Newer
Older
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"LINEAR.ipynb","provenance":[],"collapsed_sections":[],"mount_file_id":"1BWAT5BW2l5voWNsVvG4JcopvGHq_F820","authorship_tag":"ABX9TyN2NRmoPdME0YfjzA55mRuL"},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"code","metadata":{"id":"5zn9maoCzLyT","colab_type":"code","outputId":"d4f591f2-0b8b-4c12-8d74-f625c89806a7","executionInfo":{"status":"ok","timestamp":1582148253078,"user_tz":0,"elapsed":743,"user":{"displayName":"Andy Wang","photoUrl":"","userId":"02776860930356410397"}},"colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["import torch\n","import torch.nn as nn\n","import numpy as np\n","from torchvision import datasets, transforms\n","\n","import seaborn as sns\n","import matplotlib.pyplot as plt\n","\n","GPU = True\n","device_idx = 0\n","if GPU:\n"," device = torch.device(\"cuda:\" + str(device_idx) if torch.cuda.is_available() else \"cpu\")\n","else:\n"," device = torch.device(\"cpu\")\n","print(device)\n","\n","# Set default dtype for model weights\n","torch.set_default_dtype(torch.double)"],"execution_count":15,"outputs":[{"output_type":"stream","text":["cuda:0\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"Lhv5MZ3JyH59","colab_type":"code","colab":{}},"source":["# Load the data\n","import os\n","os.chdir(\"/content/drive/My Drive/Colab Notebooks/NLP/coursework\")\n","import pickle\n","with open(\"en_sentence_emb.pk\", \"rb\") as f:\n"," en_sentences_vectors = pickle.load(f)\n","with open(\"de_sentence_emb.pk\", \"rb\") as f:\n"," de_sentences_vectors = pickle.load(f)\n","with open(\"scores.pk\", \"rb\") as f:\n"," scores = pickle.load(f)"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"brKjYFeuV94x","colab_type":"code","colab":{}},"source":["sentences = torch.cat((torch.tensor(en_sentences_vectors).double(), torch.tensor(de_sentences_vectors).double()), dim=(-1))"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"q43POV9QytX_","colab_type":"code","colab":{}},"source":["class Sentences(torch.utils.data.Dataset):\n"," def __init__(self, sentences, scores):\n"," super(Sentences, self).__init__()\n"," self.sentences = sentences.double().to(device)\n"," self.scores = torch.tensor(scores, device=device, dtype=torch.double)\n","\n"," def __len__(self):\n"," return len(self.scores)\n","\n"," def __getitem__(self, idx):\n"," if torch.is_tensor(idx):\n"," idx = idx.tolist()\n"," \n"," sen, scores = self.sentences[idx], self.scores[idx]\n"," return sen, scores"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"KOd75PS_fGoA","colab_type":"code","colab":{}},"source":["# Create train and test sets\n","train_pct = 0.8\n","train_len = int(train_pct * len(sentences))\n","\n","dataset = Sentences(sentences, scores)\n","\n","train, test = torch.utils.data.random_split(dataset, [train_len, len(sentences) - train_len])"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"fkw5SZx1fTZZ","colab_type":"code","colab":{}},"source":["# Create Dataloaders\n","batch_size = 32\n","\n","loader_train = torch.utils.data.DataLoader(train, batch_size=batch_size, shuffle=True)\n","loader_test = torch.utils.data.DataLoader(test, batch_size=batch_size, shuffle=False)"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"O7keA4QQ0x0t","colab_type":"code","colab":{}},"source":["# https://stackoverflow.com/questions/53010465/bidirectional-lstm-output-question-in-pytorch\n","# Hyper Parameters\n","embed_dim = 300\n","reg_hid_dim1 = 500\n","reg_hid_dim2 = 400\n","reg_hid_dim3 = 300\n","reg_hid_dim4 = 200\n","reg_hid_dim5 = 100\n","reg_hid_dim6 = 50\n","learning_rate = 1e-4\n","\n","class NN(torch.nn.Module):\n"," def __init__(self):\n"," super(NN, self).__init__()\n"," self.reg = torch.nn.Sequential(\n"," nn.Dropout(),\n"," nn.Linear(embed_dim * 2, reg_hid_dim1),\n"," nn.ReLU(),\n"," nn.Linear(reg_hid_dim1, reg_hid_dim2),\n"," nn.ReLU(),\n"," nn.Dropout(),\n"," nn.Linear(reg_hid_dim2, reg_hid_dim3),\n"," nn.ReLU(),\n"," nn.Linear(reg_hid_dim3, reg_hid_dim4),\n"," nn.ReLU(),\n"," nn.Linear(reg_hid_dim4, reg_hid_dim5),\n"," nn.ReLU(),\n"," nn.Linear(reg_hid_dim5, reg_hid_dim6),\n"," nn.ReLU(),\n"," nn.Linear(reg_hid_dim6, 1)\n"," )\n","\n"," def forward(self, sentences):\n"," # inputs should be 3D, BATCH, NUM_WORDS, lstm_hidden_dim\n"," out = self.reg(sentences)\n"," return out\n","\n"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"tFgnl4bonSGg","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":340},"outputId":"c0cf6dcf-4003-4627-ddf5-9b62b65c7513","executionInfo":{"status":"ok","timestamp":1582148779105,"user_tz":0,"elapsed":1423,"user":{"displayName":"Andy Wang","photoUrl":"","userId":"02776860930356410397"}}},"source":["# Create model\n","model = NN()\n","model = model.to(device)\n","model.double()"],"execution_count":53,"outputs":[{"output_type":"execute_result","data":{"text/plain":["NN(\n"," (reg): Sequential(\n"," (0): Dropout(p=0.5, inplace=False)\n"," (1): Linear(in_features=600, out_features=500, bias=True)\n"," (2): ReLU()\n"," (3): Linear(in_features=500, out_features=400, bias=True)\n"," (4): ReLU()\n"," (5): Dropout(p=0.5, inplace=False)\n"," (6): Linear(in_features=400, out_features=300, bias=True)\n"," (7): ReLU()\n"," (8): Linear(in_features=300, out_features=200, bias=True)\n"," (9): ReLU()\n"," (10): Linear(in_features=200, out_features=100, bias=True)\n"," (11): ReLU()\n"," (12): Linear(in_features=100, out_features=50, bias=True)\n"," (13): ReLU()\n"," (14): Linear(in_features=50, out_features=1, bias=True)\n"," )\n",")"]},"metadata":{"tags":[]},"execution_count":53}]},{"cell_type":"code","metadata":{"id":"uDNWHcEh0lxP","colab_type":"code","colab":{}},"source":["# Optimiser\n","optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"M3IepQDap5AW","colab_type":"code","colab":{}},"source":["# Loss