip(arr, (1-mx)/9, mx)"
]
},
{
"cell_type": "code",
"execution_count": 83,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [],
"source": [
"predictTest = denseModel.predict(predict_test_on_VggConv)"
]
},
{
"cell_type": "code",
"execution_count": 87,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"clip_test_predict = do_clip(predictTest, 0.93)"
]
},
{
"cell_type": "code",
"execution_count": 89,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 0.0078, 0.93 , 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078], dtype=float32)"
]
},
"execution_count": 89,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"clip_test_predict[70000]"
]
},
{
"cell_type": "code",
"execution_count": 90,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"subm_name = path+'results/subm.gz'"
]
},
{
"cell_type": "code",
"execution_count": 107,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"subm_name = path+'results/subm.csv'"
]
},
{
"cell_type": "code",
"execution_count": 100,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"classes = sorted(val_batches.class_indices, key=val_batches.class_indices.get)"
]
},
{
"cell_type": "code",
"execution_count": 101,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"['c0', 'c1', 'c2', 'c3', 'c4', 'c5', 'c6', 'c7', 'c8', 'c9']"
]
},
"execution_count": 101,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"classes"
]
},
{
"cell_type": "code",
"execution_count": 104,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0.132033</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.073707</td>\n",
" <td>0.602998</td>\n",
" <td>0.008529</td>\n",
" <td>0.014823</td>\n",
" <td>0.007778</td>\n",
" <td>0.133301</td>\n",
" <td>0.023982</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>0.007778</td>\n",
" <td>0.084785</td>\n",
" <td>0.015637</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.324112</td>\n",
" <td>0.007778</td>\n",
" <td>0.553448</td>\n",
" <td>0.009105</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.008971</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.008923</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" c0 c1 c2 c3 c4 c5 c6 \\\n",
"0 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"1 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"2 0.132033 0.007778 0.007778 0.073707 0.602998 0.008529 0.014823 \n",
"3 0.007778 0.084785 0.015637 0.007778 0.007778 0.007778 0.324112 \n",
"4 0.007778 0.007778 0.007778 0.007778 0.007778 0.008971 0.007778 \n",
"\n",
" c7 c8 c9 \n",
"0 0.007778 0.007778 0.930000 \n",
"1 0.007778 0.007778 0.930000 \n",
"2 0.007778 0.133301 0.023982 \n",
"3 0.007778 0.553448 0.009105 \n",
"4 0.007778 0.008923 0.930000 "
]
},
"execution_count": 104,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission = pd.DataFrame(clip_test_predict, columns=classes)\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 105,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>img</th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>img_81601.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>img_14887.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>img_62885.jpg</td>\n",
" <td>0.132033</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.073707</td>\n",
" <td>0.602998</td>\n",
" <td>0.008529</td>\n",
" <td>0.014823</td>\n",
" <td>0.007778</td>\n",
" <td>0.133301</td>\n",
" <td>0.023982</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>img_45125.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.084785</td>\n",
" <td>0.015637</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.324112</td>\n",
" <td>0.007778</td>\n",
" <td>0.553448</td>\n",
" <td>0.009105</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>img_22633.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.008971</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.008923</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" img c0 c1 c2 c3 c4 c5 \\\n",
"0 img_81601.jpg 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"1 img_14887.jpg 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"2 img_62885.jpg 0.132033 0.007778 0.007778 0.073707 0.602998 0.008529 \n",
"3 img_45125.jpg 0.007778 0.084785 0.015637 0.007778 0.007778 0.007778 \n",
"4 img_22633.jpg 0.007778 0.007778 0.007778 0.007778 0.007778 0.008971 \n",
"\n",
" c6 c7 c8 c9 \n",
"0 0.007778 0.007778 0.007778 0.930000 \n",
"1 0.007778 0.007778 0.007778 0.930000 \n",
"2 0.014823 0.007778 0.133301 0.023982 \n",
