{"cells":[{"cell_type":"markdown","metadata":{"id":"Ss4WkYFDOGTu"},"source":["# Contextual online advertising with user data from U.S. Census"]},{"cell_type":"markdown","metadata":{"id":"nAajOomTOSV6"},"source":["## Scenario\n","Consider an ad server that knows all of the information above about a user, except the education level. On the other hand, the ad network is managing ads that address a specific education level. For example, at any given time, the ad server has one ad that is targeting users with college education, one ad that is targeting users with elementary school education etc. If the target audience of the ad that is shown to a user matches the user's education level, there is a high probability of a click. If not, the probability of a click decreases gradually as the discrepancy between the target education level and the user's education level. In other words, the ad server is implicitly trying to predict users' education levels as close as possible."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"sNtGmziSOCzY"},"outputs":[],"source":["from collections import namedtuple\n","from numpy.random import uniform as U\n","import pandas as pd\n","import numpy as np\n","import io\n","import requests\n","from tensorflow import keras\n","from tensorflow.keras.layers import Dense, Dropout"]},{"cell_type":"markdown","metadata":{"id":"BN7P31s6Nlbg"},"source":["## Data\n","\n","We will use a dataset that was modified from the 1994 U.S. Census and adapt it to an online advertising setting. The dataset is known as \"Census Income Dataset\" and available at https://archive.ics.uci.edu/ml/datasets/Census+Income.\n","\n","In this dataset we use the following information on the individuals participated in the census: Age, work class, education, marital status, occupation, relationship, race, gender, work hours per week, native country, and income level.\n","\n","With that, let's discuss how to turn this data into an online advertising scenario."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"_J8jba_DOLre"},"outputs":[],"source":["url=\"https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.data\"\n","s=requests.get(url).content\n","names = ['age', \n"," 'workclass', \n"," 'fnlwgt', \n"," 'education',\n"," 'education_num',\n"," 'marital_status',\n"," 'occupation',\n"," 'relationship',\n"," 'race',\n"," 'gender',\n"," 'capital_gain',\n"," 'capital_loss',\n"," 'hours_per_week',\n"," 'native_country',\n"," 'income']\n","usecols = ['age', \n"," 'workclass', \n"," 'education',\n"," 'marital_status',\n"," 'occupation',\n"," 'relationship',\n"," 'race',\n"," 'gender',\n"," 'hours_per_week',\n"," 'native_country',\n"," 'income']\n","df_census = pd.read_csv(io.StringIO(s.decode('utf-8')), \n"," sep=',',\n"," skipinitialspace=True,\n"," names=names,\n"," header=None,\n"," usecols=usecols)"]},{"cell_type":"markdown","metadata":{"id":"6AMRPeLXOvZi"},"source":["Next, let's prepare the data set for our scenario."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"rMk4nUhLOfjl"},"outputs":[],"source":["# Cleanup\n","df_census = df_census.replace('?', np.nan).dropna()\n","edu_map = {'Preschool': 'Elementary',\n"," '1st-4th': 'Elementary',\n"," '5th-6th': 'Elementary',\n"," '7th-8th': 'Elementary',\n"," '9th': 'Middle',\n"," '10th': 'Middle',\n"," '11th': 'Middle',\n"," '12th': 'Middle',\n"," 'Some-college': 'Undergraduate',\n"," 'Bachelors': 'Undergraduate',\n"," 'Assoc-acdm': 'Undergraduate',\n"," 'Assoc-voc': 'Undergraduate',\n"," 'Prof-school': 'Graduate',\n"," 'Masters': 'Graduate',\n"," 'Doctorate': 'Graduate'}\n"," \n","for from_level, to_level in edu_map.items():\n"," df_census.education.replace(from_level, to_level, inplace=True)\n","\n","# Convert raw data to processed data\n","context_cols = [c for c in usecols if c != 'education']\n","df_data = pd.concat([pd.get_dummies(df_census[context_cols]),\n"," df_census['education']], axis=1)"]},{"cell_type":"markdown","metadata":{"id":"ntpHFmP0OxUm"},"source":["By doing this conversion at the beginning, we assume that we know all possible work class categories, native countries etc.\n","\n","That's it! We have the data ready. Next, we implement a logic to simulate an ad click based on the actual education level of the user and the education level the ad displayed targets."]},{"cell_type":"markdown","metadata":{"id":"gz8ZyCx-O1fs"},"source":["### Simulating ad clicks\n","In this example, the availability of ads is also random in addition to the ad clicks being stochastic. We need to come up with some logic to simulate this behavior. \n","\n","Let's start with determining the ad availability probabilities for each education category and implement the sampling of ads."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"siw_xBqYPFIp"},"outputs":[],"source":["def get_ad_inventory():\n"," ad_inv_prob = {'Elementary': 0.9, \n"," 'Middle': 0.7, \n"," 'HS-grad': 0.7, \n"," 'Undergraduate': 0.9, \n"," 'Graduate': 0.8}\n"," ad_inventory = []\n"," for level, prob in ad_inv_prob.items():\n"," if U() < prob:\n"," ad_inventory.append(level)\n"," # Make sure there are at least one ad\n"," if not ad_inventory:\n"," ad_inventory = get_ad_inventory()\n"," return ad_inventory"]},{"cell_type":"markdown","metadata":{"id":"SNW09VX-PHXY"},"source":["As mentioned above, the ad server will have at most one ad for each target group. We also ensure that there is at least one ad in the inventory.\n","\n","Then, we define a function to generate a click probabilistically, where the likelihood of a click increases to the degree that the user's education level and the ad's target match."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"tVd5ntyMPKw0"},"outputs":[],"source":["def get_ad_click_probs():\n"," base_prob = 0.8\n"," delta = 0.3\n"," ed_levels = {'Elementary': 1, \n"," 'Middle': 2, \n"," 'HS-grad': 3, \n"," 'Undergraduate': 4, \n"," 'Graduate': 5}\n"," ad_click_probs = {l1: {l2: max(0, base_prob - delta * abs(ed_levels[l1]- ed_levels[l2])) for l2 in ed_levels}\n"," for l1 in ed_levels}\n"," return ad_click_probs"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"GhJih_pyPNwl"},"outputs":[],"source":["def display_ad(ad_click_probs, user, ad):\n"," prob = ad_click_probs[ad][user['education']]\n"," click = 1 if U() < prob else 0\n"," return click"]},{"cell_type":"markdown","metadata":{"id":"lxEnZC7bPQtB"},"source":["So, when an ad is shown to a user, if the ad's target matches the user's education level, there will be chance of a click. This probability decreases by for each level of mismatch. For example, a person with a high school diploma has a chance to click on an ad that targets a user group of elementary school graduates (or college graduates). Note that this information is not known to the CB algorithm. It will be used only to simulate the clicks.\n","\n","We have the problem set up. Next, we turn to implementing a CB model."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Sj0E4CCQPYy0"},"outputs":[],"source":["def calc_regret(user, ad_inventory, ad_click_probs, ad_selected):\n"," this_p = 0\n"," max_p = 0\n"," for ad in ad_inventory:\n"," p = ad_click_probs[ad][user['education']]\n"," if ad == ad_selected:\n"," this_p = p\n"," if p > max_p:\n"," max_p = p\n"," regret = max_p - this_p\n"," return regret"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"vrwLopu9Pc9m"},"outputs":[],"source":["def get_model(n_input, dropout):\n"," inputs = keras.Input(shape=(n_input,))\n"," x = Dense(256, activation='relu')(inputs)\n"," if dropout > 0:\n"," x = Dropout(dropout)(x, training=True)\n"," x = Dense(256, activation='relu')(x)\n"," if dropout > 0:\n"," x = Dropout(dropout)(x, training=True)\n"," phat = Dense(1, activation='sigmoid')(x)\n"," model = keras.Model(inputs, phat)\n"," model.compile(loss=keras.losses.BinaryCrossentropy(),\n"," optimizer=keras.optimizers.Adam(),\n"," metrics=[keras.metrics.binary_accuracy])\n"," return model"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"XjfTZLr_Pb8A"},"outputs":[],"source":["def update_model(model, X, y):\n"," X = np.array(X)\n"," X = X.reshape((X.shape[0], X.shape[2]))\n"," y = np.array(y).reshape(-1)\n"," model.fit(X, y, epochs=10)\n"," return model"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"T6azuaN8PbJG"},"outputs":[],"source":["def ad_to_one_hot(ad):\n"," ed_levels = ['Elementary', \n"," 'Middle', \n"," 'HS-grad', \n"," 'Undergraduate', \n"," 'Graduate']\n"," ad_input = [0] * len(ed_levels)\n"," if ad in ed_levels:\n"," ad_input[ed_levels.index(ad)] = 1\n"," return ad_input"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"bt5SeP54PaaB"},"outputs":[],"source":["def select_ad(model, context, ad_inventory):\n"," selected_ad = None\n"," selected_x = None\n"," max_action_val = 0\n"," for ad in ad_inventory:\n"," ad_x = ad_to_one_hot(ad)\n"," x = np.array(context + ad_x).reshape((1, -1))\n"," action_val_pred = model.predict(x)[0][0]\n"," if action_val_pred >= max_action_val:\n"," selected_ad = ad\n"," selected_x = x\n"," max_action_val = action_val_pred\n"," return selected_ad, selected_x"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"e6e1-glpPZfw"},"outputs":[],"source":["def generate_user(df_data):\n"," user = df_data.sample(1)\n"," context = user.iloc[:, :-1].values.tolist()[0]\n"," return user.to_dict(orient='records')[0], context"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"background_save":true,"base_uri":"https://localhost:8080/"},"id":"lyq_uwq7PkQl","outputId":"35832751-06fc-471d-842e-29b5a4fb9876","collapsed":true},"outputs":[{"name":"stdout","output_type":"stream","text":["Trying with dropout: 0\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 0.8424 - binary_accuracy: 0.5600\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7275 - binary_accuracy: 0.5700\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6772 - binary_accuracy: 0.6280\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6462 - binary_accuracy: 0.6560\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6463 - binary_accuracy: 0.6440\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6442 - binary_accuracy: 0.6540\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6532 - binary_accuracy: 0.6560\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6247 - binary_accuracy: 0.6620\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6254 - binary_accuracy: 0.6800\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6935 - binary_accuracy: 0.6100\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7602 - binary_accuracy: 0.5100\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.8014 - binary_accuracy: 0.4860\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7363 - binary_accuracy: 0.5600\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6868 - binary_accuracy: 0.5860\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6865 - binary_accuracy: 0.5420\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6697 - binary_accuracy: 0.6020\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6742 - binary_accuracy: 0.5760\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6855 - binary_accuracy: 0.5700\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6727 - binary_accuracy: 0.5840\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6695 - binary_accuracy: 0.5940\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7112 - binary_accuracy: 0.4980\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7090 - binary_accuracy: 0.5460\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6822 - binary_accuracy: 0.5620\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6733 - binary_accuracy: 0.5980\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6872 - binary_accuracy: 0.5840\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6832 - binary_accuracy: 0.5740\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6683 - binary_accuracy: 0.6000\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7067 - binary_accuracy: 0.5480\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6641 - binary_accuracy: 0.6240\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6753 - binary_accuracy: 0.6000\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7047 - binary_accuracy: 0.5480\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6834 - binary_accuracy: 0.5920\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6927 - binary_accuracy: 0.5640\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6886 - binary_accuracy: 0.5700\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6585 - binary_accuracy: 0.6100\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6880 - binary_accuracy: 0.5840\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6924 - binary_accuracy: 0.5900\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6759 - binary_accuracy: 0.5760\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6593 - binary_accuracy: 0.6340\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6594 - binary_accuracy: 0.5740\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6748 - binary_accuracy: 0.6140\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6671 - binary_accuracy: 0.6380\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6687 - binary_accuracy: 0.6200\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6667 - binary_accuracy: 0.6220\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6570 - binary_accuracy: 0.6400\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6470 - binary_accuracy: 0.6320\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6573 - binary_accuracy: 0.6240\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6706 - binary_accuracy: 0.6380\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6372 - binary_accuracy: 0.6600\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6384 - binary_accuracy: 0.6500\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7047 - binary_accuracy: 0.5280\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6950 - binary_accuracy: 0.6000\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6773 - binary_accuracy: 0.5880\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6734 - binary_accuracy: 0.5920\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6644 - binary_accuracy: 0.5960\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6815 - binary_accuracy: 0.5880\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6714 - binary_accuracy: 0.5920\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6674 - binary_accuracy: 0.6000\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6491 - binary_accuracy: 0.6460\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6584 - binary_accuracy: 0.6120\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7192 - binary_accuracy: 0.5240\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6999 - binary_accuracy: 0.5240\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6960 - binary_accuracy: 0.5520\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6843 - binary_accuracy: 0.5440\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6735 - binary_accuracy: 0.5940\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6647 - binary_accuracy: 0.5920\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6557 - binary_accuracy: 0.5900\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6777 - binary_accuracy: 0.5760\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6630 - binary_accuracy: 0.5920\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6650 - binary_accuracy: 0.5980\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7114 - binary_accuracy: 0.5680\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7233 - binary_accuracy: 0.5420\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7212 - binary_accuracy: 0.4840\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7029 - binary_accuracy: 0.5300\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6793 - binary_accuracy: 0.5960\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6729 - binary_accuracy: 0.6140\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6804 - binary_accuracy: 0.5740\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6612 - binary_accuracy: 0.5900\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6541 - binary_accuracy: 0.6220\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6443 - binary_accuracy: 0.6120\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7017 - binary_accuracy: 0.5300\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6915 - binary_accuracy: 0.5300\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6861 - binary_accuracy: 0.5660\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6814 - binary_accuracy: 0.5720\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6816 - binary_accuracy: 0.5900\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6762 - binary_accuracy: 0.5940\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6685 - binary_accuracy: 0.5960\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6679 - binary_accuracy: 0.6160\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6638 - binary_accuracy: 0.6180\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6589 - binary_accuracy: 0.6040\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6787 - binary_accuracy: 0.6060\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6713 - binary_accuracy: 0.6220\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6632 - binary_accuracy: 0.6220\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6589 - binary_accuracy: 0.6240\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6574 - binary_accuracy: 0.6320\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6572 - binary_accuracy: 