function\n","def calc_loss(output, target):\n"," loss = torch.nn.functional.mse_loss(output.squeeze(), target, reduction=\"mean\")\n"," return loss\n","\n","# Test function\n","def eval_target(output, target):\n"," acc = torch.nn.functional.l1_loss(output.squeeze(), target, reduction=\"mean\")\n"," return acc"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"VieQzJimED7Y","colab_type":"code","outputId":"6048f7b3-fd13-42f8-aaae-8f9d36d7bce0","executionInfo":{"status":"ok","timestamp":1582149003297,"user_tz":0,"elapsed":218103,"user":{"displayName":"Andy Wang","photoUrl":"","userId":"02776860930356410397"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["num_epochs = 200\n","\n","verbose_print = False\n","\n","train_losses = []\n","test_losses = []\n","test_acc = []\n","test_pearson = []\n","\n","for epoch in range(num_epochs):\n"," total_steps = len(loader_train) \n"," model.train()\n"," epoch_loss = []\n"," for idx_1, (sentences, labels) in enumerate(loader_train):\n"," model.zero_grad()\n"," sentences = sentences.to(device)\n"," labels = labels.to(device)\n"," \n"," output = model(sentences)\n"," loss = calc_loss(output, labels)\n"," loss.backward()\n"," avg_batch_loss = loss.item()\n"," optimizer.step()\n"," epoch_loss.append(avg_batch_loss)\n"," if verbose_print:\n"," print(f\"Epoch {epoch}, Batch {idx_1 + 1} Train Loss: {avg_batch_loss:.4f}\")\n","\n"," \n"," avg_epoch_loss = sum(epoch_loss) / total_steps\n"," train_losses.append(avg_epoch_loss)\n"," print(f\"Average Train Loss in Epoch {epoch}: {avg_epoch_loss:.4f}\")\n","\n"," # Test on test set\n","\n"," model.eval()\n"," epoch_loss = []\n"," epoch_acc = []\n","\n"," all_outputs = []\n"," all_labels = []\n","\n"," total_steps = len(loader_test) \n"," with torch.no_grad():\n"," for idx_1, (sentences, labels) in enumerate(loader_test):\n"," # Record for Pearson\n"," all_labels.extend(labels.tolist())\n","\n"," model.zero_grad()\n"," sentences = sentences.to(device)\n"," labels = labels.to(device)\n"," \n"," output = model(sentences)\n"," # Record for Pearson\n"," all_outputs.extend(output.squeeze().tolist())\n","\n"," loss = calc_loss(output, labels)\n"," avg_batch_loss = loss.item()\n"," acc = eval_target(output, labels)\n"," avg_batch_acc= acc.item()\n"," epoch_loss.append(avg_batch_loss)\n"," epoch_acc.append(avg_batch_acc)\n"," if verbose_print:\n"," print(f\"Epoch {epoch}, Batch {idx_1 + 1} Test Loss: {avg_batch_loss:.4f}, Test Acc: {avg_batch_acc:.4f}\")\n","\n"," avg_epoch_loss = sum(epoch_loss) / total_steps\n"," avg_epoch_acc = sum(epoch_acc) / total_steps\n"," test_losses.append(avg_epoch_loss)\n"," test_acc.append(avg_epoch_acc)\n"," print(f\"Average Test Loss in Epoch {epoch}: {avg_epoch_loss:.4f}\")\n"," print(f\"Average Test Acc in Epoch {epoch}: {avg_epoch_acc:.4f}\")\n","\n"," # Calc Pearson\n"," pearson = np.corrcoef(all_outputs, all_labels)[0, 1]\n"," test_pearson.append(pearson)\n"," print(f\"Pearson Coeff in Epoch {epoch}: {pearson}\")"],"execution_count":55,"outputs":[{"output_type":"stream","text":["Average Train Loss in Epoch 0: 0.6620\n","Average Test Loss in Epoch 0: 0.7663\n","Average Test Acc in Epoch 0: 0.5039\n","Pearson Coeff in Epoch 0: 0.08995647010240566\n","Average Train Loss in Epoch 1: 0.6612\n","Average Test Loss in Epoch 1: 0.7661\n","Average Test Acc in Epoch 1: 0.4965\n","Pearson Coeff in Epoch 1: 0.10240092862132691\n","Average Train Loss in Epoch 2: 0.6602\n","Average Test Loss in Epoch 2: 0.7643\n","Average Test Acc in Epoch 2: 0.4975\n","Pearson Coeff in Epoch 2: 0.08785221790108075\n","Average Train Loss in Epoch 3: 0.6565\n","Average Test Loss in Epoch 3: 0.7622\n","Average Test Acc in Epoch 3: 0.4907\n","Pearson Coeff in Epoch 3: 0.08998507906591446\n","Average Train Loss in Epoch 4: 0.6536\n","Average Test Loss in Epoch 4: 0.7624\n","Average Test Acc in Epoch 4: 0.4950\n","Pearson Coeff in Epoch 4: 0.08051515208226027\n","Average Train Loss in Epoch 5: 0.6505\n","Average Test Loss in Epoch 5: 0.7618\n","Average Test Acc in Epoch 5: 0.4964\n","Pearson Coeff in Epoch 5: 0.08520696912063215\n","Average Train Loss in Epoch 6: 0.6502\n","Average Test Loss in Epoch 6: 0.7619\n","Average Test Acc in Epoch 6: 0.4879\n","Pearson Coeff in Epoch 6: 0.09722789775123375\n","Average Train Loss in Epoch 7: 0.6446\n","Average Test Loss in Epoch 7: 0.7609\n","Average Test Acc in Epoch 7: 0.4942\n","Pearson Coeff in Epoch 7: 0.0969587985537999\n","Average Train Loss in Epoch 8: 0.6416\n","Average Test Loss in Epoch 8: 0.7692\n","Average Test Acc in Epoch 8: 0.5303\n","Pearson Coeff in Epoch 8: 0.08483564982396344\n","Average Train Loss in Epoch 9: 0.6446\n","Average Test Loss in Epoch 9: 0.7741\n","Average Test Acc in Epoch 9: 0.5450\n","Pearson Coeff in Epoch 9: 0.08678403403190069\n","Average Train Loss in Epoch 10: 0.6398\n","Average Test Loss in Epoch 10: 0.7685\n","Average Test Acc in Epoch 10: 0.5352\n","Pearson Coeff in Epoch 10: 0.09272690396815722\n","Average Train Loss in Epoch 11: 0.6358\n","Average Test Loss in Epoch 11: 0.7755\n","Average Test Acc in Epoch 11: 0.5354\n","Pearson Coeff in Epoch 11: 0.0841079208009783\n","Average Train Loss in Epoch 12: 0.6293\n","Average Test Loss in Epoch 12: 0.7581\n","Average Test Acc in Epoch 12: 0.4998\n","Pearson Coeff in Epoch 12: 0.11257421988511655\n","Average Train Loss in Epoch 13: 0.6221\n","Average Test Loss in Epoch 13: 0.7651\n","Average Test Acc in Epoch 13: 0.5137\n","Pearson Coeff in Epoch 13: 0.0806625768077105\n","Average Train Loss in Epoch 14: 0.6202\n","Average Test Loss in Epoch 14: 0.7661\n","Average Test Acc in Epoch 14: 0.5066\n","Pearson Coeff in Epoch 14: 0.08410615566862599\n","Average Train Loss in Epoch 15: 0.6129\n","Average Test Loss in Epoch 15: 0.7747\n","Average Test