"3 0.324112 0.007778 0.553448 0.009105 \n",
"4 0.007778 0.007778 0.008923 0.930000 "
]
},
"execution_count": 105,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission.insert(0, 'img', [a[5:] for a in test_file_names])\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 108,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"submission.to_csv(subm_name, index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### train with 5 time the images"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"path = \"/home/ubuntu/data/stateFarm/\"\n",
"#path = \"/home/ubuntu/data/stateFarm/sample/\""
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"#generate an augmanted - not shuffled- img generator\n",
"imgGen = image.ImageDataGenerator(rotation_range=15, height_shift_range=0.05, \n",
" shear_range=0.1, channel_shift_range=20, width_shift_range=0.1)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 17622 images belonging to 10 classes.\n"
]
}
],
"source": [
"augmentTrainBatch = imgGen.flow_from_directory(directory=path+'train', batch_size=64, target_size=(224,224), \n",
" shuffle=False, class_mode='categorical')\n",
" "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"trainBatches = get_batches(path+'train', batch_size=batch_size, target_size=(224,224), shuffle=False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# predict X5 the augmented train batch on VGG16 conv layers\n",
"predict_5X_augment_train_vgg_conv = vggConvLayersModel.predict_generator(augmentTrainBatch, augmentTrainBatch.N*5)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"train_labels_arr = onehot(augmentTrainBatch.classes) "
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"X5_train_labels_arr = np.concatenate([train_labels_arr]*5)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Save prediction and labels"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"save_array('X5_train_labels_arr.dat', X5_train_labels_arr)"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"save_array('predict_5X_augment_train_vgg_conv.dat', predict_5X_augment_train_vgg_conv)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load prediction as carray"
]
},
{
"cell_type": "code",
"execution_count": 76,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"predict_5X_augment_train_vgg_conv = load_carray('predict_5X_augment_train_vgg_conv.dat')"
]
},
{
"cell_type": "code",
"execution_count": 75,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"X5_train_labels_arr = load_array('X5_train_labels_arr.dat')"
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(88110, 512, 14, 14)"
]
},
"execution_count": 77,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predict_5X_augment_train_vgg_conv.shape"
]
},
{
"cell_type": "code",
"execution_count": 78,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(88110, 10)"
]
},
"execution_count": 78,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X5_train_labels_arr.shape"
]
},
{
"cell_type": "code",
"execution_count": 90,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 88110 samples, validate on 4802 samples\n",
"Epoch 1/1\n",
"88110/88110 [==============================] - 59s - loss: 0.3505 - acc: 0.9447 - val_loss: 1.0957 - val_acc: 0.8380\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f47b10515d0>"
]
},
"execution_count": 90,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(predict_5X_augment_train_vgg_conv[:], X5_train_labels_arr, nb_epoch=1, verbose=1, \n",
" validation_data=(predictValidationVggConv,valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 91,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.optimizer.lr=0.000001"
]
},
{
"cell_type": "code",
"execution_count": 94,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 88110 samples, validate on 4802 samples\n",
"Epoch 1/25\n",
"88110/88110 [==============================] - 56s - loss: 0.3156 - acc: 0.9485 - val_loss: 1.0094 - val_acc: 0.8436\n",
"Epoch 2/25\n",
"88110/88110 [==============================] - 56s - loss: 0.2924 - acc: 0.9504 - val_loss: 0.9595 - val_acc: 0.8442\n",
"Epoch 3/25\n",
"88110/88110 [==============================] - 56s - loss: 0.2703 - acc: 0.9518 - val_loss: 0.9251 - val_acc: 0.8455\n",
"Epoch 4/25\n",
"88110/88110 [==============================] - 56s - loss: 0.2459 - acc: 0.9538 - val_loss: 0.8799 - val_acc: 0.8476\n",
"Epoch 5/25\n",