0.6160\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6503 - binary_accuracy: 0.6300\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6458 - binary_accuracy: 0.6380\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6434 - binary_accuracy: 0.6400\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6428 - binary_accuracy: 0.6400\n","Trying with dropout: 0.01\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 1.5034 - binary_accuracy: 0.5400\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7648 - binary_accuracy: 0.5500\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6662 - binary_accuracy: 0.5960\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7367 - binary_accuracy: 0.5400\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7202 - binary_accuracy: 0.5600\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6967 - binary_accuracy: 0.5980\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6763 - binary_accuracy: 0.5960\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7087 - binary_accuracy: 0.5440\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6692 - binary_accuracy: 0.6100\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7623 - binary_accuracy: 0.5580\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7131 - binary_accuracy: 0.5240\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7190 - binary_accuracy: 0.5260\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7115 - binary_accuracy: 0.5520\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6994 - binary_accuracy: 0.5300\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7226 - binary_accuracy: 0.5380\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7089 - binary_accuracy: 0.5420\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6813 - binary_accuracy: 0.5640\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6568 - binary_accuracy: 0.6100\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7108 - binary_accuracy: 0.5700\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7122 - binary_accuracy: 0.5660\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7689 - binary_accuracy: 0.5080\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7372 - binary_accuracy: 0.5600\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6954 - binary_accuracy: 0.5660\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6958 - binary_accuracy: 0.5720\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6695 - binary_accuracy: 0.5880\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6546 - binary_accuracy: 0.6360\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6767 - binary_accuracy: 0.5960\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6611 - binary_accuracy: 0.6040\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6645 - binary_accuracy: 0.6140\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7341 - binary_accuracy: 0.5540\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7322 - binary_accuracy: 0.5600\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7008 - binary_accuracy: 0.5800\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6948 - binary_accuracy: 0.5680\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6803 - binary_accuracy: 0.5780\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6858 - binary_accuracy: 0.5920\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6883 - binary_accuracy: 0.5560\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6837 - binary_accuracy: 0.5940\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6399 - binary_accuracy: 0.6280\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6560 - binary_accuracy: 0.6340\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6643 - binary_accuracy: 0.5980\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6871 - binary_accuracy: 0.5940\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6683 - binary_accuracy: 0.6340\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6588 - binary_accuracy: 0.6400\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6966 - binary_accuracy: 0.5480\n","Epoch 5/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6862 - binary_accuracy: 0.6120\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6529 - binary_accuracy: 0.6420\n","Epoch 7/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6500 - binary_accuracy: 0.6520\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6326 - binary_accuracy: 0.6520\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6464 - binary_accuracy: 0.6320\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6301 - binary_accuracy: 0.6540\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6894 - binary_accuracy: 0.5820\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6624 - binary_accuracy: 0.6040\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6543 - binary_accuracy: 0.6140\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6708 - binary_accuracy: 0.5960\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6855 - binary_accuracy: 0.5940\n","Epoch 6/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6892 - binary_accuracy: 0.6000\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6642 - binary_accuracy: 0.6040\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6752 - binary_accuracy: 0.5820\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6389 - binary_accuracy: 0.6220\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6285 - binary_accuracy: 0.6640\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6938 - binary_accuracy: 0.6000\n","Epoch 2/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6976 - binary_accuracy: 0.5820\n","Epoch 3/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6909 - binary_accuracy: 0.5880\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6727 - binary_accuracy: 0.5840\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6619 - binary_accuracy: 0.6220\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6533 - binary_accuracy: 0.6040\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6569 - binary_accuracy: 0.6200\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6566 - binary_accuracy: 0.6340\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6500 - binary_accuracy: 0.6320\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6476 - binary_accuracy: 0.6240\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6862 - binary_accuracy: 0.5920\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6774 - binary_accuracy: 0.6360\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6600 - binary_accuracy: 0.6060\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6576 - binary_accuracy: 0.6200\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6465 - binary_accuracy: 0.6420\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6671 - binary_accuracy: 0.5920\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6541 - binary_accuracy: 0.6160\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6515 - binary_accuracy: 0.6220\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6366 - binary_accuracy: 0.6400\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6327 - binary_accuracy: 0.6480\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6963 - binary_accuracy: 0.5540\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6808 - binary_accuracy: 0.5780\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6785 - binary_accuracy: 0.5780\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6676 - binary_accuracy: 0.6020\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6620 - binary_accuracy: 0.6040\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6663 - binary_accuracy: 0.5780\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6683 - binary_accuracy: 0.5800\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6672 - binary_accuracy: 0.5840\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6665 - binary_accuracy: 0.5800\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6581 - binary_accuracy: 0.6000\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.7103 - binary_accuracy: 0.5280\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6982 - binary_accuracy: 0.5580\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6724 - binary_accuracy: 0.5960\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6622 - binary_accuracy: 0.6160\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6637 - binary_accuracy: 0.6060\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6502 - binary_accuracy: 0.6440\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6495 - binary_accuracy: 0.6200\n","Epoch 8/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6643 - binary_accuracy: 0.5860\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6513 - binary_accuracy: 0.6220\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6470 - binary_accuracy: 