Acc in Epoch 15: 0.5199\n","Pearson Coeff in Epoch 15: 0.06966817902277206\n","Average Train Loss in Epoch 16: 0.6006\n","Average Test Loss in Epoch 16: 0.7781\n","Average Test Acc in Epoch 16: 0.5088\n","Pearson Coeff in Epoch 16: 0.03440325713561202\n","Average Train Loss in Epoch 17: 0.5983\n","Average Test Loss in Epoch 17: 0.8182\n","Average Test Acc in Epoch 17: 0.5629\n","Pearson Coeff in Epoch 17: 0.05078903341904453\n","Average Train Loss in Epoch 18: 0.5958\n","Average Test Loss in Epoch 18: 0.7805\n","Average Test Acc in Epoch 18: 0.5146\n","Pearson Coeff in Epoch 18: 0.024712050023470166\n","Average Train Loss in Epoch 19: 0.5833\n","Average Test Loss in Epoch 19: 0.7768\n","Average Test Acc in Epoch 19: 0.5218\n","Pearson Coeff in Epoch 19: 0.05124042242328605\n","Average Train Loss in Epoch 20: 0.5950\n","Average Test Loss in Epoch 20: 0.7891\n","Average Test Acc in Epoch 20: 0.5496\n","Pearson Coeff in Epoch 20: 0.08154238806288969\n","Average Train Loss in Epoch 21: 0.5749\n","Average Test Loss in Epoch 21: 0.7920\n","Average Test Acc in Epoch 21: 0.5464\n","Pearson Coeff in Epoch 21: 0.05144618518549028\n","Average Train Loss in Epoch 22: 0.5867\n","Average Test Loss in Epoch 22: 0.7953\n","Average Test Acc in Epoch 22: 0.5533\n","Pearson Coeff in Epoch 22: 0.04073405813112652\n","Average Train Loss in Epoch 23: 0.5710\n","Average Test Loss in Epoch 23: 0.7905\n","Average Test Acc in Epoch 23: 0.5351\n","Pearson Coeff in Epoch 23: 0.03522195693267434\n","Average Train Loss in Epoch 24: 0.5485\n","Average Test Loss in Epoch 24: 0.8063\n","Average Test Acc in Epoch 24: 0.5381\n","Pearson Coeff in Epoch 24: 0.026407780518169263\n","Average Train Loss in Epoch 25: 0.5632\n","Average Test Loss in Epoch 25: 0.7948\n","Average Test Acc in Epoch 25: 0.5408\n","Pearson Coeff in Epoch 25: 0.022041170348504224\n","Average Train Loss in Epoch 26: 0.5446\n","Average Test Loss in Epoch 26: 0.7969\n","Average Test Acc in Epoch 26: 0.5213\n","Pearson Coeff in Epoch 26: 0.014284607028673586\n","Average Train Loss in Epoch 27: 0.5446\n","Average Test Loss in Epoch 27: 0.7968\n","Average Test Acc in Epoch 27: 0.5415\n","Pearson Coeff in Epoch 27: 0.017645903319366257\n","Average Train Loss in Epoch 28: 0.5512\n","Average Test Loss in Epoch 28: 0.8061\n","Average Test Acc in Epoch 28: 0.5393\n","Pearson Coeff in Epoch 28: 0.016177243560314827\n","Average Train Loss in Epoch 29: 0.5445\n","Average Test Loss in Epoch 29: 0.8120\n","Average Test Acc in Epoch 29: 0.5490\n","Pearson Coeff in Epoch 29: 0.008574518927335087\n","Average Train Loss in Epoch 30: 0.5238\n","Average Test Loss in Epoch 30: 0.8005\n","Average Test Acc in Epoch 30: 0.5308\n","Pearson Coeff in Epoch 30: 0.013056171186541494\n","Average Train Loss in Epoch 31: 0.5185\n","Average Test Loss in Epoch 31: 0.8078\n","Average Test Acc in Epoch 31: 0.5415\n","Pearson Coeff in Epoch 31: 0.01275860119060391\n","Average Train Loss in Epoch 32: 0.5065\n","Average Test Loss in Epoch 32: 0.7927\n","Average Test Acc in Epoch 32: 0.5245\n","Pearson Coeff in Epoch 32: 0.0152326559873169\n","Average Train Loss in Epoch 33: 0.5066\n","Average Test Loss in Epoch 33: 0.8005\n","Average Test Acc in Epoch 33: 0.5276\n","Pearson Coeff in Epoch 33: 0.007219439791991584\n","Average Train Loss in Epoch 34: 0.5180\n","Average Test Loss in Epoch 34: 0.8622\n","Average Test Acc in Epoch 34: 0.5891\n","Pearson Coeff in Epoch 34: 0.002905471554344473\n","Average Train Loss in Epoch 35: 0.5234\n","Average Test Loss in Epoch 35: 0.8437\n","Average Test Acc in Epoch 35: 0.5635\n","Pearson Coeff in Epoch 35: 0.015007979123126564\n","Average Train Loss in Epoch 36: 0.5118\n","Average Test Loss in Epoch 36: 0.7981\n","Average Test Acc in Epoch 36: 0.5214\n","Pearson Coeff in Epoch 36: 0.028934001382114247\n","Average Train Loss in Epoch 37: 0.4881\n","Average Test Loss in Epoch 37: 0.8097\n","Average Test Acc in Epoch 37: 0.5557\n","Pearson Coeff in Epoch 37: 0.027247186939146937\n","Average Train Loss in Epoch 38: 0.4953\n","Average Test Loss in Epoch 38: 0.8343\n","Average Test Acc in Epoch 38: 0.5599\n","Pearson Coeff in Epoch 38: 0.02454688777169875\n","Average Train Loss in Epoch 39: 0.4949\n","Average Test Loss in Epoch 39: 0.8675\n","Average Test Acc in Epoch 39: 0.5836\n","Pearson Coeff in Epoch 39: 0.030065013476109424\n","Average Train Loss in Epoch 40: 0.4888\n","Average Test Loss in Epoch 40: 0.7936\n","Average Test Acc in Epoch 40: 0.5359\n","Pearson Coeff in Epoch 40: 0.0318880490476026\n","Average Train Loss in Epoch 41: 0.4794\n","Average Test Loss in Epoch 41: 0.8675\n","Average Test Acc in Epoch 41: 0.5806\n","Pearson Coeff in Epoch 41: 0.02266074017277038\n","Average Train Loss in Epoch 42: 0.4831\n","Average Test Loss in Epoch 42: 0.8867\n","Average Test Acc in Epoch 42: 0.5945\n","Pearson Coeff in Epoch 42: 0.02875054245736079\n","Average Train Loss in Epoch 43: 0.4831\n","Average Test Loss in Epoch 43: 0.7951\n","Average Test Acc in Epoch 43: 0.5134\n","Pearson Coeff in Epoch 43: 0.035082200319616036\n","Average Train Loss in Epoch 44: 0.4674\n","Average Test Loss in Epoch 44: 0.8326\n","Average Test Acc in Epoch 44: 0.5523\n","Pearson Coeff in Epoch 44: 0.03252811454523643\n","Average Train Loss in Epoch 45: 0.4953\n","Average Test Loss in Epoch 45: 0.7978\n","Average Test Acc in Epoch 45: 0.5265\n","Pearson Coeff in Epoch 45: 0.041670121907479034\n","Average Train Loss in Epoch 46: 0.4766\n","Average Test Loss in Epoch 46: 0.8297\n","Average Test Acc in Epoch 46: 0.5604\n","Pearson Coeff in Epoch 46: 0.03257827669208175\n","Average Train Loss in Epoch 47: 0.4641\n","Average Test Loss in Epoch 47: 0.8111\n","Average Test Acc in Epoch 47: 0.5253\n","Pearson Coeff in Epoch 47: 0.033840661305344494\n","Average Train