"88110/88110 [==============================] - 56s - loss: 0.2278 - acc: 0.9554 - val_loss: 0.8386 - val_acc: 0.8476\n",
"Epoch 6/25\n",
"88110/88110 [==============================] - 56s - loss: 0.2045 - acc: 0.9572 - val_loss: 0.7962 - val_acc: 0.8482\n",
"Epoch 7/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1973 - acc: 0.9569 - val_loss: 0.7375 - val_acc: 0.8501\n",
"Epoch 8/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1822 - acc: 0.9577 - val_loss: 0.6977 - val_acc: 0.8496\n",
"Epoch 9/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1694 - acc: 0.9582 - val_loss: 0.6617 - val_acc: 0.8503\n",
"Epoch 10/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1620 - acc: 0.9582 - val_loss: 0.6277 - val_acc: 0.8480\n",
"Epoch 11/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1494 - acc: 0.9599 - val_loss: 0.5813 - val_acc: 0.8515\n",
"Epoch 12/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1424 - acc: 0.9610 - val_loss: 0.5543 - val_acc: 0.8517\n",
"Epoch 13/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1348 - acc: 0.9614 - val_loss: 0.5217 - val_acc: 0.8553\n",
"Epoch 14/25\n",
"88110/88110 [==============================] - 57s - loss: 0.1272 - acc: 0.9629 - val_loss: 0.5178 - val_acc: 0.8515\n",
"Epoch 15/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1250 - acc: 0.9632 - val_loss: 0.5086 - val_acc: 0.8519\n",
"Epoch 16/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1232 - acc: 0.9633 - val_loss: 0.4939 - val_acc: 0.8538\n",
"Epoch 17/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1203 - acc: 0.9640 - val_loss: 0.4817 - val_acc: 0.8553\n",
"Epoch 18/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1203 - acc: 0.9639 - val_loss: 0.4911 - val_acc: 0.8507\n",
"Epoch 19/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1173 - acc: 0.9648 - val_loss: 0.4797 - val_acc: 0.8526\n",
"Epoch 20/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1147 - acc: 0.9656 - val_loss: 0.4846 - val_acc: 0.8544\n",
"Epoch 21/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1114 - acc: 0.9670 - val_loss: 0.4798 - val_acc: 0.8532\n",
"Epoch 22/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1103 - acc: 0.9659 - val_loss: 0.4817 - val_acc: 0.8530\n",
"Epoch 23/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1098 - acc: 0.9666 - val_loss: 0.4768 - val_acc: 0.8549\n",
"Epoch 24/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1050 - acc: 0.9683 - val_loss: 0.4791 - val_acc: 0.8544\n",
"Epoch 25/25\n",
"88110/88110 [==============================] - 56s - loss: 0.1059 - acc: 0.9679 - val_loss: 0.4696 - val_acc: 0.8567\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f47b1dc4d90>"
]
},
"execution_count": 94,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(predict_5X_augment_train_vgg_conv[:], X5_train_labels_arr, nb_epoch=25, verbose=1, \n",
" validation_data=(predictValidationVggConv,valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.save_weights('State_Farm_Trained_val_acc_0.8567_over_fit.h5')"
]
},
{
"cell_type": "code",
"execution_count": 126,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.load_weights('State_Farm_Trained_val_acc_0.8567_over_fit.h5')"
]
},
{
"cell_type": "code",
"execution_count": 127,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"4800/4802 [============================>.] - ETA: 0s"
]
},
{
"data": {
"text/plain": [
"[0.46957097133417541, 0.8567263640149938]"
]
},
"execution_count": 127,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"## validate val_acc correct after loading weights\n",
"denseModel.evaluate(predictValidationVggConv,valid_label_arr)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"## 2st Submit - score 0.80768 (with clipping of 0.93)\n",
"(without clipping it is worse 0.869...)\n"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def do_clip(arr, mx): return np.clip(arr, ((1-mx)/9.0), mx)"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [],
"source": [
"predictTest = denseModel.predict(predict_test_on_VggConv)"
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"clip_test_predict = do_clip(predictTest, 0.93)"
]
},
{
"cell_type": "code",
"execution_count": 78,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 0.0105, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0092, 0.93 ],\n",
" [ 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.93 ]], dtype=float32)"
]
},
"execution_count": 78,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"clip_test_predict[:2]"