0.6320\n","Trying with dropout: 0.05\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 0.8016 - binary_accuracy: 0.5700\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.8029 - binary_accuracy: 0.5700\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7046 - binary_accuracy: 0.6120\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6696 - binary_accuracy: 0.6200\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6876 - binary_accuracy: 0.5980\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6445 - binary_accuracy: 0.6280\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6531 - binary_accuracy: 0.6180\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6353 - binary_accuracy: 0.6460\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6844 - binary_accuracy: 0.6080\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6466 - binary_accuracy: 0.6560\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7171 - binary_accuracy: 0.5400\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7333 - binary_accuracy: 0.5060\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6868 - binary_accuracy: 0.5580\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6944 - binary_accuracy: 0.5540\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6704 - binary_accuracy: 0.5760\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6894 - binary_accuracy: 0.5620\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6757 - binary_accuracy: 0.5660\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6497 - binary_accuracy: 0.6080\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6460 - binary_accuracy: 0.6240\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6662 - binary_accuracy: 0.5700\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7188 - binary_accuracy: 0.5440\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6956 - binary_accuracy: 0.5680\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7030 - binary_accuracy: 0.5560\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6991 - binary_accuracy: 0.5640\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6999 - binary_accuracy: 0.5540\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6921 - binary_accuracy: 0.5700\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6708 - binary_accuracy: 0.5980\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6667 - binary_accuracy: 0.6000\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6702 - binary_accuracy: 0.6000\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6865 - binary_accuracy: 0.5700\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7352 - binary_accuracy: 0.5580\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6909 - binary_accuracy: 0.5500\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7130 - binary_accuracy: 0.5740\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6912 - binary_accuracy: 0.5460\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6799 - binary_accuracy: 0.5900\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6799 - binary_accuracy: 0.5980\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6567 - binary_accuracy: 0.6140\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6618 - binary_accuracy: 0.6120\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6665 - binary_accuracy: 0.6060\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7091 - binary_accuracy: 0.5680\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7086 - binary_accuracy: 0.5400\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6824 - binary_accuracy: 0.5800\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6741 - binary_accuracy: 0.6040\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6949 - binary_accuracy: 0.5620\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6811 - binary_accuracy: 0.5780\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6588 - binary_accuracy: 0.5980\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7057 - binary_accuracy: 0.5560\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7061 - binary_accuracy: 0.5580\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6653 - binary_accuracy: 0.6200\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6638 - binary_accuracy: 0.6020\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6878 - binary_accuracy: 0.5820\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6770 - binary_accuracy: 0.5720\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6869 - binary_accuracy: 0.5700\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6680 - binary_accuracy: 0.5860\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6654 - binary_accuracy: 0.5980\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6540 - binary_accuracy: 0.6400\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6570 - binary_accuracy: 0.6220\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6701 - binary_accuracy: 0.6120\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6388 - binary_accuracy: 0.6360\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6397 - binary_accuracy: 0.6520\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7046 - binary_accuracy: 0.5380\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6993 - binary_accuracy: 0.5480\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7059 - binary_accuracy: 0.5420\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6962 - binary_accuracy: 0.5480\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6715 - binary_accuracy: 0.5800\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6710 - binary_accuracy: 0.5800\n","Epoch 7/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6701 - binary_accuracy: 0.5800\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6563 - binary_accuracy: 0.6020\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6519 - binary_accuracy: 0.6060\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6578 - binary_accuracy: 0.6160\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6873 - binary_accuracy: 0.5660\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6753 - binary_accuracy: 0.5680\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6653 - binary_accuracy: 0.5880\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6629 - binary_accuracy: 0.5880\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6724 - binary_accuracy: 0.5780\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6657 - binary_accuracy: 0.6000\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6574 - binary_accuracy: 0.5760\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6437 - binary_accuracy: 0.6240\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6496 - binary_accuracy: 0.6100\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6472 - binary_accuracy: 0.6380\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7081 - binary_accuracy: 0.5620\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6928 - binary_accuracy: 0.5540\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6886 - binary_accuracy: 0.5860\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6752 - binary_accuracy: 0.5760\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6925 - binary_accuracy: 0.5640\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6689 - binary_accuracy: 0.5700\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6720 - binary_accuracy: 0.5740\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6671 - binary_accuracy: 0.5940\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6657 - binary_accuracy: 0.5920\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6640 - binary_accuracy: 0.6080\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6828 - binary_accuracy: 0.5960\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6700 - binary_accuracy: 0.5820\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6599 - binary_accuracy: 0.6040\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6520 - binary_accuracy: 0.6240\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6579 - binary_accuracy: 0.6220\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6452 - binary_accuracy: 0.6380\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6448 - binary_accuracy: 0.6220\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6395 - binary_accuracy: 0.6420\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6481 - binary_accuracy: 0.6020\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6379 - binary_accuracy: 0.6520\n","Trying with dropout: 0.1\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 0.9884 - binary_accuracy: 0.4940\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7877 - binary_accuracy: 0.5840\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6771 - binary_accuracy: 0.5860\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6991 - binary_accuracy: 0.5820\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6906 - binary_accuracy: 0.5980\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7296 - binary_accuracy: 0.5380\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6380 - binary_accuracy: 0.6400\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6671 - binary_accuracy: 0.6140\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6816 - binary_accuracy: 0.6280\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6397 - binary_accuracy: 0.6280\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7042 - binary_accuracy: 0.5280\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7512 - binary_accuracy: 0.5220\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6795 - binary_accuracy: 0.5680\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6938 - binary_accuracy: 0.5640\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7466 - binary_accuracy: 0.5640\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6878 - binary_accuracy: 0.5840\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6762 - binary_accuracy: 0.5700\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6555 - binary_accuracy: 0.6160\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6798 - binary_accuracy: 0.5720\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6696 - binary_accuracy: 0.5760\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7740 - binary_accuracy: 0.5060\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.8210 - binary_accuracy: 0.5000\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6919 - binary_accuracy: 0.5800\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7914 - binary_accuracy: 0.5460\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6909 - binary_accuracy: 0.5520\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6774 - binary_accuracy: 0.5820\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6563 - binary_accuracy: 0.6180\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6597 - binary_accuracy: 0.6000\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6667 - binary_accuracy: 0.5960\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6701 - binary_accuracy: 0.5840\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7469 - binary_accuracy: 0.5380\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6965 - binary_accuracy: 0.5600\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6811 - binary_accuracy: 0.5820\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6811 - binary_accuracy: 0.6100\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6682 - binary_accuracy: 0.5940\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6606 - binary_accuracy: 0.6100\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6767 - binary_accuracy: 0.5900\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6800 - binary_accuracy: 0.6200\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6595 - binary_accuracy: 0.6200\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7340 - binary_accuracy: 0.5320\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7531 - binary_accuracy: 0.5240\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6901 - binary_accuracy: 0.5520\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6713 - binary_accuracy: 0.5980\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6775 - binary_accuracy: 0.6020\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6795 - binary_accuracy: 0.5800\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6597 - binary_accuracy: 0.6020\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6707 - binary_accuracy: 0.6060\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6644 - binary_accuracy: 0.5860\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6541 - binary_accuracy: 0.6020\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6560 - binary_accuracy: 0.6240\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6898 - binary_accuracy: 0.5780\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6807 - binary_accuracy: 0.5960\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6743 - binary_accuracy: 0.5800\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6842 - binary_accuracy: 0.6040\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6664 - binary_accuracy: 0.6280\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6678 - binary_accuracy: 0.6260\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6553 - binary_accuracy: 0.5960\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6554 - binary_accuracy: 0.6140\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6459 - binary_accuracy: 0.6400\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6478 - binary_accuracy: 0.6180\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7144 - binary_accuracy: 0.5320\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7065 - binary_accuracy: 0.5120\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7025 - binary_accuracy: 0.5760\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6843 - binary_accuracy: 0.5600\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6799 - binary_accuracy: 0.5480\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6753 - binary_accuracy: 0.5860\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6646 - binary_accuracy: 0.5980\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6763 - binary_accuracy: 0.5600\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6714 - binary_accuracy: 0.5880\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6625 - binary_accuracy: 0.5960\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7030 - binary_accuracy: 0.5580\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7124 - binary_accuracy: 0.5160\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6750 - binary_accuracy: 0.5620\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6727 - binary_accuracy: 0.5720\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6641 - binary_accuracy: 0.6240\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6603 - binary_accuracy: 0.6060\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6532 - binary_accuracy: 0.6020\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6373 - binary_accuracy: 0.6100\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6415 - binary_accuracy: 0.6200\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6376 - binary_accuracy: 0.6060\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7164 - binary_accuracy: 0.5260\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6955 - binary_accuracy: 0.5740\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6788 - binary_accuracy: 0.5720\n","Epoch 4/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6748 - binary_accuracy: 0.5860\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6714 - binary_accuracy: 0.5860\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6652 - binary_accuracy: 0.6100\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6619 - binary_accuracy: 0.6060\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6608 - binary_accuracy: 0.6120\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6599 - binary_accuracy: 0.5980\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6496 - binary_accuracy: 0.6240\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6688 - binary_accuracy: 0.6060\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6715 - binary_accuracy: 0.5860\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6571 - binary_accuracy: 0.6040\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6531 - binary_accuracy: 0.6360\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6430 - binary_accuracy: 0.6220\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6498 - binary_accuracy: 0.6220\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6383 - binary_accuracy: 0.6360\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6485 - binary_accuracy: 0.6140\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6403 - binary_accuracy: 0.6400\n","Epoch 10/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6414 - binary_accuracy: 0.6420\n","Trying with dropout: 0.2\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 1.4584 - binary_accuracy: 0.5060\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.9217 - binary_accuracy: 0.4840\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7640 - binary_accuracy: 0.5220\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6963 - binary_accuracy: 0.5580\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6786 - binary_accuracy: 0.5840\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6725 - binary_accuracy: 0.6020\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6635 - binary_accuracy: 0.5920\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7161 - binary_accuracy: 0.5880\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6580 - binary_accuracy: 0.5840\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6423 - binary_accuracy: 0.6340\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7148 - binary_accuracy: 0.5420\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7069 - binary_accuracy: 0.5600\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7121 - binary_accuracy: 0.5400\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7156 - binary_accuracy: 0.5740\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7200 - binary_accuracy: 0.5620\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7199 - binary_accuracy: 0.5560\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6778 - binary_accuracy: 0.6040\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6632 - binary_accuracy: 0.5940\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6599 - binary_accuracy: 0.6260\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6367 - binary_accuracy: 0.6200\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6940 - binary_accuracy: 0.5540\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7129 - binary_accuracy: 0.5500\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6761 - binary_accuracy: 0.5860\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7081 - binary_accuracy: 0.5340\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6923 - binary_accuracy: 0.5780\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7198 - binary_accuracy: 0.5460\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7137 - binary_accuracy: 0.5680\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6967 - binary_accuracy: 0.5940\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6552 - binary_accuracy: 0.6180\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6607 - binary_accuracy: 0.6340\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6966 - binary_accuracy: 0.5620\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7105 - binary_accuracy: 0.5800\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7163 - binary_accuracy: 0.5920\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6696 - binary_accuracy: 0.5840\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6504 - binary_accuracy: 0.6300\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6688 - binary_accuracy: 0.6080\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6640 - binary_accuracy: 0.6340\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6558 - binary_accuracy: 0.6260\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6435 - binary_accuracy: 0.6400\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6349 - binary_accuracy: 0.6200\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7073 - binary_accuracy: 0.5900\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6805 - binary_accuracy: 0.6060\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6748 - binary_accuracy: 0.6040\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6645 - binary_accuracy: 0.6060\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6674 - binary_accuracy: 0.6140\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6551 - binary_accuracy: 0.6240\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6527 - binary_accuracy: 0.6160\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6659 - binary_accuracy: 0.6040\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6660 - binary_accuracy: 0.6160\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6415 - binary_accuracy: 0.6360\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6946 - binary_accuracy: 0.5920\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6908 - binary_accuracy: 0.5840\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6937 - binary_accuracy: 0.5620\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6862 - binary_accuracy: 0.5760\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6970 - binary_accuracy: 0.5740\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7177 - binary_accuracy: 0.5600\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6763 - binary_accuracy: 0.5820\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6582 - binary_accuracy: 0.6120\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6620 - binary_accuracy: 0.6040\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6574 - binary_accuracy: 0.6160\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6969 - binary_accuracy: 0.5600\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6870 - binary_accuracy: 0.5700\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6798 - binary_accuracy: 0.5680\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6741 - binary_accuracy: 0.6180\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6715 - binary_accuracy: 0.5960\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6594 - binary_accuracy: 0.6280\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6649 - binary_accuracy: 0.6060\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6922 - binary_accuracy: 0.5440\n","Epoch 9/10\n","16/16 [==============================] - 0s 2ms/step - loss: 0.6498 - binary_accuracy: 0.6320\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6347 - binary_accuracy: 0.6480\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6784 - binary_accuracy: 0.5860\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6692 - binary_accuracy: 0.5840\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7043 - binary_accuracy: 0.5880\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6630 - binary_accuracy: 0.5920\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6511 - binary_accuracy: 0.6280\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6511 - binary_accuracy: 0.6220\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6392 - binary_accuracy: 0.6500\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6514 - binary_accuracy: 0.5940\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6353 - binary_accuracy: 0.6320\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6330 - binary_accuracy: 0.6340\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7154 - binary_accuracy: 0.5240\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7046 - binary_accuracy: 0.5360\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6845 - binary_accuracy: 0.5620\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6823 - binary_accuracy: 0.5660\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6804 - binary_accuracy: 0.5740\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6737 - binary_accuracy: 0.5640\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6827 - binary_accuracy: 0.5660\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6837 - binary_accuracy: 0.5500\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6638 - binary_accuracy: 0.5860\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6593 - binary_accuracy: 0.6000\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6801 - binary_accuracy: 0.5900\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6607 - binary_accuracy: 0.6200\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6633 - binary_accuracy: 0.6220\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6600 - binary_accuracy: 0.6100\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6486 - binary_accuracy: 0.6380\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6485 - binary_accuracy: 0.6280\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6332 - binary_accuracy: 0.6400\n","Epoch 8/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6357 - binary_accuracy: 0.6360\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6339 - binary_accuracy: 0.6380\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6239 - binary_accuracy: 0.6560\n","Trying with dropout: 0.4\n","# of impressions: 0\n","# of impressions: 20\n","# of impressions: 40\n","# of impressions: 60\n","# of impressions: 80\n","# of impressions: 100\n","# of impressions: 120\n","# of impressions: 140\n","# of impressions: 160\n","# of impressions: 180\n","# of impressions: 200\n","# of impressions: 220\n","# of impressions: 240\n","# of impressions: 260\n","# of impressions: 280\n","# of impressions: 300\n","# of impressions: 320\n","# of impressions: 340\n","# of impressions: 360\n","# of impressions: 380\n","# of impressions: 400\n","# of impressions: 420\n","# of impressions: 440\n","# of impressions: 460\n","# of impressions: 480\n","Updating the model at 500\n","Epoch 1/10\n","16/16 [==============================] - 1s 3ms/step - loss: 1.2578 - binary_accuracy: 0.5080\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.8418 - binary_accuracy: 0.5180\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.8006 - binary_accuracy: 0.4920\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7543 - binary_accuracy: 0.5360\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6835 - binary_accuracy: 0.5580\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6767 - binary_accuracy: 0.6060\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6759 - binary_accuracy: 0.5760\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6739 - binary_accuracy: 0.5960\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6969 - binary_accuracy: 0.5740\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6963 - binary_accuracy: 0.6100\n","# of impressions: 500\n","# of impressions: 520\n","# of impressions: 540\n","# of impressions: 560\n","# of impressions: 580\n","# of impressions: 600\n","# of impressions: 620\n","# of impressions: 640\n","# of impressions: 660\n","# of impressions: 680\n","# of impressions: 700\n","# of impressions: 720\n","# of impressions: 740\n","# of impressions: 760\n","# of impressions: 780\n","# of impressions: 800\n","# of impressions: 820\n","# of impressions: 840\n","# of impressions: 860\n","# of impressions: 880\n","# of impressions: 900\n","# of impressions: 920\n","# of impressions: 940\n","# of impressions: 960\n","# of impressions: 980\n","Updating the model at 1000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7266 - binary_accuracy: 0.5360\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7078 - binary_accuracy: 0.5380\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7576 - binary_accuracy: 0.5300\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7017 - binary_accuracy: 0.5420\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6974 - binary_accuracy: 0.5380\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7270 - binary_accuracy: 0.5320\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7037 - binary_accuracy: 0.5580\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6529 - binary_accuracy: 0.6100\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6757 - binary_accuracy: 0.5880\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6808 - binary_accuracy: 0.5840\n","# of impressions: 1000\n","# of impressions: 1020\n","# of impressions: 1040\n","# of impressions: 1060\n","# of impressions: 1080\n","# of impressions: 1100\n","# of impressions: 1120\n","# of impressions: 1140\n","# of impressions: 1160\n","# of impressions: 1180\n","# of impressions: 1200\n","# of impressions: 1220\n","# of impressions: 1240\n","# of impressions: 1260\n","# of impressions: 1280\n","# of impressions: 1300\n","# of impressions: 1320\n","# of impressions: 1340\n","# of impressions: 1360\n","# of impressions: 1380\n","# of impressions: 1400\n","# of impressions: 1420\n","# of impressions: 1440\n","# of impressions: 1460\n","# of impressions: 1480\n","Updating the model at 1500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7177 - binary_accuracy: 0.5220\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7246 - binary_accuracy: 0.5480\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7166 - binary_accuracy: 0.5480\n","Epoch 4/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6939 - binary_accuracy: 0.5580\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6732 - binary_accuracy: 0.6100\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7052 - binary_accuracy: 0.5720\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6656 - binary_accuracy: 0.6000\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6968 - binary_accuracy: 0.5680\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6641 - binary_accuracy: 0.6140\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6701 - binary_accuracy: 0.5980\n","# of impressions: 1500\n","# of impressions: 1520\n","# of impressions: 1540\n","# of impressions: 1560\n","# of impressions: 1580\n","# of impressions: 1600\n","# of impressions: 1620\n","# of impressions: 1640\n","# of impressions: 1660\n","# of impressions: 1680\n","# of impressions: 1700\n","# of impressions: 1720\n","# of impressions: 1740\n","# of impressions: 1760\n","# of impressions: 1780\n","# of impressions: 1800\n","# of impressions: 1820\n","# of impressions: 1840\n","# of impressions: 1860\n","# of impressions: 1880\n","# of impressions: 1900\n","# of impressions: 1920\n","# of impressions: 1940\n","# of impressions: 1960\n","# of impressions: 1980\n","Updating the model at 2000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7108 - binary_accuracy: 0.5920\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6958 - binary_accuracy: 0.5560\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7078 - binary_accuracy: 0.5560\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6714 - binary_accuracy: 0.5880\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6547 - binary_accuracy: 0.6200\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6507 - binary_accuracy: 0.6100\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6550 - binary_accuracy: 0.6120\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6668 - binary_accuracy: 0.6420\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6764 - binary_accuracy: 0.6020\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6449 - binary_accuracy: 0.6220\n","# of impressions: 2000\n","# of impressions: 2020\n","# of impressions: 2040\n","# of impressions: 2060\n","# of impressions: 2080\n","# of impressions: 2100\n","# of impressions: 2120\n","# of impressions: 2140\n","# of impressions: 2160\n","# of impressions: 2180\n","# of impressions: 2200\n","# of impressions: 2220\n","# of impressions: 2240\n","# of impressions: 2260\n","# of impressions: 2280\n","# of impressions: 2300\n","# of impressions: 2320\n","# of impressions: 2340\n","# of impressions: 2360\n","# of impressions: 2380\n","# of impressions: 2400\n","# of impressions: 2420\n","# of impressions: 2440\n","# of impressions: 2460\n","# of impressions: 2480\n","Updating the model at 2500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7030 - binary_accuracy: 0.5680\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6916 - binary_accuracy: 0.5800\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6891 - binary_accuracy: 0.5820\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6771 - binary_accuracy: 0.5980\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6717 - binary_accuracy: 0.5780\n","Epoch 6/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6956 - binary_accuracy: 0.5660\n","Epoch 7/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6649 - binary_accuracy: 0.6200\n","Epoch 8/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6590 - binary_accuracy: 0.6040\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6549 - binary_accuracy: 0.6180\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6528 - binary_accuracy: 0.6200\n","# of impressions: 2500\n","# of impressions: 2520\n","# of impressions: 2540\n","# of impressions: 2560\n","# of impressions: 2580\n","# of impressions: 2600\n","# of impressions: 2620\n","# of impressions: 2640\n","# of impressions: 2660\n","# of impressions: 2680\n","# of impressions: 2700\n","# of impressions: 2720\n","# of impressions: 2740\n","# of impressions: 2760\n","# of impressions: 2780\n","# of impressions: 2800\n","# of impressions: 2820\n","# of impressions: 2840\n","# of impressions: 2860\n","# of impressions: 2880\n","# of impressions: 2900\n","# of impressions: 2920\n","# of impressions: 2940\n","# of impressions: 2960\n","# of impressions: 2980\n","Updating the model at 3000\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6941 - binary_accuracy: 0.5740\n","Epoch 2/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.7142 - binary_accuracy: 0.5320\n","Epoch 3/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6967 - binary_accuracy: 0.5560\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7075 - binary_accuracy: 0.5340\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6652 - binary_accuracy: 0.6220\n","Epoch 6/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6733 - binary_accuracy: 0.5860\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6597 - binary_accuracy: 0.6160\n","Epoch 8/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6649 - binary_accuracy: 0.6080\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6575 - binary_accuracy: 0.6100\n","Epoch 10/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6543 - binary_accuracy: 0.6220\n","# of impressions: 3000\n","# of impressions: 3020\n","# of impressions: 3040\n","# of impressions: 3060\n","# of impressions: 3080\n","# of impressions: 3100\n","# of impressions: 3120\n","# of impressions: 3140\n","# of impressions: 3160\n","# of impressions: 3180\n","# of impressions: 3200\n","# of impressions: 3220\n","# of impressions: 3240\n","# of impressions: 3260\n","# of impressions: 3280\n","# of impressions: 3300\n","# of impressions: 3320\n","# of impressions: 3340\n","# of impressions: 3360\n","# of impressions: 3380\n","# of impressions: 3400\n","# of impressions: 3420\n","# of impressions: 3440\n","# of impressions: 3460\n","# of impressions: 3480\n","Updating the model at 3500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7185 - binary_accuracy: 0.5280\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6977 - binary_accuracy: 0.5580\n","Epoch 3/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6851 - binary_accuracy: 0.5780\n","Epoch 4/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6784 - binary_accuracy: 0.5920\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6674 - binary_accuracy: 0.6120\n","Epoch 6/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6563 - binary_accuracy: 0.6240\n","Epoch 7/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6550 - binary_accuracy: 0.6260\n","Epoch 8/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6576 - binary_accuracy: 0.6100\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6577 - binary_accuracy: 0.6100\n","Epoch 10/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6429 - binary_accuracy: 0.6440\n","# of impressions: 3500\n","# of impressions: 3520\n","# of impressions: 3540\n","# of impressions: 3560\n","# of impressions: 3580\n","# of impressions: 3600\n","# of impressions: 3620\n","# of impressions: 3640\n","# of impressions: 3660\n","# of impressions: 3680\n","# of impressions: 3700\n","# of impressions: 3720\n","# of impressions: 3740\n","# of impressions: 3760\n","# of impressions: 3780\n","# of impressions: 3800\n","# of impressions: 3820\n","# of impressions: 3840\n","# of impressions: 3860\n","# of impressions: 3880\n","# of impressions: 3900\n","# of impressions: 3920\n","# of impressions: 3940\n","# of impressions: 3960\n","# of impressions: 3980\n","Updating the model at 4000\n","Epoch 1/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6994 - binary_accuracy: 0.5740\n","Epoch 2/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6851 - binary_accuracy: 0.5780\n","Epoch 3/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6551 - binary_accuracy: 0.6000\n","Epoch 4/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6564 - binary_accuracy: 0.6300\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6596 - binary_accuracy: 0.5980\n","Epoch 6/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6509 - binary_accuracy: 0.6260\n","Epoch 7/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6498 - binary_accuracy: 0.6120\n","Epoch 8/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6359 - binary_accuracy: 0.6360\n","Epoch 9/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6306 - binary_accuracy: 0.6360\n","Epoch 10/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6317 - binary_accuracy: 0.6500\n","# of impressions: 4000\n","# of impressions: 4020\n","# of impressions: 4040\n","# of impressions: 4060\n","# of impressions: 4080\n","# of impressions: 4100\n","# of impressions: 4120\n","# of impressions: 4140\n","# of impressions: 4160\n","# of impressions: 4180\n","# of impressions: 4200\n","# of impressions: 4220\n","# of impressions: 4240\n","# of impressions: 4260\n","# of impressions: 4280\n","# of impressions: 4300\n","# of impressions: 4320\n","# of impressions: 4340\n","# of impressions: 4360\n","# of impressions: 4380\n","# of impressions: 4400\n","# of impressions: 4420\n","# of impressions: 4440\n","# of impressions: 4460\n","# of impressions: 4480\n","Updating the model at 4500\n","Epoch 1/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.7003 - binary_accuracy: 0.5380\n","Epoch 2/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6897 - binary_accuracy: 0.5540\n","Epoch 3/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6833 - binary_accuracy: 0.5660\n","Epoch 4/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6748 - binary_accuracy: 0.5700\n","Epoch 5/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6815 - binary_accuracy: 0.5760\n","Epoch 6/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6805 - binary_accuracy: 0.5660\n","Epoch 7/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6892 - binary_accuracy: 0.5640\n","Epoch 8/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6758 - binary_accuracy: 0.5960\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6561 - binary_accuracy: 0.6180\n","Epoch 10/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6594 - binary_accuracy: 0.6240\n","# of impressions: 4500\n","# of impressions: 4520\n","# of impressions: 4540\n","# of impressions: 4560\n","# of impressions: 4580\n","# of impressions: 4600\n","# of impressions: 4620\n","# of impressions: 4640\n","# of impressions: 4660\n","# of impressions: 4680\n","# of impressions: 4700\n","# of impressions: 4720\n","# of impressions: 4740\n","# of impressions: 4760\n","# of impressions: 4780\n","# of impressions: 4800\n","# of impressions: 4820\n","# of impressions: 4840\n","# of impressions: 4860\n","# of impressions: 4880\n","# of impressions: 4900\n","# of impressions: 4920\n","# of impressions: 4940\n","# of impressions: 4960\n","# of impressions: 4980\n","Updating the model at 5000\n","Epoch 1/10\n","16/16 [==============================] - 0s 5ms/step - loss: 0.6957 - binary_accuracy: 0.5700\n","Epoch 2/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6753 - binary_accuracy: 0.6080\n","Epoch 3/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6657 - binary_accuracy: 0.6060\n","Epoch 4/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6588 - binary_accuracy: 0.6260\n","Epoch 5/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6611 - binary_accuracy: 0.6180\n","Epoch 6/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6545 - binary_accuracy: 0.6160\n","Epoch 7/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6479 - binary_accuracy: 0.6360\n","Epoch 8/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6563 - binary_accuracy: 0.6120\n","Epoch 9/10\n","16/16 [==============================] - 0s 4ms/step - loss: 0.6465 - binary_accuracy: 0.6320\n","Epoch 10/10\n","16/16 [==============================] - 0s 3ms/step - loss: 0.6458 - binary_accuracy: 0.6240\n"]}],"source":["#collapse-hide\n","ad_click_probs = get_ad_click_probs()\n","df_cbandits = pd.DataFrame()\n","dropout_levels = [0, 0.01, 0.05, 0.1, 0.2, 0.4]\n","\n","for d in dropout_levels:\n"," print(\"Trying with dropout:\", d)\n"," np.random.seed(0)\n"," context_n = df_data.shape[1] - 1\n"," ad_input_n = df_data.education.nunique()\n"," model = get_model(context_n + ad_input_n, 0.01)\n"," X = []\n"," y = []\n"," regret_vec = []\n"," total_regret = 0\n"," for i in range(5000):\n"," if i % 20 == 0:\n"," print(\"# of impressions:\", i)\n"," user, context = generate_user(df_data)\n"," ad_inventory = get_ad_inventory()\n"," ad, x = select_ad(model, context, ad_inventory)\n"," click = display_ad(ad_click_probs, user, ad)\n"," regret = calc_regret(user, ad_inventory, ad_click_probs, ad)\n"," total_regret += regret\n"," regret_vec.append(total_regret)\n"," X.append(x)\n"," y.append(click)\n"," if (i + 1) % 500 == 0:\n"," print('Updating the model at', i+1)\n"," model = update_model(model, X, y)\n"," X = []\n"," y = []\n"," \n"," df_cbandits['dropout: '+str(d)] = regret_vec"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Vf7v4dQgPoc5"},"outputs":[],"source":["df_cbandits.iplot(dash = ['dash', 'solid', 'dashdot', \n"," 'dot', 'longdash', 'longdashdot'],\n"," xTitle='Impressions', \n"," yTitle='Cumulative Regret')"]},{"cell_type":"markdown","metadata":{"id":"0u56gxY-PsPo"},"source":["![image.png](data:image/png;base64,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)"]},{"cell_type":"markdown","metadata":{"id":"amgt8394PtRG"},"source":["*Comparison of cumulative regret with various dropout rates*"]},{"cell_type":"markdown","metadata":{"id":"Kbi4wGKXP1oV"},"source":["The results on the above figure show that our bandit models learn after some observations how to select the ads given the user characteristics. As various dropout rates have led to different algorithm performances, an important question again becomes how to select the dropout rate. One obvious answer is to try different rates over time identify what works the best in similar online advertising problems. This approach usually works if the business has to solve similar problems again and again over a long time period. A better approach though is to learn the optimal dropout rate."]}],"metadata":{"accelerator":"GPU","colab":{"collapsed_sections":[],"name":"2022-01-24-contextual-mab.ipynb","provenance":[{"file_id":"https://github.com/recohut/nbs/blob/main/raw/T273396%20%7C%20Contextual%20online%20advertising%20with%20user%20data%20from%20U.S.%20Census.ipynb","timestamp":1644668852780}],"toc_visible":true,"authorship_tag":"ABX9TyM5xF93WDs5M5XRE3uy37c0"},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"}},"nbformat":4,"nbformat_minor":0}