Loss in Epoch 48: 0.4647\n","Average Test Loss in Epoch 48: 0.8275\n","Average Test Acc in Epoch 48: 0.5534\n","Pearson Coeff in Epoch 48: 0.061214436140912216\n","Average Train Loss in Epoch 49: 0.4619\n","Average Test Loss in Epoch 49: 0.8818\n","Average Test Acc in Epoch 49: 0.5923\n","Pearson Coeff in Epoch 49: 0.023892527794506912\n","Average Train Loss in Epoch 50: 0.4589\n","Average Test Loss in Epoch 50: 0.8358\n","Average Test Acc in Epoch 50: 0.5659\n","Pearson Coeff in Epoch 50: 0.033884055620729624\n","Average Train Loss in Epoch 51: 0.4452\n","Average Test Loss in Epoch 51: 0.8367\n","Average Test Acc in Epoch 51: 0.5612\n","Pearson Coeff in Epoch 51: 0.039885668456414954\n","Average Train Loss in Epoch 52: 0.4497\n","Average Test Loss in Epoch 52: 0.8894\n","Average Test Acc in Epoch 52: 0.5956\n","Pearson Coeff in Epoch 52: 0.026320381149949678\n","Average Train Loss in Epoch 53: 0.4483\n","Average Test Loss in Epoch 53: 0.9157\n","Average Test Acc in Epoch 53: 0.6273\n","Pearson Coeff in Epoch 53: 0.04487338270180374\n","Average Train Loss in Epoch 54: 0.4414\n","Average Test Loss in Epoch 54: 0.9090\n","Average Test Acc in Epoch 54: 0.6082\n","Pearson Coeff in Epoch 54: 0.04365285696226019\n","Average Train Loss in Epoch 55: 0.4468\n","Average Test Loss in Epoch 55: 0.8733\n","Average Test Acc in Epoch 55: 0.5899\n","Pearson Coeff in Epoch 55: 0.02936432993728556\n","Average Train Loss in Epoch 56: 0.4483\n","Average Test Loss in Epoch 56: 0.8155\n","Average Test Acc in Epoch 56: 0.5312\n","Pearson Coeff in Epoch 56: 0.03910649350221317\n","Average Train Loss in Epoch 57: 0.4481\n","Average Test Loss in Epoch 57: 0.8303\n","Average Test Acc in Epoch 57: 0.5599\n","Pearson Coeff in Epoch 57: 0.02479665383461297\n","Average Train Loss in Epoch 58: 0.4343\n","Average Test Loss in Epoch 58: 0.8369\n","Average Test Acc in Epoch 58: 0.5662\n","Pearson Coeff in Epoch 58: 0.03409512500190778\n","Average Train Loss in Epoch 59: 0.4294\n","Average Test Loss in Epoch 59: 0.8021\n","Average Test Acc in Epoch 59: 0.5221\n","Pearson Coeff in Epoch 59: 0.04638170919893269\n","Average Train Loss in Epoch 60: 0.4295\n","Average Test Loss in Epoch 60: 0.9009\n","Average Test Acc in Epoch 60: 0.5913\n","Pearson Coeff in Epoch 60: 0.039379406811226016\n","Average Train Loss in Epoch 61: 0.4235\n","Average Test Loss in Epoch 61: 0.8332\n","Average Test Acc in Epoch 61: 0.5442\n","Pearson Coeff in Epoch 61: 0.022491304387294266\n","Average Train Loss in Epoch 62: 0.4337\n","Average Test Loss in Epoch 62: 0.8606\n","Average Test Acc in Epoch 62: 0.5851\n","Pearson Coeff in Epoch 62: 0.031017748369119878\n","Average Train Loss in Epoch 63: 0.4027\n","Average Test Loss in Epoch 63: 0.8958\n","Average Test Acc in Epoch 63: 0.5869\n","Pearson Coeff in Epoch 63: 0.02301825683478988\n","Average Train Loss in Epoch 64: 0.4151\n","Average Test Loss in Epoch 64: 0.8059\n","Average Test Acc in Epoch 64: 0.5461\n","Pearson Coeff in Epoch 64: 0.05307670742820803\n","Average Train Loss in Epoch 65: 0.4103\n","Average Test Loss in Epoch 65: 1.0036\n","Average Test Acc in Epoch 65: 0.6233\n","Pearson Coeff in Epoch 65: 0.01828948002529151\n","Average Train Loss in Epoch 66: 0.4116\n","Average Test Loss in Epoch 66: 0.8169\n","Average Test Acc in Epoch 66: 0.5373\n","Pearson Coeff in Epoch 66: 0.037872171994512885\n","Average Train Loss in Epoch 67: 0.4169\n","Average Test Loss in Epoch 67: 0.8641\n","Average Test Acc in Epoch 67: 0.5726\n","Pearson Coeff in Epoch 67: 0.020018259888135912\n","Average Train Loss in Epoch 68: 0.4057\n","Average Test Loss in Epoch 68: 0.8141\n","Average Test Acc in Epoch 68: 0.5228\n","Pearson Coeff in Epoch 68: 0.03190141645794034\n","Average Train Loss in Epoch 69: 0.3937\n","Average Test Loss in Epoch 69: 0.8589\n","Average Test Acc in Epoch 69: 0.5543\n","Pearson Coeff in Epoch 69: 0.021311779542357148\n","Average Train Loss in Epoch 70: 0.4023\n","Average Test Loss in Epoch 70: 0.8439\n","Average Test Acc in Epoch 70: 0.5536\n","Pearson Coeff in Epoch 70: 0.03493846595947967\n","Average Train Loss in Epoch 71: 0.4032\n","Average Test Loss in Epoch 71: 0.9147\n","Average Test Acc in Epoch 71: 0.5975\n","Pearson Coeff in Epoch 71: 0.019059264431991896\n","Average Train Loss in Epoch 72: 0.3974\n","Average Test Loss in Epoch 72: 0.8382\n","Average Test Acc in Epoch 72: 0.5401\n","Pearson Coeff in Epoch 72: 0.039460914428605984\n","Average Train Loss in Epoch 73: 0.3925\n","Average Test Loss in Epoch 73: 0.9295\n","Average Test Acc in Epoch 73: 0.6170\n","Pearson Coeff in Epoch 73: 0.02540440149353885\n","Average Train Loss in Epoch 74: 0.3913\n","Average Test Loss in Epoch 74: 0.9259\n","Average Test Acc in Epoch 74: 0.5852\n","Pearson Coeff in Epoch 74: 0.0074979394752139735\n","Average Train Loss in Epoch 75: 0.4027\n","Average Test Loss in Epoch 75: 0.9145\n","Average Test Acc in Epoch 75: 0.5933\n","Pearson Coeff in Epoch 75: 0.022137208452189132\n","Average Train Loss in Epoch 76: 0.3969\n","Average Test Loss in Epoch 76: 0.9044\n","Average Test Acc in Epoch 76: 0.5898\n","Pearson Coeff in Epoch 76: 0.023222165014897454\n","Average Train Loss in Epoch 77: 0.3869\n","Average Test Loss in Epoch 77: 0.9021\n","Average Test Acc in Epoch 77: 0.5731\n","Pearson Coeff in Epoch 77: 0.021957051064203644\n","Average Train Loss in Epoch 78: 0.3827\n","Average Test Loss in Epoch 78: 0.9041\n","Average Test Acc in Epoch 78: 0.5990\n","Pearson Coeff in Epoch 78: 0.033689201468040456\n","Average Train Loss in Epoch 79: 0.3662\n","Average Test Loss in Epoch 79: 0.9130\n","Average Test Acc in Epoch 79: 0.5981\n","Pearson Coeff in Epoch 79: 0.044404753261681286\n","Average Train Loss in Epoch 80: 0.3707\n","Average Test Loss in Epoch 80: 0.8732\n","Average Test Acc in Epoch 80: 0.5679\n","Pearson Coeff in Epoch 80: 0.0351305188146265\n","Average Train Loss in Epoch 81: 0.3772\n","Average Test Loss in Epoch 81: 0.9262\n","Average Test Acc in Epoch 81: 0.6065\n","Pearson Coeff in Epoch 81: 0.04523515286939734\n","Average Train Loss in Epoch 82: 0.3715\n","Average Test Loss in Epoch 82: 0.8224\n","Average Test Acc in Epoch 82: 0.5498\n","Pearson Coeff in Epoch 82: 0.061525212818027177\n","Average Train Loss in Epoch 83: 0.3674\n","Average Test Loss in Epoch 83: 0.9042\n","Average Test Acc in Epoch 83: 0.6203\n","Pearson Coeff in Epoch 83: 0.04609860293508482\n","Average Train Loss in Epoch 84: 0.3716\n","Average Test Loss in Epoch 84: 0.9046\n","Average Test Acc in Epoch 84: 0.5862\n","Pearson Coeff in Epoch 84: 0.04578201265320917\n","Average Train Loss in Epoch 85: 0.3654\n","Average Test Loss in Epoch 85: 0.7952\n","Average Test Acc in Epoch 85: 0.5128\n","Pearson Coeff in Epoch 85: 0.07501947403884338\n","Average Train Loss in Epoch 86: 0.3576\n","Average Test Loss in Epoch 86: 0.8008\n","Average Test Acc in Epoch 86: 0.5406\n","Pearson Coeff in Epoch 86: 0.07014740944601133\n","Average Train Loss in Epoch 87: 0.3631\n","Average Test Loss in Epoch 87: 0.8389\n","Average Test Acc in Epoch 87: 0.5655\n","Pearson Coeff in Epoch 87: 0.05410292869233262\n","Average Train Loss in Epoch 88: 0.3791\n","Average Test Loss in Epoch 88: 0.8911\n","Average Test Acc in Epoch 88: 0.5849\n","Pearson Coeff in Epoch 88: 0.04343526043008413\n","Average Train Loss in Epoch 89: 0.3520\n","Average Test Loss in Epoch 89: 0.9272\n","Average Test Acc in Epoch 89: 0.6098\n","Pearson Coeff in Epoch 89: 0.02612316759059236\n","Average Train Loss in Epoch 90: 0.3720\n","Average Test Loss in Epoch 90: 0.8563\n","Average Test Acc in Epoch 90: 0.5678\n","Pearson Coeff in Epoch 90: 0.05558137817248038\n","Average Train Loss in Epoch 91: 0.3532\n","Average Test Loss in Epoch 91: 0.8102\n","Average Test Acc in Epoch 91: 0.5314\n","Pearson Coeff in Epoch 91: 0.05389148480041872\n","Average Train Loss in Epoch 92: 0.3458\n","Average Test Loss in Epoch 92: 0.8223\n","Average Test Acc in Epoch 92: 0.5377\n","Pearson Coeff in Epoch 92: 0.050529016317738484\n","Average Train Loss in Epoch 93: 0.3481\n","Average Test Loss in Epoch 93: 0.8632\n","Average Test Acc in Epoch 93: 0.5552\n","Pearson Coeff in Epoch 93: 0.03298678671838119\n","Average Train Loss in Epoch 94: 0.3550\n","Average Test Loss in Epoch 94: 0.8943\n","Average Test Acc in Epoch 94: 0.5834\n","Pearson Coeff in Epoch 94: 0.042159121830747756\n","Average Train Loss in Epoch 95: 0.3504\n","Average Test Loss in Epoch 95: 0.8424\n","Average Test Acc in Epoch 95: 0.5544\n","Pearson Coeff in Epoch 95: 0.04859467404111373\n","Average Train Loss in Epoch 96: 0.3531\n","Average Test Loss in Epoch 96: 0.8929\n","Average Test Acc in Epoch 96: 0.5769\n","Pearson Coeff in Epoch 96: 0.0291886215342315\n","Average Train Loss in Epoch 97: 0.3484\n","Average Test Loss in Epoch 97: 0.8524\n","Average Test Acc in Epoch 97: 0.5607\n","Pearson Coeff in Epoch 97: 0.033727133507403285\n","Average Train Loss in Epoch 98: 0.3409\n","Average Test Loss in Epoch 98: 0.9328\n","Average Test Acc in Epoch 98: 0.5905\n","Pearson Coeff in Epoch 98: 0.028486974034068427\n","Average Train Loss in Epoch 99: 0.3411\n","Average Test Loss in Epoch 99: 0.9115\n","Average Test Acc in Epoch 99: 0.6182\n","Pearson Coeff in Epoch 99: 0.043974053097916195\n","Average Train Loss in Epoch 100: 0.3433\n","Average Test Loss in Epoch 100: 0.8757\n","Average Test Acc in Epoch 100: 0.5681\n","Pearson Coeff in Epoch 100: 0.03582088960457004\n","Average Train Loss in Epoch 101: 0.3320\n","Average Test Loss in Epoch 101: 0.8550\n","Average Test Acc in Epoch 101: 0.5624\n","Pearson Coeff in Epoch 101: 0.01781143151220178\n","Average Train Loss in Epoch 102: 0.3412\n","Average Test Loss in Epoch 102: 0.9036\n","Average Test Acc in Epoch 102: 0.5772\n","Pearson Coeff in Epoch 102: 0.034928829292705445\n","Average Train Loss in Epoch 103: 0.3350\n","Average Test Loss in Epoch 103: 0.9253\n","Average Test Acc in Epoch 103: 0.5789\n","Pearson Coeff in Epoch 103: 0.030859698149130543\n","Average Train Loss in Epoch 104: 0.3337\n","Average Test Loss in Epoch 104: 0.8454\n","Average Test Acc in Epoch 104: 0.5412\n","Pearson Coeff in Epoch 104: 0.03020864455911108\n","Average Train Loss in Epoch 105: 0.3347\n","Average Test Loss in Epoch 105: 0.8287\n","Average Test Acc in Epoch 105: 0.5680\n","Pearson Coeff in Epoch 105: 0.036381000729738794\n","Average Train Loss in Epoch 106: 0.3361\n","Average Test Loss in Epoch 106: 0.8664\n","Average Test Acc in Epoch 106: 0.5727\n","Pearson Coeff in Epoch 106: 0.028513613295388193\n","Average Train Loss in Epoch 107: 0.3291\n","Average Test Loss in Epoch 107: 0.8928\n","Average Test Acc in Epoch 107: 0.5918\n","Pearson Coeff in Epoch 107: 0.0319642990980195\n","Average Train Loss in Epoch 108: 0.3288\n","Average Test Loss in Epoch 108: 0.9965\n","Average Test Acc in Epoch 108: 0.6418\n","Pearson Coeff in Epoch 108: 0.0379174975900099\n","Average Train Loss in Epoch 109: 0.3140\n","Average Test Loss in Epoch 109: 0.9025\n","Average Test Acc in Epoch 109: 0.5674\n","Pearson Coeff in Epoch 109: 0.03807412674721549\n","Average Train Loss in Epoch 110: 0.3184\n","Average Test Loss in Epoch 110: 0.9120\n","Average Test Acc in Epoch 110: 0.5899\n","Pearson Coeff in Epoch 110: 0.030576066158500014\n","Average Train Loss in Epoch 111: 0.3127\n","Average Test Loss in Epoch 111: 0.8263\n","Average Test Acc in Epoch 111: 0.5279\n","Pearson Coeff in Epoch 111: 0.042400969385593554\n","Average Train Loss in Epoch 112: 0.3125\n","Average Test Loss in Epoch 112: 0.8366\n","Average Test Acc in Epoch 112: 0.5495\n","Pearson Coeff in Epoch 112: 0.03801355367858849\n","Average Train Loss in Epoch 113: 0.3204\n","Average Test Loss in Epoch 113: 0.8579\n","Average Test Acc in Epoch 113: 0.5610\n","Pearson Coeff in Epoch 113: 0.03690562699852183\n","Average Train Loss in Epoch 114: 0.3020\n","Average Test Loss in Epoch 114: 0.8763\n","Average Test Acc in Epoch 114: 0.5615\n","Pearson Coeff in Epoch 114: 0.03387086636386486\n","Average Train Loss in Epoch 115: 0.2976\n","Average Test Loss in Epoch 115: 0.8585\n","Average Test Acc in Epoch 115: 0.5741\n","Pearson Coeff in Epoch 115: 0.04286772382612775\n","Average Train Loss in Epoch 116: 0.3070\n","Average Test Loss in Epoch 116: 0.8318\n","Average Test Acc in Epoch 116: 0.5285\n","Pearson Coeff in Epoch 116: 0.02016396345988365\n","Average Train Loss in Epoch 117: 0.3106\n","Average Test Loss in Epoch 117: 0.8875\n","Average Test Acc in Epoch 117: 0.5825\n","Pearson Coeff in Epoch 117: 0.01742958565946704\n","Average Train Loss in Epoch 118: 0.3008\n","Average Test Loss in Epoch 118: 0.8866\n","Average Test Acc in Epoch 118: 0.5783\n","Pearson Coeff in Epoch 118: 0.011894168245774562\n","Average Train Loss in Epoch 119: 0.2965\n","Average Test Loss in Epoch 119: 0.9024\n","Average Test Acc in Epoch 119: 0.5804\n","Pearson Coeff in Epoch 119: 0.017901122695037613\n","Average Train Loss in Epoch 120: 0.2985\n","Average Test Loss in Epoch 120: 0.8390\n","Average Test Acc in Epoch 120: 0.5300\n","Pearson Coeff in Epoch 120: 0.03323263164114658\n","Average Train Loss in Epoch 121: 0.2960\n","Average Test Loss in Epoch 121: 0.8871\n","Average Test Acc in Epoch 121: 0.5666\n","Pearson Coeff in Epoch 121: 0.05489411250307537\n","Average Train Loss in Epoch 122: 0.3048\n","Average Test Loss in Epoch 122: 0.9183\n","Average Test Acc in Epoch 122: 0.5892\n","Pearson Coeff in Epoch 122: 0.015821877382426755\n","Average Train Loss in Epoch 123: 0.2923\n","Average Test Loss in Epoch 123: 0.8219\n","Average Test Acc in Epoch 123: 0.5236\n","Pearson Coeff in Epoch 123: 0.023606486105931596\n","Average Train Loss in Epoch 124: 0.3007\n","Average Test Loss in Epoch 124: 0.9175\n","Average Test Acc in Epoch 124: 0.5832\n","Pearson Coeff in Epoch 124: 0.015491203855874276\n","Average Train Loss in Epoch 125: 0.2883\n","Average Test Loss in Epoch 125: 0.8938\n","Average Test Acc in Epoch 125: 0.5706\n","Pearson Coeff in Epoch 125: 0.026858006096465933\n","Average Train Loss in Epoch 126: 0.2790\n","Average Test Loss in Epoch 126: 0.8695\n","Average Test Acc in Epoch 126: 0.5530\n","Pearson Coeff in Epoch 126: 0.034156660410896704\n","Average Train Loss in Epoch 127: 0.2934\n","Average Test Loss in Epoch 127: 0.9234\n","Average Test Acc in Epoch 127: 0.5672\n","Pearson Coeff in Epoch 127: 0.026996303384766024\n","Average Train Loss in Epoch 128: 0.2766\n","Average Test Loss in Epoch 128: 0.9271\n","Average Test Acc in Epoch 128: 0.5799\n","Pearson Coeff in Epoch 128: 0.025754403180389178\n","Average Train Loss in Epoch 129: 0.2816\n","Average Test Loss in Epoch 129: 0.9538\n","Average Test Acc in Epoch 129: 0.6020\n","Pearson Coeff in Epoch 129: 0.030846062002874364\n","Average Train Loss in Epoch 130: 0.2806\n","Average Test Loss in Epoch 130: 0.8838\n","Average Test Acc in Epoch 130: 0.5659\n","Pearson Coeff in Epoch 130: 0.00602962370483143\n","Average Train Loss in Epoch 131: 0.2858\n","Average Test Loss in Epoch 131: 0.8466\n","Average Test Acc in Epoch 131: 0.5445\n","Pearson Coeff in Epoch 131: 0.030561253042213306\n","Average Train Loss in Epoch 132: 0.2758\n","Average Test Loss in Epoch 132: 0.8672\n","Average Test Acc in Epoch 132: 0.5718\n","Pearson Coeff in Epoch 132: 0.034908364026499356\n","Average Train Loss in Epoch 133: 0.2724\n","Average Test Loss in Epoch 133: 0.8281\n","Average Test Acc in Epoch 133: 0.5363\n","Pearson Coeff in Epoch 133: 0.022957392216816627\n","Average Train Loss in Epoch 134: 0.2712\n","Average Test Loss in Epoch 134: 0.8526\n","Average Test Acc in Epoch 134: 0.5405\n","Pearson Coeff in Epoch 134: 0.02383911929644182\n","Average Train Loss in Epoch 135: 0.2789\n","Average Test Loss in Epoch 135: 0.8286\n","Average Test Acc in Epoch 135: 0.5373\n","Pearson Coeff in Epoch 135: 0.016508881129617282\n","Average Train Loss in Epoch 136: 0.2662\n","Average Test Loss in Epoch 136: 0.8486\n","Average Test Acc in Epoch 136: 0.5451\n","Pearson Coeff in Epoch 136: 0.01304662224248889\n","Average Train Loss in Epoch 137: 0.2821\n","Average Test Loss in Epoch 137: 0.8651\n","Average Test Acc in Epoch 137: 0.5547\n","Pearson Coeff in Epoch 137: 0.0019586850589532337\n","Average Train Loss in Epoch 138: 0.2610\n","Average Test Loss in Epoch 138: 0.8613\n","Average Test Acc in Epoch 138: 0.5575\n","Pearson Coeff in Epoch 138: 0.00796999716469933\n","Average Train Loss in Epoch 139: 0.2628\n","Average Test Loss in Epoch 139: 0.8724\n","Average Test Acc in Epoch 139: 0.5743\n","Pearson Coeff in Epoch 139: 0.007572669802658232\n","Average Train Loss in Epoch 140: 0.2631\n","Average Test Loss in Epoch 140: 0.9085\n","Average Test Acc in Epoch 140: 0.5870\n","Pearson Coeff in Epoch 140: 0.00649244152151307\n","Average Train Loss in Epoch 141: 0.2649\n","Average Test Loss in Epoch 141: 0.8736\n","Average Test Acc in Epoch 141: 0.5615\n","Pearson Coeff in Epoch 141: -0.002194289680742873\n","Average Train Loss in Epoch 142: 0.2550\n","Average Test Loss in Epoch 142: 0.8635\n","Average Test Acc in Epoch 142: 0.5524\n","Pearson Coeff in Epoch 142: -0.019209799159054797\n","Average Train Loss in Epoch 143: 0.2721\n","Average Test Loss in Epoch 143: 0.8674\n","Average Test Acc in Epoch 143: 0.5725\n","Pearson Coeff in Epoch 143: 0.011841560309330507\n","Average Train Loss in Epoch 144: 0.2612\n","Average Test Loss in Epoch 144: 0.8933\n","Average Test Acc in Epoch 144: 0.5783\n","Pearson Coeff in Epoch 144: 0.015335295655636239\n","Average Train Loss in Epoch 145: 0.2460\n","Average Test Loss in Epoch 145: 0.8534\n","Average Test Acc in Epoch 145: 0.5462\n","Pearson Coeff in Epoch 145: 0.012246244481480338\n","Average Train Loss in Epoch 146: 0.2487\n","Average Test Loss in Epoch 146: 0.8794\n","Average Test Acc in Epoch 146: 0.5649\n","Pearson Coeff in Epoch 146: 0.0030086032406489993\n","Average Train Loss in Epoch 147: 0.2668\n","Average Test Loss in Epoch 147: 0.8430\n","Average Test Acc in Epoch 147: 0.5478\n","Pearson Coeff in Epoch 147: 0.02185243567968775\n","Average Train Loss in Epoch 148: 0.2471\n","Average Test Loss in Epoch 148: 0.8671\n","Average Test Acc in Epoch 148: 0.5531\n","Pearson Coeff in Epoch 148: 0.012109224438763081\n","Average Train Loss in Epoch 149: 0.2527\n","Average Test Loss in Epoch 149: 0.8855\n","Average Test Acc in Epoch 149: 0.5573\n","Pearson Coeff in Epoch 149: 0.01792265970954428\n","Average Train Loss in Epoch 150: 0.2468\n","Average Test Loss in Epoch 150: 0.9030\n","Average Test Acc in Epoch 150: 0.5738\n","Pearson Coeff in Epoch 150: 0.01928680428916586\n","Average Train Loss in Epoch 151: 0.2584\n","Average Test Loss in Epoch 151: 0.8661\n","Average Test Acc in Epoch 151: 0.5534\n","Pearson Coeff in Epoch 151: 0.026805307370708856\n","Average Train Loss in Epoch 152: 0.2538\n","Average Test Loss in Epoch 152: 0.8394\n","Average Test Acc in Epoch 152: 0.5333\n","Pearson Coeff in Epoch 152: 0.021167035711144467\n","Average Train Loss in Epoch 153: 0.2502\n","Average Test Loss in Epoch 153: 0.9108\n","Average Test Acc in Epoch 153: 0.5862\n","Pearson Coeff in Epoch 153: 0.01386693755224287\n","Average Train Loss in Epoch 154: 0.2417\n","Average Test Loss in Epoch 154: 0.8577\n","Average Test Acc in Epoch 154: 0.5404\n","Pearson Coeff in Epoch 154: 0.02032267393579301\n","Average Train Loss in Epoch 155: 0.2487\n","Average Test Loss in Epoch 155: 0.8480\n","Average Test Acc in Epoch 155: 0.5435\n","Pearson Coeff in Epoch 155: 0.020721608833848708\n","Average Train Loss in Epoch 156: 0.2309\n","Average Test Loss in Epoch 156: 0.8632\n","Average Test Acc in Epoch 156: 0.5529\n","Pearson Coeff in Epoch 156: 0.010718219262776878\n","Average Train Loss in Epoch 157: 0.2405\n","Average Test Loss in Epoch 157: 0.8542\n","Average Test Acc in Epoch 157: 0.5461\n","Pearson Coeff in Epoch 157: 0.004773610137089486\n","Average Train Loss in Epoch 158: 0.2400\n","Average Test Loss in Epoch 158: 0.8659\n","Average Test Acc in Epoch 158: 0.5455\n","Pearson Coeff in Epoch 158: 0.007376322065120279\n","Average Train Loss in Epoch 159: 0.2398\n","Average Test Loss in Epoch 159: 0.9124\n","Average Test Acc in Epoch 159: 0.5820\n","Pearson Coeff in Epoch 159: -0.00015178024885565756\n","Average Train Loss in Epoch 160: 0.2278\n","Average Test Loss in Epoch 160: 0.8595\n","Average Test Acc in Epoch 160: 0.5479\n","Pearson Coeff in Epoch 160: 0.007502927891681227\n","Average Train Loss in Epoch 161: 0.2369\n","Average Test Loss in Epoch 161: 0.8597\n","Average Test Acc in Epoch 161: 0.5452\n","Pearson Coeff in Epoch 161: 0.00014284301804946774\n","Average Train Loss in Epoch 162: 0.2382\n","Average Test Loss in Epoch 162: 0.8550\n","Average Test Acc in Epoch 162: 0.5483\n","Pearson Coeff in Epoch 162: 0.004656237306182772\n","Average Train Loss in Epoch 163: 0.2345\n","Average Test Loss in Epoch 163: 0.9096\n","Average Test Acc in Epoch 163: 0.5909\n","Pearson Coeff in Epoch 163: -0.01871534517722504\n","Average Train Loss in Epoch 164: 0.2343\n","Average Test Loss in Epoch 164: 0.8792\n","Average Test Acc in Epoch 164: 0.5686\n","Pearson Coeff in Epoch 164: 0.008573083740883362\n","Average Train Loss in Epoch 165: 0.2402\n","Average Test Loss in Epoch 165: 0.9202\n","Average Test Acc in Epoch 165: 0.6156\n","Pearson Coeff in Epoch 165: 0.024334965089025763\n","Average Train Loss in Epoch 166: 0.2197\n","Average Test Loss in Epoch 166: 0.9754\n","Average Test Acc in Epoch 166: 0.6057\n","Pearson Coeff in Epoch 166: 0.011725717948630187\n","Average Train Loss in Epoch 167: 0.2276\n","Average Test Loss in Epoch 167: 0.8546\n","Average Test Acc in Epoch 167: 0.5442\n","Pearson Coeff in Epoch 167: 0.011046510368505513\n","Average Train Loss in Epoch 168: 0.2298\n","Average Test Loss in Epoch 168: 0.9135\n","Average Test Acc in Epoch 168: 0.5927\n","Pearson Coeff in Epoch 168: 0.03389040022876638\n","Average Train Loss in Epoch 169: 0.2212\n","Average Test Loss in Epoch 169: 0.8773\n","Average Test Acc in Epoch 169: 0.5600\n","Pearson Coeff in Epoch 169: 0.015457445454646576\n","Average Train Loss in Epoch 170: 0.2258\n","Average Test Loss in Epoch 170: 0.8647\n","Average Test Acc in Epoch 170: 0.5630\n","Pearson Coeff in Epoch 170: 0.016376017286977584\n","Average Train Loss in Epoch 171: 0.2282\n","Average Test Loss in Epoch 171: 0.8831\n","Average Test Acc in Epoch 171: 0.5477\n","Pearson Coeff in Epoch 171: 0.005316351527280453\n","Average Train Loss in Epoch 172: 0.2286\n","Average Test Loss in Epoch 172: 0.8483\n","Average Test Acc in Epoch 172: 0.5297\n","Pearson Coeff in Epoch 172: 0.01813497726183776\n","Average Train Loss in Epoch 173: 0.2255\n","Average Test Loss in Epoch 173: 0.9039\n","Average Test Acc in Epoch 173: 0.5771\n","Pearson Coeff in Epoch 173: -0.00016817483772310537\n","Average Train Loss in Epoch 174: 0.2200\n","Average Test Loss in Epoch 174: 0.8891\n","Average Test Acc in Epoch 174: 0.5726\n","Pearson Coeff in Epoch 174: -0.010752135405116008\n","Average Train Loss in Epoch 175: 0.2166\n","Average Test Loss in Epoch 175: 0.8750\n","Average Test Acc in Epoch 175: 0.5434\n","Pearson Coeff in Epoch 175: -0.01824576821273713\n","Average Train Loss in Epoch 176: 0.2163\n","Average Test Loss in Epoch 176: 0.8738\n","Average Test Acc in Epoch 176: 0.5589\n","Pearson Coeff in Epoch 176: 0.01635199408345275\n","Average Train Loss in Epoch 177: 0.2183\n","Average Test Loss in Epoch 177: 0.9015\n","Average Test Acc in Epoch 177: 0.5719\n","Pearson Coeff in Epoch 177: 0.0014618970029829984\n","Average Train Loss in Epoch 178: 0.2082\n","Average Test Loss in Epoch 178: 0.8748\n","Average Test Acc in Epoch 178: 0.5615\n","Pearson Coeff in Epoch 178: 0.02031770062120775\n","Average Train Loss in Epoch 179: 0.2169\n","Average Test Loss in Epoch 179: 0.8446\n","Average Test Acc in Epoch 179: 0.5288\n","Pearson Coeff in Epoch 179: 0.002650380371354998\n","Average Train Loss in Epoch 180: 0.2099\n","Average Test Loss in Epoch 180: 0.9186\n","Average Test Acc in Epoch 180: 0.5683\n","Pearson Coeff in Epoch 180: 0.001495790614385095\n","Average Train Loss in Epoch 181: 0.2144\n","Average Test Loss in Epoch 181: 0.9052\n","Average Test Acc in Epoch 181: 0.5705\n","Pearson Coeff in Epoch 181: 0.0009878489152257259\n","Average Train Loss in Epoch 182: 0.2132\n","Average Test Loss in Epoch 182: 0.9153\n","Average Test Acc in Epoch 182: 0.5923\n","Pearson Coeff in Epoch 182: -0.004482148221832333\n","Average Train Loss in Epoch 183: 0.2085\n","Average Test Loss in Epoch 183: 0.8781\n","Average Test Acc in Epoch 183: 0.5646\n","Pearson Coeff in Epoch 183: 0.004185669015919783\n","Average Train Loss in Epoch 184: 0.2098\n","Average Test Loss in Epoch 184: 0.8667\n","Average Test Acc in Epoch 184: 0.5359\n","Pearson Coeff in Epoch 184: -0.0036715986324464193\n","Average Train Loss in Epoch 185: 0.2042\n","Average Test Loss in Epoch 185: 0.8785\n","Average Test Acc in Epoch 185: 0.5551\n","Pearson Coeff in Epoch 185: 0.00024429492684952794\n","Average Train Loss in Epoch 186: 0.2071\n","Average Test Loss in Epoch 186: 0.9252\n","Average Test Acc in Epoch 186: 0.5882\n","Pearson Coeff in Epoch 186: 0.0019387904762869135\n","Average Train Loss in Epoch 187: 0.2048\n","Average Test Loss in Epoch 187: 0.8968\n","Average Test Acc in Epoch 187: 0.5701\n","Pearson Coeff in Epoch 187: 0.01304660487221574\n","Average Train Loss in Epoch 188: 0.2034\n","Average Test Loss in Epoch 188: 0.8969\n","Average Test Acc in Epoch 188: 0.5705\n","Pearson Coeff in Epoch 188: -0.002628562552341833\n","Average Train Loss in Epoch 189: 0.2002\n","Average Test Loss in Epoch 189: 0.9538\n","Average Test Acc in Epoch 189: 0.6014\n","Pearson Coeff in Epoch 189: -0.0002931281365674012\n","Average Train Loss in Epoch 190: 0.2053\n","Average Test Loss in Epoch 190: 0.8964\n","Average Test Acc in Epoch 190: 0.5615\n","Pearson Coeff in Epoch 190: -0.000550425294872905\n","Average Train Loss in Epoch 191: 0.2035\n","Average Test Loss in Epoch 191: 0.9473\n","Average Test Acc in Epoch 191: 0.5742\n","Pearson Coeff in Epoch 191: -0.012136436033133689\n","Average Train Loss in Epoch 192: 0.2115\n","Average Test Loss in Epoch 192: 0.9701\n","Average Test Acc in Epoch 192: 0.6156\n","Pearson Coeff in Epoch 192: 0.008599511910649242\n","Average Train Loss in Epoch 193: 0.2052\n","Average Test Loss in Epoch 193: 0.9361\n","Average Test Acc in Epoch 193: 0.5961\n","Pearson Coeff in Epoch 193: -0.003807488787164131\n","Average Train Loss in Epoch 194: 0.2009\n","Average Test Loss in Epoch 194: 0.9240\n","Average Test Acc in Epoch 194: 0.5793\n","Pearson Coeff in Epoch 194: -0.014016988558772583\n","Average Train Loss in Epoch 195: 0.2007\n","Average Test Loss in Epoch 195: 0.9297\n","Average Test Acc in Epoch 195: 0.6005\n","Pearson Coeff in Epoch 195: -0.02436993689798892\n","Average Train Loss in Epoch 196: 0.1944\n","Average Test Loss in Epoch 196: 0.9071\n","Average Test Acc in Epoch 196: 0.5960\n","Pearson Coeff in Epoch 196: -0.00836166056593199\n","Average Train Loss in Epoch 197: 0.2018\n","Average Test Loss in Epoch 197: 0.8509\n","Average Test Acc in Epoch 197: 0.5325\n","Pearson Coeff in Epoch 197: -0.004897029825483201\n","Average Train Loss in Epoch 198: 0.1994\n","Average Test Loss in Epoch 198: 0.9199\n","Average Test Acc in Epoch 198: 0.5856\n","Pearson Coeff in Epoch 198: -0.01277005291088132\n","Average Train Loss in Epoch 199: 0.1955\n","Average Test Loss in Epoch 199: 0.9363\n","Average Test Acc in Epoch 199: 0.6088\n","Pearson Coeff in Epoch 199: -0.004917070464038325\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"OV-ZxbuwBw76","colab_type":"code","outputId":"60493bb2-a76d-4710-e546-2a6847e56149","executionInfo":{"status":"ok","timestamp":1582148545399,"user_tz":0,"elapsed":63214,"user":{"displayName":"Andy Wang","photoUrl":"","userId":"02776860930356410397"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["# plot stats\n","plt.plot(range(len(train_losses)), train_losses)\n","plt.xlabel(\"Epochs\")\n","plt.title('Train losses')\n","plt.show()\n","\n","plt.plot(range(len(test_losses)), test_losses)\n","plt.xlabel(\"Epochs\")\n","plt.title('Test losses')\n","plt.show()\n","\n","plt.plot(range(len(test_acc)), test_acc)\n","plt.xlabel(\"Epochs\")\n","plt.title('Test acc')\n","plt.show()\n","\n","plt.plot(range(len(test_pearson)), test_pearson)\n","plt.xlabel(\"Epochs\")\n","plt.title('Test pearson')\n","plt.show()\n"],"execution_count":50,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEWCAYAAABollyxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0BtjjJ+zoDfGGD/3/8CuQEXPVI/PAAAAAElFTkSuQmCC\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEWCAYAAABollyxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGGMCnAV6Y4wJcBbojTEmwP1/dh0WAyOngfEAAAAASUVORK5CYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEWCAYAAABollyxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM7PE/R+0FXyXb7USlgAAAABJRU5ErkJggg==\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]}]}