]
},
{
"cell_type": "code",
"execution_count": 87,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"subm_name = path+'results/thirg_subm.gz'"
]
},
{
"cell_type": "code",
"execution_count": 88,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"classes = sorted(val_batches.class_indices, key=val_batches.class_indices.get)"
]
},
{
"cell_type": "code",
"execution_count": 89,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"['c0', 'c1', 'c2', 'c3', 'c4', 'c5', 'c6', 'c7', 'c8', 'c9']"
]
},
"execution_count": 89,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"classes"
]
},
{
"cell_type": "code",
"execution_count": 90,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0.010526</td>\n",
" <td>0.000124</td>\n",
" <td>0.000098</td>\n",
" <td>0.000021</td>\n",
" <td>0.000189</td>\n",
" <td>0.000122</td>\n",
" <td>0.002255</td>\n",
" <td>0.006140</td>\n",
" <td>0.009152</td>\n",
" <td>0.971373</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.005590</td>\n",
" <td>0.000041</td>\n",
" <td>0.000099</td>\n",
" <td>0.000006</td>\n",
" <td>0.000023</td>\n",
" <td>0.001325</td>\n",
" <td>0.000013</td>\n",
" <td>0.000013</td>\n",
" <td>0.000647</td>\n",
" <td>0.992241</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0.004217</td>\n",
" <td>0.000018</td>\n",
" <td>0.000778</td>\n",
" <td>0.011714</td>\n",
" <td>0.975617</td>\n",
" <td>0.002211</td>\n",
" <td>0.000222</td>\n",
" <td>0.000002</td>\n",
" <td>0.004648</td>\n",
" <td>0.000575</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>0.000139</td>\n",
" <td>0.000503</td>\n",
" <td>0.002671</td>\n",
" <td>0.000013</td>\n",
" <td>0.000220</td>\n",
" <td>0.000007</td>\n",
" <td>0.351552</td>\n",
" <td>0.002226</td>\n",
" <td>0.641944</td>\n",
" <td>0.000726</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>0.004608</td>\n",
" <td>0.002961</td>\n",
" <td>0.001624</td>\n",
" <td>0.000098</td>\n",
" <td>0.000630</td>\n",
" <td>0.002191</td>\n",
" <td>0.000471</td>\n",
" <td>0.006208</td>\n",
" <td>0.046550</td>\n",
" <td>0.934658</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" c0 c1 c2 c3 c4 c5 c6 \\\n",
"0 0.010526 0.000124 0.000098 0.000021 0.000189 0.000122 0.002255 \n",
"1 0.005590 0.000041 0.000099 0.000006 0.000023 0.001325 0.000013 \n",
"2 0.004217 0.000018 0.000778 0.011714 0.975617 0.002211 0.000222 \n",
"3 0.000139 0.000503 0.002671 0.000013 0.000220 0.000007 0.351552 \n",
"4 0.004608 0.002961 0.001624 0.000098 0.000630 0.002191 0.000471 \n",
"\n",
" c7 c8 c9 \n",
"0 0.006140 0.009152 0.971373 \n",
"1 0.000013 0.000647 0.992241 \n",
"2 0.000002 0.004648 0.000575 \n",
"3 0.002226 0.641944 0.000726 \n",
"4 0.006208 0.046550 0.934658 "
]
},
"execution_count": 90,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission = pd.DataFrame(predictTest, columns=classes)\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 91,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>img</th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>img_81601.jpg</td>\n",
" <td>0.010526</td>\n",
" <td>0.000124</td>\n",
" <td>0.000098</td>\n",
" <td>0.000021</td>\n",
" <td>0.000189</td>\n",
" <td>0.000122</td>\n",
" <td>0.002255</td>\n",
" <td>0.006140</td>\n",
" <td>0.009152</td>\n",
" <td>0.971373</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>img_14887.jpg</td>\n",
" <td>0.005590</td>\n",
" <td>0.000041</td>\n",
" <td>0.000099</td>\n",
" <td>0.000006</td>\n",
" <td>0.000023</td>\n",
" <td>0.001325</td>\n",
" <td>0.000013</td>\n",
" <td>0.000013</td>\n",
" <td>0.000647</td>\n",
" <td>0.992241</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>img_62885.jpg</td>\n",
" <td>0.004217</td>\n",
" <td>0.000018</td>\n",
" <td>0.000778</td>\n",
" <td>0.011714</td>\n",
" <td>0.975617</td>\n",
" <td>0.002211</td>\n",
" <td>0.000222</td>\n",
" <td>0.000002</td>\n",
" <td>0.004648</td>\n",
" <td>0.000575</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>img_45125.jpg</td>\n",
" <td>0.000139</td>\n",
" <td>0.000503</td>\n",
" <td>0.002671</td>\n",
" <td>0.000013</td>\n",
" <td>0.000220</td>\n",
" <td>0.000007</td>\n",
" <td>0.351552</td>\n",
" <td>0.002226</td>\n",
" <td>0.641944</td>\n",
" <td>0.000726</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>img_22633.jpg</td>\n",
" <td>0.004608</td>\n",
" <td>0.002961</td>\n",
" <td>0.001624</td>\n",
" <td>0.000098</td>\n",
" <td>0.000630</td>\n",
" <td>0.002191</td>\n",
" <td>0.000471</td>\n",
" <td>0.006208</td>\n",
" <td>0.046550</td>\n",
" <td>0.934658</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" img c0 c1 c2 c3 c4 c5 \\\n",
"0 img_81601.jpg 0.010526 0.000124 0.000098 0.000021 0.000189 0.000122 \n",
"1 img_14887.jpg 0.005590 0.000041 0.000099 0.000006 0.000023 0.001325 \n",
"2 img_62885.jpg 0.004217 0.000018 0.000778 0.011714 0.975617 0.002211 \n",
"3 img_45125.jpg 0.000139 0.000503 0.002671 0.000013 0.000220 0.000007 \n",
"4 img_22633.jpg 0.004608 0.002961 0.001624 0.000098 0.000630 0.002191 \n",
"\n",
" c6 c7 c8 c9 \n",
"0 0.002255 0.006140 0.009152 0.971373 \n",
"1 0.000013 0.000013 0.000647 0.992241 \n",
"2 0.000222 0.000002 0.004648 0.000575 \n",
"3 0.351552 0.002226 0.641944 0.000726 \n",
"4 0.000471 0.006208 0.046550 0.934658 "
]
},
"execution_count": 91,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission.insert(0, 'img', [a[5:] for a in test_file_names_arr])\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 92,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"submission.to_csv(subm_name, index=False, compression='gzip')"
]
},
{
"cell_type": "code",
"execution_count": 85,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<a href='/home/ubuntu/data/stateFarm/results/second_subm.gz' target='_blank'>/home/ubuntu/data/stateFarm/results/second_subm.gz</a><br>"
],
"text/plain": [
"/home/ubuntu/data/stateFarm/results/second_subm.gz"
]
},
"execution_count": 85,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"FileLink(subm_name)"
]
},
{
"cell_type": "code",
"execution_count": 94,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import random\n",
"\n",
"def shuffle_img_lable_arrays(imgs, lables):\n",
" if (len(imgs) != len(lables)):\n",
" return\n",
" \n",
" imgLabelPairList = []\n",
" \n",
" for i in range(len(lables)):\n",
" single_img=imgs[i]\n",
" single_label=lables[i]\n",
" imgLabelPairList.append((single_img,single_label))\n",
" \n",
" # shuffel the list\n",
" random.shuffle(imgLabelPairList)\n",
" \n",
" #split list to image and label np array\n",
" len_list = len(imgLabelPairList)\n",
" #for i in range(len_list):\n",
" shuffeledImgArray = np.stack((np.array(imgLabelPairList[i][0]) for i in range(len_list)))\n",
" shuffeledLblArray = np.stack((np.array(imgLabelPairList[i][1]) for i in range(len_list)))\n",
" \n",
" return (shuffeledImgArray, shuffeledLblArray)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"## train with psuado labeled test DATA"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### predict test images and use as psuado labeled data"
]
},
{
"cell_type": "code",
"execution_count": 102,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"test_set_prediction = np.round(predictTest)"
]
},
{
"cell_type": "code",
"execution_count": 105,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(79726, 10)"
]
},
"execution_count": 105,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test_set_prediction.shape"
]
},
{
"cell_type": "code",
"execution_count": 104,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.], dtype=float32)"
]
},
"execution_count": 104,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test_set_prediction[70]"
]
},
{
"cell_type": "code",
"execution_count": 112,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"train_w_pasudo_input = predict_test_on_VggConv.copy()"
]
},
{
"cell_type": "code",
"execution_count": 115,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train_w_pasudo_input.append(predictTrainVggConv)"
]
},
{
"cell_type": "code",
"execution_count": 122,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"train_w_pasudo_labels = np.concatenate([test_set_prediction, train_label_arr])"
]
},
{
"cell_type": "code",
"execution_count": 116,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(97348, 512, 14, 14)"
]
},
"execution_count": 116,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train_w_pasudo_input.shape"
]
},
{
"cell_type": "code",
"execution_count": 123,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"(97348, 10)"
]
},
"execution_count": 123,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"train_w_pasudo_labels.shape"
]
},
{
"cell_type": "code",
"execution_count": 128,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.optimizer.lr=0.0001"
]
},
{
"cell_type": "code",
"execution_count": 129,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 97348 samples, validate on 4802 samples\n",
"Epoch 1/10\n",
"97348/97348 [==============================] - 130s - loss: 0.4880 - acc: 0.8160 - val_loss: 0.4485 - val_acc: 0.8580\n",
"Epoch 2/10\n",
"97348/97348 [==============================] - 130s - loss: 0.4378 - acc: 0.8237 - val_loss: 0.4258 - val_acc: 0.8599\n",
"Epoch 3/10\n",
"97348/97348 [==============================] - 130s - loss: 0.4203 - acc: 0.8269 - val_loss: 0.4175 - val_acc: 0.8619\n",
"Epoch 4/10\n",
"97348/97348 [==============================] - 130s - loss: 0.4042 - acc: 0.8341 - val_loss: 0.4004 - val_acc: 0.8655\n",
"Epoch 5/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3952 - acc: 0.8380 - val_loss: 0.4056 - val_acc: 0.8638\n",
"Epoch 6/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3809 - acc: 0.8407 - val_loss: 0.4009 - val_acc: 0.8653\n",
"Epoch 7/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3720 - acc: 0.8456 - val_loss: 0.4022 - val_acc: 0.8655\n",
"Epoch 8/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3659 - acc: 0.8468 - val_loss: 0.3964 - val_acc: 0.8676\n",
"Epoch 9/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3590 - acc: 0.8506 - val_loss: 0.3976 - val_acc: 0.8663\n",
"Epoch 10/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3489 - acc: 0.8531 - val_loss: 0.3935 - val_acc: 0.8669\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f97b985ae50>"
]
},
"execution_count": 129,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(train_w_pasudo_input, train_w_pasudo_labels, nb_epoch=10, verbose=1, \n",
" validation_data=(predictValidationVggConv,valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 130,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 97348 samples, validate on 4802 samples\n",
"Epoch 1/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3418 - acc: 0.8559 - val_loss: 0.3877 - val_acc: 0.8715\n",
"Epoch 2/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3352 - acc: 0.8588 - val_loss: 0.3927 - val_acc: 0.8698\n",
"Epoch 3/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3345 - acc: 0.8588 - val_loss: 0.3946 - val_acc: 0.8692\n",
"Epoch 4/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3281 - acc: 0.8601 - val_loss: 0.3840 - val_acc: 0.8723\n",
"Epoch 5/10\n",
"97348/97348 [==============================] - 129s - loss: 0.3254 - acc: 0.8613 - val_loss: 0.3808 - val_acc: 0.8738\n",
"Epoch 6/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3199 - acc: 0.8653 - val_loss: 0.3888 - val_acc: 0.8711\n",
"Epoch 7/10\n",
"97348/97348 [==============================] - 129s - loss: 0.3159 - acc: 0.8655 - val_loss: 0.3841 - val_acc: 0.8723\n",
"Epoch 8/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3098 - acc: 0.8679 - val_loss: 0.3830 - val_acc: 0.8732\n",
"Epoch 9/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3106 - acc: 0.8680 - val_loss: 0.3823 - val_acc: 0.8732\n",
"Epoch 10/10\n",
"97348/97348 [==============================] - 130s - loss: 0.3043 - acc: 0.8690 - val_loss: 0.3778 - val_acc: 0.8723\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f98379cc3d0>"
]
},
"execution_count": 130,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"denseModel.fit(train_w_pasudo_input, train_w_pasudo_labels, nb_epoch=10, verbose=1, \n",
" validation_data=(predictValidationVggConv,valid_label_arr))"
]
},
{
"cell_type": "code",
"execution_count": 131,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"denseModel.save_weights('Psodu_labels_State_Farm_Trained_val_acc_0.8723.h5')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3rd Submit - 0.71513"
]
},
{
"cell_type": "code",
"execution_count": 132,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [],
"source": [
"predictTest = denseModel.predict(predict_test_on_VggConv)"
]
},
{
"cell_type": "code",
"execution_count": 133,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"clip_test_predict = do_clip(predictTest, 0.93)"
]
},
{
"cell_type": "code",
"execution_count": 134,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 0.0097, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0107, 0.93 ],\n",
" [ 0.0378, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.0078, 0.93 ]], dtype=float32)"
]
},
"execution_count": 134,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"clip_test_predict[:2]"
]
},
{
"cell_type": "code",
"execution_count": 135,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"subm_name = path+'results/fourth_subm.gz'"
]
},
{
"cell_type": "code",
"execution_count": 136,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"classes = sorted(val_batches.class_indices, key=val_batches.class_indices.get)"
]
},
{
"cell_type": "code",
"execution_count": 137,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"['c0', 'c1', 'c2', 'c3', 'c4', 'c5', 'c6', 'c7', 'c8', 'c9']"
]
},
"execution_count": 137,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"classes"
]
},
{
"cell_type": "code",
"execution_count": 139,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0.009737</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010736</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0.037769</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010362</td>\n",
" <td>0.930000</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.304589</td>\n",
" <td>0.007778</td>\n",
" <td>0.687587</td>\n",
" <td>0.007778</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>0.014843</td>\n",
" <td>0.011130</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010767</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.215934</td>\n",
" <td>0.724175</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" c0 c1 c2 c3 c4 c5 c6 \\\n",
"0 0.009737 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"1 0.037769 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"2 0.007778 0.007778 0.007778 0.010362 0.930000 0.007778 0.007778 \n",
"3 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 0.304589 \n",
"4 0.014843 0.011130 0.007778 0.007778 0.007778 0.010767 0.007778 \n",
"\n",
" c7 c8 c9 \n",
"0 0.007778 0.010736 0.930000 \n",
"1 0.007778 0.007778 0.930000 \n",
"2 0.007778 0.007778 0.007778 \n",
"3 0.007778 0.687587 0.007778 \n",
"4 0.007778 0.215934 0.724175 "
]
},
"execution_count": 139,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission = pd.DataFrame(clip_test_predict, columns=classes)\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 140,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>img</th>\n",
" <th>c0</th>\n",
" <th>c1</th>\n",
" <th>c2</th>\n",
" <th>c3</th>\n",
" <th>c4</th>\n",
" <th>c5</th>\n",
" <th>c6</th>\n",
" <th>c7</th>\n",
" <th>c8</th>\n",
" <th>c9</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>img_81601.jpg</td>\n",
" <td>0.009737</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010736</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>img_14887.jpg</td>\n",
" <td>0.037769</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.930000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>img_62885.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010362</td>\n",
" <td>0.930000</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>img_45125.jpg</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.304589</td>\n",
" <td>0.007778</td>\n",
" <td>0.687587</td>\n",
" <td>0.007778</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>img_22633.jpg</td>\n",
" <td>0.014843</td>\n",
" <td>0.011130</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.010767</td>\n",
" <td>0.007778</td>\n",
" <td>0.007778</td>\n",
" <td>0.215934</td>\n",
" <td>0.724175</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" img c0 c1 c2 c3 c4 c5 \\\n",
"0 img_81601.jpg 0.009737 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"1 img_14887.jpg 0.037769 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"2 img_62885.jpg 0.007778 0.007778 0.007778 0.010362 0.930000 0.007778 \n",
"3 img_45125.jpg 0.007778 0.007778 0.007778 0.007778 0.007778 0.007778 \n",
"4 img_22633.jpg 0.014843 0.011130 0.007778 0.007778 0.007778 0.010767 \n",
"\n",
" c6 c7 c8 c9 \n",
"0 0.007778 0.007778 0.010736 0.930000 \n",
"1 0.007778 0.007778 0.007778 0.930000 \n",
"2 0.007778 0.007778 0.007778 0.007778 \n",
"3 0.304589 0.007778 0.687587 0.007778 \n",
"4 0.007778 0.007778 0.215934 0.724175 "
]
},
"execution_count": 140,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"submission.insert(0, 'img', [a[5:] for a in test_file_names_arr])\n",
"submission.head()"
]
},
{
"cell_type": "code",
"execution_count": 141,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"submission.to_csv(subm_name, index=False, compression='gzip')"
]
}
],
"metadata": {
"anaconda-cloud": {},
"kernelspec": {
"display_name": "Python [conda root]",
"language": "python",
"name": "conda-root-py"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.12"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment