diff --git a/lSTM-with-confusion-end.ipynb b/lSTM-with-confusion-end.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..a06df6bb80536fadf894666d579779c5a497b254
--- /dev/null
+++ b/lSTM-with-confusion-end.ipynb
@@ -0,0 +1 @@
+{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"tpu1vmV38","dataSources":[],"dockerImageVersionId":30699,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np\nimport torch\nimport torchtext\nimport seaborn as sns\nfrom matplotlib import pyplot as plt\nfrom sklearn.metrics import confusion_matrix\n\n# Training hyperparameters\nBATCH_SIZE: int = 1\nN_EPOCHS = 30\n\nSEED = 1234\nDEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\ntorch.manual_seed(SEED)\ntorch.backends.cudnn.deterministic = True\n\nprint(\"PyTorch Version: \", torch.__version__)\nprint(\"torchtext Version: \", torchtext.__version__)\nprint(f\"Using {'GPU' if str(DEVICE) == 'cuda' else 'CPU'}.\")\n\n\"\"\"## Preparing the Data\"\"\"\n\n# Load dataset\n# train_data, test_data = IMDB(root=\"./\", split=(\"train\", \"test\"))\nfrom datasets import load_dataset, load_metric\nfrom transformers import AutoTokenizer, RobertaTokenizerFast\nimport torch\nfrom torch.utils.data import DataLoader, Dataset\nimport torch.nn as nn\n\ndataset = load_dataset(\"surrey-nlp/PLOD-CW\")\ntrain_data = dataset[\"train\"]\nvalid_data = dataset[\"validation\"]\n\ntest_data = dataset[\"test\"]\n\nlabel_encoding = {\"B-O\": 0, \"B-AC\": 1, \"B-LF\": 2, \"I-LF\": 3}\n\nlabel_list = []\nfor sample in train_data[\"ner_tags\"]:\n    label_list.append([label_encoding[tag] for tag in sample])\n\nval_label_list = []\nfor sample in valid_data[\"ner_tags\"]:\n    val_label_list.append([label_encoding[tag] for tag in sample])\n\ntest_label_list = []\nfor sample in test_data[\"ner_tags\"]:\n    test_label_list.append([label_encoding[tag] for tag in sample])\n\n# tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')  # FixMe: Could produce bugs in the script\ntokenizer = RobertaTokenizerFast.from_pretrained(\"roberta-base\", add_prefix_space=True)\n\n\ndef tokenize_and_align_labels(short_dataset, list_name):\n    tokenized_inputs = tokenizer(short_dataset[\"tokens\"], truncation=True,\n                                 is_split_into_words=True)  ## For some models, you may need to set max_length to approximately 500.\n\n    labels = []\n    for i, label in enumerate(list_name):\n        word_ids = tokenized_inputs.word_ids(batch_index=i)\n        previous_word_idx = None\n        label_ids = []\n        for word_idx in word_ids:\n            # Special tokens have a word id that is None. We set the label to -100 so they are automatically\n            # ignored in the loss function.\n            if word_idx is None:\n                label_ids.append(-100)\n            # We set the label for the first token of each word.\n            elif word_idx != previous_word_idx:\n                label_ids.append(label[word_idx])\n            # For the other tokens in a word, we set the label to either the current label or -100, depending on\n            # the label_all_tokens flag.\n            else:\n                label_ids.append(label[word_idx])\n            previous_word_idx = word_idx\n\n        labels.append(label_ids)\n\n    tokenized_inputs[\"labels\"] = labels\n    return tokenized_inputs\n\n\ntokenized_datasets = tokenize_and_align_labels(train_data, label_list)\ntokenized_val_datasets = tokenize_and_align_labels(valid_data, val_label_list)\ntokenized_test_datasets = tokenize_and_align_labels(test_data, test_label_list)\n\n\nclass NERDataset(Dataset):\n    def __init__(self, encodings):\n        self.encodings = encodings\n\n    def __getitem__(self, idx):\n        label_tensor: torch.Tensor = torch.tensor(self.encodings[\"labels\"][idx])\n        input_id_tensor = torch.tensor(self.encodings[\"input_ids\"][idx])\n        attention_mask_tensor = torch.tensor(self.encodings[\"attention_mask\"][idx])\n        return input_id_tensor, attention_mask_tensor, label_tensor\n\n    def __len__(self):\n        return len(self.encodings[\"input_ids\"])\n\n\ntrain_dataset = NERDataset(tokenized_datasets)\nval_dataset = NERDataset(tokenized_val_datasets)\ntrain_loader = DataLoader(train_dataset, batch_size=BATCH_SIZE, shuffle=True)\nval_loader = DataLoader(val_dataset, batch_size=BATCH_SIZE, shuffle=True)\n\n\"\"\"Just like last time we'll create some useful utilities for processing pipelines so we can tokenize with spaCy and get the lengths post-tokenization to use packed padded sequences.\"\"\"\n\nfrom torchtext.data.utils import get_tokenizer\n\n\nclass SpacyTokenizer(torch.nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.tokenizer = get_tokenizer(\"spacy\", language=\"en_core_web_sm\")\n\n    def forward(self, input):\n        if isinstance(input, list):\n            tokens = []\n            for text in input:\n                tokens.append(self.tokenizer(text))\n            return tokens\n        elif isinstance(input, str):\n            return self.tokenizer(input)\n        raise ValueError(f\"Type {type(input)} is not supported.\")\n\n\nclass ToLengths(torch.nn.Module):\n    def forward(self, input):\n        if isinstance(input[0], list):\n            lengths = []\n            for text in input:\n                lengths.append(len(text))\n            return lengths\n        elif isinstance(input, list):\n            return len(input)\n        raise ValueError(f\"Type {type(input)} is not supported.\")\n\n\n\"\"\"Next is the use of pre-trained word embeddings. Instead of building a vocabulary and then generating word embeddings for each vocabulary token, with the word embeddings being initialized randomly, we will use pre-trained vectors.\n\nWe get these vectors by specifying which vectors we want and passing it as an argument to `torchtext.vocab.Vectors` subclasses `vocab.GloVe`, `vocab.FastText` or `vocab.CharNGram`. `TorchText` handles downloading the vectors.\n\nHere, we'll be using the GloVe `\"6B\"` vectors. GloVe is the algorithm used to calculate the vectors, go [here](https://nlp.stanford.edu/projects/glove/) for more. `6B` indicates these vectors were trained on 6 billion tokens. We wil also specify that we want these vectors to be $100$-dimensional.\n\nYou can see the other available vectors [here](https://pytorch.org/text/stable/vocab.html#pretrained-word-embeddings).\n\nThe theory is that these pre-trained vectors already have words with similar semantic meaning close together in vector space, e.g. \"terrible\", \"awful\", \"dreadful\" are nearby. This gives our embedding layer a good initialization as it does not have to learn these relations from scratch.\n\n**Note**: these vectors are about 862MB, so watch out if you have a limited internet connection.\n\"\"\"\n\nfrom torchtext import vocab\n\nMAX_VOCAB_SIZE = 25_000\n\nglove_vectors = vocab.GloVe(\n    name=\"6B\",\n    dim=100,\n    max_vectors=MAX_VOCAB_SIZE\n)\n\n\"\"\" from torchtext.vocab import FastText\nDiffrent exprement\nMAX_VOCAB_SIZE = 25_000\n\nfasttext_vectors = FastText(\n    language='en',  # Language for embeddings\n    max_vectors=MAX_VOCAB_SIZE\n)\"\"\"\n\n\"\"\"So now that we have the vectors downloaded, how do we *actually* use them?\n\nIn torchtext, `Vectors` objects are wrappers over a particular vocabulary (accessed with the `.stoi` member) and associated pre-trained word embeddings (accessed with the `.vectors` member). We want to decouple this into a vocabulary we can use in our data processing pipelines, and embeddings we can feed to the `Embedding` layer in our model so they are not initialized randomly.\n\nAs our `<unk>` and `<pad>` token aren't in the pre-trained vocabulary it is preferable to initialize them both to all zeros to explicitly tell our model that, initially, they are irrelevant for determining sentiment.\n\nWe do this by manually setting their row in the embedding matrix to zeros.\n\"\"\"\n\nfrom torchtext.vocab import vocab\n\ntext_vocab = vocab(glove_vectors.stoi, min_freq=0, specials=(\"<unk>\", \"<pad>\"), special_first=True)\ntext_vocab.set_default_index(text_vocab[\"<unk>\"])\n\npretrained_embeddings = glove_vectors.vectors\npretrained_embeddings = torch.cat([\n    torch.empty(1, glove_vectors.dim).normal_(),  # unk token vector\n    torch.zeros(1, glove_vectors.dim),  # padding token vector\n    pretrained_embeddings\n])\n\nprint(\"Vocab size: \", len(text_vocab))\nprint(\"Pretrained vectors shape: \", pretrained_embeddings.shape)\nprint(\"<unk> vector: \", pretrained_embeddings[text_vocab[\"<unk>\"]])\nprint(\"<pad> vector: \", pretrained_embeddings[text_vocab[\"<pad>\"]])\n\n\"\"\"Let's also define our label vocabulary.\"\"\"\n\nfrom collections import OrderedDict\n\nlabel_vocab = vocab(OrderedDict([(\"B-O\", 0), (\"B-AC\", 1), (\"B-LF\", 2), (\"I-LF\", 3)]))\n\n\"\"\"We can now define the rest of our pipelines like last time.\"\"\"\n\nimport torchtext.transforms as T\n\ntext_transform = T.Sequential(\n    SpacyTokenizer(),  # Tokenize\n    T.VocabTransform(text_vocab),  # Conver to vocab IDs\n    T.ToTensor(padding_value=text_vocab[\"<pad>\"]),  # Convert to tensor and pad\n)\n\nlabel_transform = T.Sequential(\n    T.LabelToIndex(label_vocab.get_itos()),  # Convert to integer\n    T.ToTensor(),  # Convert to tensor\n)\n\nlengths_transform = T.Sequential(\n    SpacyTokenizer(),\n    ToLengths(),\n    T.ToTensor(),\n)\n\nimport torch.nn as nn\n\n\nclass RNN(nn.Module):\n    def __init__(self, pretrained_embeddings, hidden_dim, output_dim, n_layers, bidirectional, dropout, pad_idx):\n        super().__init__()\n\n        self.num_directions = 2 if bidirectional else 1\n\n        self.embedding = nn.Embedding.from_pretrained(pretrained_embeddings, freeze=True, padding_idx=pad_idx)\n        self.rnn = nn.LSTM(pretrained_embeddings.shape[1],\n                           hidden_dim,\n                           num_layers=n_layers,\n                           bidirectional=bidirectional,\n                           dropout=dropout)\n        self.fc = nn.Linear(hidden_dim * self.num_directions, output_dim)\n\n        self.dropout = nn.Dropout(dropout)\n\n    def forward(self, text, lengths):\n        embedded = self.dropout(self.embedding(text))  # VV note that lengths need to be on the CPU\n        packed_embedded = nn.utils.rnn.pack_padded_sequence(embedded, lengths.cpu(), batch_first=True,\n                                                            enforce_sorted=False)\n\n        packed_output, (hidden, cell) = self.rnn(packed_embedded)\n\n        if self.num_directions == 2:  # if bidirectional\n            # Concat the final forward (hidden[-2,:,:]) and backward (hidden[-1,:,:]) hidden layers\n            # and apply dropout\n            hidden = self.dropout(torch.cat((hidden[-2, :, :], hidden[-1, :, :]), dim=1))\n        else:\n            hidden = self.dropout(hidden[-1, :, :])\n\n        return self.fc(hidden)\n\n\n\"\"\"Like before, we'll create an instance of our RNN class, with the new parameters and arguments for the number of layers, bidirectionality and dropout probability.\n\nTo ensure the pre-trained vectors are loaded into the model, we pass the decoupled vectors (`pretrained_embeddings` which we created earlier) to it.\n\nFinally, we get our pad token index from the vocabulary.\n\"\"\"\n\nHIDDEN_DIM = 256\nOUTPUT_DIM = 1\nN_LAYERS = 2\nBIDIRECTIONAL = True\nDROPOUT = 0.5\nPAD_IDX = text_vocab[\"<pad>\"]\n\nmodel = RNN(\n    pretrained_embeddings,\n    HIDDEN_DIM,\n    OUTPUT_DIM,\n    N_LAYERS,\n    BIDIRECTIONAL,\n    DROPOUT,\n    PAD_IDX\n)\n\n\"\"\"We'll print out the number of parameters in our model.\"\"\"\n\n\ndef count_parameters(model):\n    return sum(p.numel() for p in model.parameters() if p.requires_grad)\n\n\nprint(f'The model has {count_parameters(model):,} trainable parameters')\n\n\"\"\"## Train the Model\n\nNow to training the model.\n\nThe only change we'll make here is changing the optimizer from `SGD` to `Adam`. SGD updates all parameters with the same learning rate and choosing this learning rate can be tricky. `Adam` adapts the learning rate for each parameter, giving parameters that are updated more frequently lower learning rates and parameters that are updated infrequently higher learning rates. More information about `Adam` (and other optimizers) can be found [here](http://ruder.io/optimizing-gradient-descent/index.html).\n\nTo change `SGD` to `Adam`, we simply change `optim.SGD` to `optim.Adam`, also note how we do not have to provide an initial learning rate for Adam as PyTorch specifies a sensibile default initial learning rate.\n\"\"\"\n\nimport torch.optim as optim\n\n\"\"\"The rest of the steps for training the model are unchanged.\n\nWe define the criterion and place the model and criterion on the GPU (if available)...\n\"\"\"\n\ncriterion = nn.CrossEntropyLoss(ignore_index=-100)\n\n\"\"\"We implement the function to calculate accuracy...\"\"\"\n\n\ndef calculate_accuracy(prediction_tensors: torch.Tensor, ground_truth_tensor: torch.Tensor):\n    \"\"\"\n    Returns accuracy per batch, i.e. if you get 8/10 right, this returns 0.8, NOT 8\n    \"\"\"\n\n    # round predictions to the closest integer\n    # rounded_preds = torch.round(torch.sigmoid(preds))\n    preds = torch.argmax(prediction_tensors, dim=1)\n    correct = (preds == ground_truth_tensor).float()  # convert into float for division\n    acc: torch.Tensor = correct.sum() / len(correct)\n\n    # cm = confusion_matrix(ground_truth_tensor.numpy(), preds.numpy())\n\n    return acc\n\n\ndef train(model, iterator, optimizer, criterion, scheduler):\n    epoch_loss = 0\n    epoch_acc = 0\n    all_predictions, all_labels = [], []\n    model.train().to(DEVICE)\n\n    for batch in tqdm(iterator):\n        optimizer.zero_grad()\n        input_id_tensor, attention_mask_tensor, label_tensor = batch  # Note that this has to match the order in collate_batch\n        label_tensor = label_tensor.to(DEVICE)\n        outputs = model(input_id_tensor.to(DEVICE))\n        outputs = outputs.view(-1, outputs.shape[-1])  # Flatten output for loss calculation\n        label_tensor = label_tensor.view(-1)  # Flatten labels\n        # Backward pass and optimization\n        # optimizer.zero_grad()\n        loss = criterion(outputs, label_tensor)\n        loss.backward()\n        optimizer.step()\n\n        # Determine the model accuracy\n        batch_accuracy = calculate_accuracy(outputs, label_tensor)\n        epoch_loss += loss.item()\n        epoch_acc += batch_accuracy.item()\n\n        all_predictions.extend(torch.argmax(outputs, dim=1).cpu().numpy())\n        all_labels.extend(label_tensor.cpu().numpy())\n\n    scheduler.step()\n    computed_learning_rate = optimizer.param_groups[0]['lr']\n    print(f\"Epoch {epoch}, Train Loss: {epoch_loss / len(iterator)}\")\n    print(f\"Epoch {epoch}, Train Accuracy: {epoch_acc / len(iterator)}\")\n    print(f\"Epoch {epoch + 1}, Current Learning Rate: {computed_learning_rate}\")\n    return epoch_loss / len(iterator), epoch_acc / len(iterator), scheduler, all_predictions, all_labels\n\n\n\"\"\"Then we define a function for testing our model.\n\n**Note**: as we are now using dropout, we must remember to use `model.eval()` to ensure the dropout is \"turned off\" while evaluating.\n\"\"\"\n\ndef evaluation(model, iterator, optimizer, criterion):\n    epoch_loss = 0\n    epoch_acc = 0\n    all_predictions, all_labels = [], []\n    model.eval().to(DEVICE)\n\n    for batch in tqdm(iterator):\n        optimizer.zero_grad()\n        input_id_tensor, attention_mask_tensor, label_tensor = batch  # Note that this has to match the order in collate_batch\n        label_tensor = label_tensor.to(DEVICE)\n        outputs = model(input_id_tensor.to(DEVICE))\n        \n        \n        \n        outputs = outputs.view(-1, outputs.shape[-1])  # Flatten output for loss calculation\n        label_tensor = label_tensor.view(-1)  # Flatten labels\n        # Backward pass and optimization\n        # optimizer.zero_grad()\n        loss = criterion(outputs, label_tensor)\n        \n\n        # Determine the model accuracy\n        batch_accuracy = calculate_accuracy(outputs, label_tensor)\n        epoch_loss += loss.item()\n        epoch_acc += batch_accuracy.item()\n\n        all_predictions.extend(torch.argmax(outputs, dim=1).cpu().numpy())\n        all_labels.extend(label_tensor.cpu().numpy())\n\n    print(f\"Epoch {epoch}, Val Loss: {epoch_loss / len(iterator)}\")\n    print(f\"Epoch {epoch}, Val Accuracy: {epoch_acc / len(iterator)}\")\n    return epoch_loss / len(iterator), epoch_acc / len(iterator), scheduler, all_predictions, all_labels\n\n\nfrom tqdm import tqdm\n\n\ndef evaluate(model, iterator, criterion):\n    epoch_loss = 0\n    epoch_acc = 0\n\n    model.eval()\n\n    with torch.no_grad():\n        for batch in tqdm(iterator, desc=\"\\tEvaluation\"):\n            labels, texts, lengths = batch  # Note that this has to match the order in collate_batch\n            predictions = model(texts, lengths).squeeze(1)\n            loss = criterion(predictions, labels)\n            acc = calculate_accuracy(predictions, labels)\n\n            epoch_loss += loss.item()\n            epoch_acc += acc.item()\n\n    return epoch_loss / len(iterator), epoch_acc / len(iterator)\n\n\n\"\"\"And also create a nice function to tell us how long our epochs are taking.\"\"\"\n\nimport time\n\n\ndef epoch_time(start_time, end_time):\n    elapsed_time = end_time - start_time\n    elapsed_mins = int(elapsed_time / 60)\n    elapsed_secs = int(elapsed_time - (elapsed_mins * 60))\n    return elapsed_mins, elapsed_secs\n\n\n\"\"\"Finally, we train our model...\"\"\"\n\nbest_valid_loss = float('inf')\nprint(f\"Using {'GPU' if str(DEVICE) == 'cuda' else 'CPU'} for training.\")\n\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\n\n\ndef calculate_confusion_matrix(preds_: np.ndarray, ground_truths_: np.ndarray, data_portion_name: str = \"train\"):\n    cm = confusion_matrix(ground_truths_, preds_, labels=[0,1,2,3])\n    # Assuming 'cm' is your confusion matrix and 'class_names' are the names of the classes\n    class_names = ['B-O', 'B-AC', 'B-LF', 'I-LF']  # TODO: To be replaced by the actual names of the classes\n\n    plt.figure(figsize=(10, 7))\n    sns.heatmap(cm, annot=True, fmt='g', cmap='Blues', xticklabels=class_names, yticklabels=class_names)\n\n    plt.xlabel('Predicted')\n    plt.ylabel('True')\n    plt.title('Confusion Matrix')\n    plt.savefig(f'confusion_matrix_{data_portion_name}.png', format='png', dpi=300)\n    plt.show()\n\n    return cm\n\n\nclass LSTMForNER(nn.Module):\n    def __init__(self, vocab_size, embedding_dim, hidden_dim, num_labels):\n        super(LSTMForNER, self).__init__()\n        self.embedding = nn.Embedding(vocab_size, embedding_dim)\n        self.lstm = nn.LSTM(embedding_dim, hidden_dim, batch_first=True, bidirectional=True)\n        self.fc = nn.Linear(hidden_dim * 2, num_labels)  # Multiply by 2 for bidirectional\n\n    def forward(self, input_ids, attention_mask=None):\n        embedded = self.embedding(input_ids)\n        # Optionally use attention_mask to ignore padded areas in LSTM processing\n        lstm_output, _ = self.lstm(embedded)\n        logits = self.fc(lstm_output)\n        return logits\n\n\nvocab_size = tokenizer.vocab_size  # Assuming you have a tokenizer\nembedding_dim = 128\nhidden_dim = 256\nnum_labels = 4\nLEARNING_RATE: float = 5e-4\n\n# Instantiate the model\nmodel = LSTMForNER(vocab_size, embedding_dim, hidden_dim, num_labels)\noptimizer = optim.Adam(model.parameters(), lr=LEARNING_RATE)\nscheduler = optim.lr_scheduler.StepLR(optimizer, step_size=30, gamma=0.1)\nmodel = model.to(DEVICE)\ncriterion = criterion.to(DEVICE)\ntrain_loss_list = []\nbest_train_loss = np.inf\n\n\nfor epoch in range(N_EPOCHS):\n    print(f'Epoch: {epoch + 1:02}')\n\n    train_loss, train_acc, scheduler, preds, ground_truths = train(model, train_loader,\n                                                                   optimizer, criterion, scheduler)\n    val_loss, val_acc, _, val_preds, val_ground_truths =evaluation(model, val_loader, optimizer, criterion)\n    \n#added  for computing  to save the model\nif train_loss < best_train_loss:\n        best_train_loss = train_loss\n        torch.save(model.state_dict(), 'lstm_model.pt')\n        \n# Calculate the confusion matrix\nconfusion_matrix_train = calculate_confusion_matrix(preds, ground_truths, \"train\")\nconfusion_matrix_train = calculate_confusion_matrix(val_preds, val_ground_truths, \"val\")","metadata":{"execution":{"iopub.status.busy":"2024-05-02T08:32:09.623601Z","iopub.execute_input":"2024-05-02T08:32:09.623966Z","iopub.status.idle":"2024-05-02T08:39:20.515991Z","shell.execute_reply.started":"2024-05-02T08:32:09.623937Z","shell.execute_reply":"2024-05-02T08:39:20.515060Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"PyTorch Version:  2.1.2\ntorchtext Version:  0.16.2\nUsing GPU.\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Downloading readme:   0%|          | 0.00/8.37k [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"af943ab1e0a64dcf89f1d1a87378bd5d"}},"metadata":{}},{"name":"stderr","text":"Downloading data: 100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 188k/188k [00:00<00:00, 946kB/s]\nDownloading data: 100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 28.4k/28.4k [00:00<00:00, 302kB/s]\nDownloading data: 100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 28.7k/28.7k [00:00<00:00, 295kB/s]\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Generating train split:   0%|          | 0/1072 [00:00<?, ? examples/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"2b683cc47fbc4b27bd75f52ff4cad5aa"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Generating validation split:   0%|          | 0/126 [00:00<?, ? examples/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"b3c217c6f04242b4a16c592b62459550"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Generating test split:   0%|          | 0/153 [00:00<?, ? examples/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"02ea8fbca81c436d98606c0ac8a202a2"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenizer_config.json:   0%|          | 0.00/25.0 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"b78e7aae9e414ddaa90e0216da797ab4"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"vocab.json:   0%|          | 0.00/899k [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"79fd7dde5f8c488db93d87c61671178e"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"merges.txt:   0%|          | 0.00/456k [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"7474aa6dfbe04dcaa35da9e837229543"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenizer.json:   0%|          | 0.00/1.36M [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"c09e86111d7542598097d49e52a2349a"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"config.json:   0%|          | 0.00/481 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"a111f390c8b742f9a6bf6e8f59f82fbb"}},"metadata":{}},{"name":"stderr","text":".vector_cache/glove.6B.zip: 862MB [02:40, 5.39MB/s]                               \n100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–‰| 24999/25000 [00:01<00:00, 13891.97it/s]\n","output_type":"stream"},{"name":"stdout","text":"Vocab size:  25002\nPretrained vectors shape:  torch.Size([25002, 100])\n<unk> vector:  tensor([-0.1117, -0.4966,  0.1631, -0.8817,  0.0539,  0.6684, -0.0597, -0.4675,\n        -0.2153,  0.8840, -0.7584, -0.3689, -0.3424, -1.4020,  0.3206, -1.0219,\n         0.7988, -0.0923, -0.7049, -1.6024,  0.2891,  0.4899, -0.3853, -0.7120,\n        -0.1706, -1.4594,  0.2207,  0.2463, -1.3248,  0.6970, -0.6631,  1.2158,\n        -1.4949,  0.8810, -1.1786, -0.9340, -0.5675, -0.2772, -2.1834,  0.3668,\n         0.9380,  0.0078, -0.3139, -1.1567,  1.8409, -1.0174,  1.2192,  0.1601,\n         1.5985, -0.0469, -1.5270, -2.0143, -1.5173,  0.3877, -1.1849,  0.6897,\n         1.3232,  1.8169,  0.6808,  0.7244,  0.0323, -1.6593, -1.8773,  0.7372,\n         0.9257,  0.9247,  0.1825, -0.0737,  0.3147, -1.0369,  0.2100,  0.6144,\n         0.0628, -0.3297, -1.7970,  0.8728,  0.7670, -0.1138, -0.9428,  0.7540,\n         0.1407, -0.6937, -0.6159, -0.7295,  1.3204,  1.5997, -1.0792, -0.3396,\n        -1.4538, -2.6740,  1.5984,  0.8021,  0.5722,  0.0653, -0.0235,  0.8876,\n         1.4689,  1.2647, -0.2753, -0.1325])\n<pad> vector:  tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n        0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n        0., 0., 0., 0.])\nThe model has 2,310,657 trainable parameters\nUsing GPU for training.\nEpoch: 01\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:08<00:00, 123.85it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Train Loss: 0.5378376107170757\nEpoch 0, Train Accuracy: 0.7637579962611198\nEpoch 1, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 476.47it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Val Loss: 0.43633632580675774\nEpoch 0, Val Accuracy: 0.8020249538951449\nEpoch: 02\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Train Loss: 0.3542693723788061\nEpoch 1, Train Accuracy: 0.821125294832485\nEpoch 2, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 472.68it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Val Loss: 0.40308288271938053\nEpoch 1, Val Accuracy: 0.8094576183292601\nEpoch: 03\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.03it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Train Loss: 0.245138624473893\nEpoch 2, Train Accuracy: 0.859023123590359\nEpoch 3, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 476.97it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Val Loss: 0.40182039046305273\nEpoch 2, Val Accuracy: 0.8186003423872448\nEpoch: 04\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.75it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Train Loss: 0.15904736903981323\nEpoch 3, Train Accuracy: 0.8908468737030653\nEpoch 4, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 482.78it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Val Loss: 0.3922580880009466\nEpoch 3, Val Accuracy: 0.8220328357484605\nEpoch: 05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.67it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Train Loss: 0.09476250534430365\nEpoch 4, Train Accuracy: 0.9122282597444841\nEpoch 5, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 491.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Val Loss: 0.4187763231836023\nEpoch 4, Val Accuracy: 0.8250781728161706\nEpoch: 06\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.18it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Train Loss: 0.058220347971786296\nEpoch 5, Train Accuracy: 0.9248734261117765\nEpoch 6, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 487.44it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Val Loss: 0.5489389347004896\nEpoch 5, Val Accuracy: 0.8175863611792761\nEpoch: 07\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.11it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Train Loss: 0.031926905943439055\nEpoch 6, Train Accuracy: 0.9339040190315069\nEpoch 7, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 472.49it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Val Loss: 0.5828908763766761\nEpoch 6, Val Accuracy: 0.8134996467639529\nEpoch: 08\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.67it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Train Loss: 0.02033462768020346\nEpoch 7, Train Accuracy: 0.9375229229344361\nEpoch 8, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 498.70it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Val Loss: 0.6006780061094711\nEpoch 7, Val Accuracy: 0.812131943920302\nEpoch: 09\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.53it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Train Loss: 0.013233047518459539\nEpoch 8, Train Accuracy: 0.9394805885628977\nEpoch 9, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 445.33it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Val Loss: 0.6171971858166794\nEpoch 8, Val Accuracy: 0.8173968176993113\nEpoch: 10\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.82it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Train Loss: 0.007332582025128908\nEpoch 9, Train Accuracy: 0.9410677894727507\nEpoch 10, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 500.13it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Val Loss: 0.7318806546769799\nEpoch 9, Val Accuracy: 0.811647787926689\nEpoch: 11\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.20it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Train Loss: 0.007478771130815606\nEpoch 10, Train Accuracy: 0.940814954798613\nEpoch 11, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 481.10it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Val Loss: 0.76983389327668\nEpoch 10, Val Accuracy: 0.8151468733946482\nEpoch: 12\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.86it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Train Loss: 0.013028364628756055\nEpoch 11, Train Accuracy: 0.9388737696320263\nEpoch 12, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 487.48it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Val Loss: 0.7426299993687916\nEpoch 11, Val Accuracy: 0.8211086619467962\nEpoch: 13\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.87it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Train Loss: 0.002948059140625031\nEpoch 12, Train Accuracy: 0.9419927316815105\nEpoch 13, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 475.58it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Val Loss: 0.7678524402509783\nEpoch 12, Val Accuracy: 0.8185589857517727\nEpoch: 14\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.20it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Train Loss: 0.005475965297017784\nEpoch 13, Train Accuracy: 0.9416941048494026\nEpoch 14, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 495.69it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Val Loss: 0.7603848825522055\nEpoch 13, Val Accuracy: 0.8224161631531186\nEpoch: 15\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 146.54it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Train Loss: 0.0032841120088824296\nEpoch 14, Train Accuracy: 0.9420272960369267\nEpoch 15, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 491.96it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Val Loss: 0.8330103429177561\nEpoch 14, Val Accuracy: 0.8209936942846056\nEpoch: 16\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.11it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Train Loss: 0.005000714661248532\nEpoch 15, Train Accuracy: 0.9413787947773044\nEpoch 16, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 451.32it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Val Loss: 0.8684990309796247\nEpoch 15, Val Accuracy: 0.8195822588981144\nEpoch: 17\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.86it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Train Loss: 0.0045358700730734995\nEpoch 16, Train Accuracy: 0.9415207223883316\nEpoch 17, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 451.00it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Val Loss: 0.8449212101978547\nEpoch 16, Val Accuracy: 0.822124484512541\nEpoch: 18\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.36it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Train Loss: 0.0058594094509109\nEpoch 17, Train Accuracy: 0.9411018344559776\nEpoch 18, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 483.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Val Loss: 0.7832681014290771\nEpoch 17, Val Accuracy: 0.8233477142122057\nEpoch: 19\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.05it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Train Loss: 0.001846997112536294\nEpoch 18, Train Accuracy: 0.9421012358211759\nEpoch 19, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 405.14it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Val Loss: 0.807956216209046\nEpoch 18, Val Accuracy: 0.8239923531100863\nEpoch: 20\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.27it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Train Loss: 0.003568819639176546\nEpoch 19, Train Accuracy: 0.9418126366365311\nEpoch 20, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 493.91it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Val Loss: 0.8588715430619592\nEpoch 19, Val Accuracy: 0.8230729382189493\nEpoch: 21\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 151.23it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Train Loss: 0.0018788331402510771\nEpoch 20, Train Accuracy: 0.9421245405700669\nEpoch 21, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 445.39it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Val Loss: 0.8689107714654667\nEpoch 20, Val Accuracy: 0.8196191806641836\nEpoch: 22\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.06it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Train Loss: 0.0013674746341673111\nEpoch 21, Train Accuracy: 0.9422978197238339\nEpoch 22, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 490.99it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Val Loss: 0.845870930288728\nEpoch 21, Val Accuracy: 0.8213692875135512\nEpoch: 23\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.91it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Train Loss: 0.0017260686945525525\nEpoch 22, Train Accuracy: 0.9421808592307923\nEpoch 23, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 452.01it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Val Loss: 0.9360425495688105\nEpoch 22, Val Accuracy: 0.8202547790512206\nEpoch: 24\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.75it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Train Loss: 0.004007385134599921\nEpoch 23, Train Accuracy: 0.9414804940904254\nEpoch 24, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 497.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Val Loss: 0.8707952222395097\nEpoch 23, Val Accuracy: 0.8238974251444378\nEpoch: 25\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 153.87it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Train Loss: 0.0015232587040166195\nEpoch 24, Train Accuracy: 0.9421719765818831\nEpoch 25, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 496.76it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Val Loss: 0.8641515623132277\nEpoch 24, Val Accuracy: 0.820838998707514\nEpoch: 26\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.24it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Train Loss: 0.000570197188484729\nEpoch 25, Train Accuracy: 0.9425044051969229\nEpoch 26, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 502.13it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Val Loss: 0.886853411184434\nEpoch 25, Val Accuracy: 0.8274933605913132\nEpoch: 27\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 153.20it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Train Loss: 0.00040390731117412483\nEpoch 26, Train Accuracy: 0.9425652305692879\nEpoch 27, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 487.97it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Val Loss: 0.8894238545651046\nEpoch 26, Val Accuracy: 0.8313691890428937\nEpoch: 28\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Train Loss: 0.005877325156521041\nEpoch 27, Train Accuracy: 0.9416653525028655\nEpoch 28, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 489.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Val Loss: 0.8571161116957112\nEpoch 27, Val Accuracy: 0.8224501051600017\nEpoch: 29\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Train Loss: 0.000722844751019223\nEpoch 28, Train Accuracy: 0.9424859811899378\nEpoch 29, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 495.35it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Val Loss: 0.8992603149472704\nEpoch 28, Val Accuracy: 0.8289874988415885\nEpoch: 30\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.64it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Train Loss: 0.00026297410271528774\nEpoch 29, Train Accuracy: 0.9425799431307103\nEpoch 30, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 475.02it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Val Loss: 0.8697680176436922\nEpoch 29, Val Accuracy: 0.8304375805078991\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x700 with 2 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x700 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"learning_rate_list = []\nfor epoch in range(N_EPOCHS):\n    print(f'Epoch: {epoch + 1:02}')\n    train_loss, train_acc, scheduler, preds, ground_truths = train(model, train_loader,\n                                                                   optimizer, criterion, scheduler)\n    val_loss, val_acc, _, val_preds, val_ground_truths = evaluation(model, val_loader, optimizer, criterion)\n    \n    # Append the current epoch's loss and learning rate to their respective lists\n    train_loss_list.append(train_loss)\n    learning_rate_list.append(scheduler.get_last_lr()[0])  # Assuming you're interested in the last computed LR\n\n    # Update best train loss and save model if current loss is lower\n    if train_loss < best_train_loss:\n        best_train_loss = train_loss\n        torch.save(model.state_dict(), 'lstm_model.pt')\n\n# Plotting the training loss\nplt.figure(figsize=(10, 5))\nplt.plot(train_loss_list, label='Training Loss')\nplt.xlabel('Epoch')\nplt.ylabel('Loss')\nplt.title('Training Loss Over Epochs')\nplt.legend()\nplt.grid(True)\nplt.show()\n\n# Optionally, plot the learning rate evolution\nplt.figure(figsize=(10, 5))\nplt.plot(learning_rate_list, label='Learning Rate', color='red')\nplt.xlabel('Epoch')\nplt.ylabel('Learning Rate')\nplt.title('Learning Rate Over Epochs')\nplt.legend()\nplt.grid(True)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-02T08:40:58.921809Z","iopub.execute_input":"2024-05-02T08:40:58.922735Z","iopub.status.idle":"2024-05-02T08:44:42.125023Z","shell.execute_reply.started":"2024-05-02T08:40:58.922701Z","shell.execute_reply":"2024-05-02T08:44:42.124084Z"},"trusted":true},"execution_count":3,"outputs":[{"name":"stdout","text":"Epoch: 01\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.72it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Train Loss: 4.526616297180755e-05\nEpoch 0, Train Accuracy: 0.942625673015171\nEpoch 1, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 448.94it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Val Loss: 0.8867771479982323\nEpoch 0, Val Accuracy: 0.8323488762927433\nEpoch: 02\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.92it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Train Loss: 3.9669827986092675e-05\nEpoch 1, Train Accuracy: 0.942625673015171\nEpoch 2, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 498.18it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Val Loss: 0.8918501265377725\nEpoch 1, Val Accuracy: 0.8314725666765183\nEpoch: 03\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.80it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Train Loss: 3.4382006494212595e-05\nEpoch 2, Train Accuracy: 0.942625673015171\nEpoch 3, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 458.52it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Val Loss: 0.898444967603104\nEpoch 2, Val Accuracy: 0.8314255608452691\nEpoch: 04\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.89it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Train Loss: 2.8882143100514415e-05\nEpoch 3, Train Accuracy: 0.942625673015171\nEpoch 4, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 470.24it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Val Loss: 0.9081117287350935\nEpoch 3, Val Accuracy: 0.8312850722244808\nEpoch: 05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.69it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Train Loss: 2.3270712742212458e-05\nEpoch 4, Train Accuracy: 0.942625673015171\nEpoch 5, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 454.10it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Val Loss: 0.9250490475366989\nEpoch 4, Val Accuracy: 0.8320322661172777\nEpoch: 06\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 146.71it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Train Loss: 1.7906465676511818e-05\nEpoch 5, Train Accuracy: 0.942625673015171\nEpoch 6, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 483.29it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Val Loss: 0.9429263433579879\nEpoch 5, Val Accuracy: 0.8321595963031526\nEpoch: 07\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.48it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Train Loss: 1.313514025311006e-05\nEpoch 6, Train Accuracy: 0.942625673015171\nEpoch 7, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 489.96it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Val Loss: 0.9645917093318804\nEpoch 6, Val Accuracy: 0.8309820441026536\nEpoch: 08\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.23it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Train Loss: 9.220313894217086e-06\nEpoch 7, Train Accuracy: 0.942625673015171\nEpoch 8, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 486.15it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Val Loss: 0.9926202725192337\nEpoch 7, Val Accuracy: 0.8298234551671951\nEpoch: 09\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.42it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Train Loss: 6.242224222773434e-06\nEpoch 8, Train Accuracy: 0.942625673015171\nEpoch 9, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 440.08it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Val Loss: 1.0207844377372177\nEpoch 8, Val Accuracy: 0.8293291633091275\nEpoch: 10\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.43it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Train Loss: 4.1057451886393374e-06\nEpoch 9, Train Accuracy: 0.942625673015171\nEpoch 10, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 476.39it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Val Loss: 1.056980770354621\nEpoch 9, Val Accuracy: 0.8294156051817394\nEpoch: 11\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.15it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Train Loss: 2.649138292794388e-06\nEpoch 10, Train Accuracy: 0.942625673015171\nEpoch 11, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 486.64it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Val Loss: 1.0879872088905633\nEpoch 10, Val Accuracy: 0.82927082549958\nEpoch: 12\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.68it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Train Loss: 1.708642657113197e-06\nEpoch 11, Train Accuracy: 0.942625673015171\nEpoch 12, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 446.29it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Val Loss: 1.1229980312044872\nEpoch 11, Val Accuracy: 0.8291556858827197\nEpoch: 13\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.90it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Train Loss: 1.0924102828605547e-06\nEpoch 12, Train Accuracy: 0.942625673015171\nEpoch 13, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 483.92it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Val Loss: 1.1587616219500296\nEpoch 12, Val Accuracy: 0.829144971711295\nEpoch: 14\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 145.87it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Train Loss: 6.991456202282971e-07\nEpoch 13, Train Accuracy: 0.942625673015171\nEpoch 14, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 491.46it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Val Loss: 1.1929318488905882\nEpoch 13, Val Accuracy: 0.8292445445817614\nEpoch: 15\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.04it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Train Loss: 4.5061216416916365e-07\nEpoch 14, Train Accuracy: 0.942625673015171\nEpoch 15, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 428.59it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Val Loss: 1.2289685530072925\nEpoch 14, Val Accuracy: 0.8292137720282116\nEpoch: 16\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 151.05it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Train Loss: 2.933639295455865e-07\nEpoch 15, Train Accuracy: 0.942625673015171\nEpoch 16, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 495.48it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Val Loss: 1.2627835709305102\nEpoch 15, Val Accuracy: 0.8294254366367583\nEpoch: 17\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.31it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Train Loss: 1.9700575959085343e-07\nEpoch 16, Train Accuracy: 0.942625673015171\nEpoch 17, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 480.15it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Val Loss: 1.2945007570173805\nEpoch 16, Val Accuracy: 0.8292032267366137\nEpoch: 18\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.64it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Train Loss: 1.6527173846583077e-07\nEpoch 17, Train Accuracy: 0.942625673015171\nEpoch 18, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 460.93it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Val Loss: 1.3163788109969385\nEpoch 17, Val Accuracy: 0.83022926156483\nEpoch: 19\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 152.81it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Train Loss: 1.7777460030724146e-05\nEpoch 18, Train Accuracy: 0.942611539152576\nEpoch 19, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 497.03it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Val Loss: 1.3526700322150278\nEpoch 18, Val Accuracy: 0.8286873423863971\nEpoch: 20\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.43it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Train Loss: 1.1161666626519737e-07\nEpoch 19, Train Accuracy: 0.942625673015171\nEpoch 20, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 498.60it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Val Loss: 1.360815836476245\nEpoch 19, Val Accuracy: 0.8286873423863971\nEpoch: 21\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 155.89it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Train Loss: 9.600557573940325e-08\nEpoch 20, Train Accuracy: 0.942625673015171\nEpoch 21, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 481.76it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Val Loss: 1.369495807162651\nEpoch 20, Val Accuracy: 0.8287132931134057\nEpoch: 22\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 154.16it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Train Loss: 8.467038085779948e-08\nEpoch 21, Train Accuracy: 0.942625673015171\nEpoch 22, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 505.72it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Val Loss: 1.3779058292922823\nEpoch 21, Val Accuracy: 0.828629671108155\nEpoch: 23\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 157.00it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Train Loss: 7.529130604725939e-08\nEpoch 22, Train Accuracy: 0.942625673015171\nEpoch 23, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 486.57it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Val Loss: 1.3863400134637467\nEpoch 22, Val Accuracy: 0.8289712071418762\nEpoch: 24\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 155.57it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Train Loss: 6.72193361601882e-08\nEpoch 23, Train Accuracy: 0.942625673015171\nEpoch 24, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 492.87it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Val Loss: 1.3945708935095826\nEpoch 23, Val Accuracy: 0.828720264018528\nEpoch: 25\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Train Loss: 5.978162048963566e-08\nEpoch 24, Train Accuracy: 0.942625673015171\nEpoch 25, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 506.21it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Val Loss: 1.4026416455508766\nEpoch 24, Val Accuracy: 0.8290142085817125\nEpoch: 26\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.19it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Train Loss: 5.323785371274512e-08\nEpoch 25, Train Accuracy: 0.942625673015171\nEpoch 26, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 492.78it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Val Loss: 1.4115896797089393\nEpoch 25, Val Accuracy: 0.828956697668348\nEpoch: 27\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 155.17it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Train Loss: 4.742366287169286e-08\nEpoch 26, Train Accuracy: 0.942625673015171\nEpoch 27, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 499.49it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Val Loss: 1.4201813464536979\nEpoch 26, Val Accuracy: 0.8290984209567781\nEpoch: 28\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.50it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Train Loss: 4.198963294209773e-08\nEpoch 27, Train Accuracy: 0.942625673015171\nEpoch 28, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 494.30it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Val Loss: 1.4295534512957702\nEpoch 27, Val Accuracy: 0.8294065443296281\nEpoch: 29\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.09it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Train Loss: 3.7035178174112636e-08\nEpoch 28, Train Accuracy: 0.942625673015171\nEpoch 29, Current Learning Rate: 5e-06\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 492.08it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Val Loss: 1.439547942418317\nEpoch 28, Val Accuracy: 0.829253919067837\nEpoch: 30\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.85it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Train Loss: 3.373143861648263e-08\nEpoch 29, Train Accuracy: 0.942625673015171\nEpoch 30, Current Learning Rate: 5e-06\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 488.24it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Val Loss: 1.4404478640779927\nEpoch 29, Val Accuracy: 0.8294428832947262\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x500 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x500 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"\n\ntrain_loss_list = []\ntrain_accuracy_list = []\nlearning_rate_list = []\n\n# Training loop\nfor epoch in range(N_EPOCHS):\n    print(f'Epoch: {epoch + 1:02}')\n    train_loss, train_acc, scheduler, preds, ground_truths = train(model, train_loader,\n                                                                   optimizer, criterion, scheduler)\n    val_loss, val_acc, _, val_preds, val_ground_truths = evaluation(model, val_loader, optimizer, criterion)\n    \n    # Append the current epoch's metrics to their respective lists\n    train_loss_list.append(train_loss)\n    train_accuracy_list.append(train_acc)\n    learning_rate_list.append(scheduler.get_last_lr()[0])\n\n    # Update best train loss and save model if current loss is lower\n    if train_loss < best_train_loss:\n        best_train_loss = train_loss\n        torch.save(model.state_dict(), 'lstm_model.pt')\n\n# Creating a plot with two y-axes\nfig, ax1 = plt.subplots(figsize=(10, 5))\n\n# Loss (primary y-axis)\ncolor = 'tab:red'\nax1.set_xlabel('Epoch')\nax1.set_ylabel('Loss', color=color)\nax1.plot(train_loss_list, label='Training Loss', color=color)\nax1.tick_params(axis='y', labelcolor=color)\n\n# Creating a second y-axis for accuracy and learning rate\nax2 = ax1.twinx()\ncolor = 'tab:blue'\nax2.set_ylabel('Accuracy', color=color)\nax2.plot(train_accuracy_list, label='Training Accuracy', color=color)\nax2.tick_params(axis='y', labelcolor=color)\n\n# Plot learning rate\ncolor = 'tab:green'\nax3 = ax1.twinx()\nax3.spines['right'].set_position(('outward', 60))  # Offset the right spine of learning rate\nax3.set_ylabel('Learning Rate', color=color)\nax3.plot(learning_rate_list, label='Learning Rate', color=color, linestyle='--')\nax3.tick_params(axis='y', labelcolor=color)\n\nfig.tight_layout()  # Adjust layout to make room\nplt.title('Training Loss, Accuracy, and Learning Rate Over Epochs')\nfig.legend(loc='upper left', bbox_to_anchor=(0.1,0.9))\nplt.show()\n","metadata":{"execution":{"iopub.status.busy":"2024-05-02T08:47:18.628313Z","iopub.execute_input":"2024-05-02T08:47:18.628674Z","iopub.status.idle":"2024-05-02T08:51:02.724512Z","shell.execute_reply.started":"2024-05-02T08:47:18.628645Z","shell.execute_reply":"2024-05-02T08:51:02.723515Z"},"trusted":true},"execution_count":7,"outputs":[{"name":"stdout","text":"Epoch: 01\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.04it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Train Loss: 0.12637709077691073\nEpoch 0, Train Accuracy: 0.9021762258081294\nEpoch 1, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 464.62it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 0, Val Loss: 0.4707501535082147\nEpoch 0, Val Accuracy: 0.7993496862195787\nEpoch: 02\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.83it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Train Loss: 0.07394607583867653\nEpoch 1, Train Accuracy: 0.9204028287151856\nEpoch 2, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 486.81it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1, Val Loss: 0.4905132136847233\nEpoch 1, Val Accuracy: 0.8137810445494122\nEpoch: 03\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.36it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Train Loss: 0.03934009883164223\nEpoch 2, Train Accuracy: 0.9317641228215018\nEpoch 3, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 439.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 2, Val Loss: 0.5491843268290044\nEpoch 2, Val Accuracy: 0.8141449819954615\nEpoch: 04\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Train Loss: 0.023647315730032415\nEpoch 3, Train Accuracy: 0.9361785006389689\nEpoch 4, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 478.70it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 3, Val Loss: 0.5931537916702736\nEpoch 3, Val Accuracy: 0.812385328941875\nEpoch: 05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 146.14it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Train Loss: 0.017446020986497238\nEpoch 4, Train Accuracy: 0.9382210891527026\nEpoch 5, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 462.14it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 4, Val Loss: 0.6148029962833601\nEpoch 4, Val Accuracy: 0.8154035447135805\nEpoch: 06\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.65it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Train Loss: 0.011484432839770837\nEpoch 5, Train Accuracy: 0.9396362768402741\nEpoch 6, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 442.55it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 5, Val Loss: 0.624878207107233\nEpoch 5, Val Accuracy: 0.8207231492750229\nEpoch: 07\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.73it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Train Loss: 0.014191484782301434\nEpoch 6, Train Accuracy: 0.9390163217573914\nEpoch 7, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 468.08it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 6, Val Loss: 0.6374132011925602\nEpoch 6, Val Accuracy: 0.8250840755682143\nEpoch: 08\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.61it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Train Loss: 0.007859415084319869\nEpoch 7, Train Accuracy: 0.9406222276278396\nEpoch 8, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 458.05it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 7, Val Loss: 0.6547639571141363\nEpoch 7, Val Accuracy: 0.8247319400783569\nEpoch: 09\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.75it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Train Loss: 0.003971068978645973\nEpoch 8, Train Accuracy: 0.9418976700016811\nEpoch 9, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 474.46it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 8, Val Loss: 0.7038532176372846\nEpoch 8, Val Accuracy: 0.8218582066751662\nEpoch: 10\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.26it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Train Loss: 0.0016106216264629828\nEpoch 9, Train Accuracy: 0.9423387186851964\nEpoch 10, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 471.73it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 9, Val Loss: 0.7601627634701019\nEpoch 9, Val Accuracy: 0.8138368378082911\nEpoch: 11\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 151.12it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Train Loss: 0.0034164498401244334\nEpoch 10, Train Accuracy: 0.9419595852160632\nEpoch 11, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 450.57it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 10, Val Loss: 0.8296723172854168\nEpoch 10, Val Accuracy: 0.8187126221637877\nEpoch: 12\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.78it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Train Loss: 0.004409786703977375\nEpoch 11, Train Accuracy: 0.941476940219082\nEpoch 12, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 489.59it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 11, Val Loss: 0.841546752131156\nEpoch 11, Val Accuracy: 0.8132878577425366\nEpoch: 13\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 146.71it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Train Loss: 0.011768929960073933\nEpoch 12, Train Accuracy: 0.9399096446473207\nEpoch 13, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 451.30it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 12, Val Loss: 0.7776519457277443\nEpoch 12, Val Accuracy: 0.8120191972407084\nEpoch: 14\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 153.40it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Train Loss: 0.005324836878452054\nEpoch 13, Train Accuracy: 0.9412995500684674\nEpoch 14, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 488.18it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 13, Val Loss: 0.8195275481995186\nEpoch 13, Val Accuracy: 0.8134601714592131\nEpoch: 15\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:06<00:00, 156.43it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Train Loss: 0.004067076610205725\nEpoch 14, Train Accuracy: 0.9417454076569471\nEpoch 15, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 491.09it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 14, Val Loss: 0.8000858290428629\nEpoch 14, Val Accuracy: 0.8157070759269927\nEpoch: 16\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 151.31it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Train Loss: 0.000659985190820385\nEpoch 15, Train Accuracy: 0.9425356547667909\nEpoch 16, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 475.51it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 15, Val Loss: 0.8308243305884446\nEpoch 15, Val Accuracy: 0.8210152768426471\nEpoch: 17\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.93it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Train Loss: 0.00025879426539827884\nEpoch 16, Train Accuracy: 0.9426004206622715\nEpoch 17, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 483.39it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 16, Val Loss: 0.8692607111254238\nEpoch 16, Val Accuracy: 0.8195109866441243\nEpoch: 18\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.12it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Train Loss: 0.00011313466532997334\nEpoch 17, Train Accuracy: 0.9426187115596302\nEpoch 18, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 476.35it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 17, Val Loss: 0.8981347273625971\nEpoch 17, Val Accuracy: 0.8208083910128426\nEpoch: 19\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.94it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Train Loss: 5.197453480496589e-05\nEpoch 18, Train Accuracy: 0.942625673015171\nEpoch 19, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 495.35it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 18, Val Loss: 0.94086069273983\nEpoch 18, Val Accuracy: 0.8195377292614134\nEpoch: 20\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 147.58it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Train Loss: 3.07755656045088e-05\nEpoch 19, Train Accuracy: 0.942625673015171\nEpoch 20, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 496.57it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 19, Val Loss: 0.9751344906070742\nEpoch 19, Val Accuracy: 0.820261917653538\nEpoch: 21\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.74it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Train Loss: 1.883192660618672e-05\nEpoch 20, Train Accuracy: 0.942625673015171\nEpoch 21, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 441.31it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 20, Val Loss: 1.010414358890254\nEpoch 20, Val Accuracy: 0.8214580220362496\nEpoch: 22\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.24it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Train Loss: 1.132750412659912e-05\nEpoch 21, Train Accuracy: 0.942625673015171\nEpoch 22, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 487.63it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 21, Val Loss: 1.0573706247671777\nEpoch 21, Val Accuracy: 0.8218008234860406\nEpoch: 23\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.01it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Train Loss: 6.750763815175928e-06\nEpoch 22, Train Accuracy: 0.942625673015171\nEpoch 23, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 475.13it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 22, Val Loss: 1.097595301135871\nEpoch 22, Val Accuracy: 0.8208476581743785\nEpoch: 24\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 151.27it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Train Loss: 0.013183046820927437\nEpoch 23, Train Accuracy: 0.9386344691043469\nEpoch 24, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 449.99it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 23, Val Loss: 0.9198854343328368\nEpoch 23, Val Accuracy: 0.813454229916845\nEpoch: 25\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.48it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Train Loss: 0.00663350996297109\nEpoch 24, Train Accuracy: 0.9412101095180903\nEpoch 25, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 481.84it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 24, Val Loss: 0.9141175631115662\nEpoch 24, Val Accuracy: 0.8132069359223048\nEpoch: 26\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.23it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Train Loss: 0.00043721114992508766\nEpoch 25, Train Accuracy: 0.9425470577135905\nEpoch 26, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 454.67it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 25, Val Loss: 0.8918011858444342\nEpoch 25, Val Accuracy: 0.8150661012956074\nEpoch: 27\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 149.95it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Train Loss: 0.00010200586531773326\nEpoch 26, Train Accuracy: 0.942625673015171\nEpoch 27, Current Learning Rate: 0.0005\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 476.35it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 26, Val Loss: 0.9195845651534794\nEpoch 26, Val Accuracy: 0.815411778906035\nEpoch: 28\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 148.66it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Train Loss: 6.0550433013461745e-05\nEpoch 27, Train Accuracy: 0.942625673015171\nEpoch 28, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 468.69it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 27, Val Loss: 0.9385028392987604\nEpoch 27, Val Accuracy: 0.8161264168364661\nEpoch: 29\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 150.43it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Train Loss: 4.278971649632078e-05\nEpoch 28, Train Accuracy: 0.942625673015171\nEpoch 29, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 435.58it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 28, Val Loss: 0.9420799977601818\nEpoch 28, Val Accuracy: 0.8162702442634673\nEpoch: 30\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 1072/1072 [00:07<00:00, 146.56it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Train Loss: 4.055016197703367e-05\nEpoch 29, Train Accuracy: 0.942625673015171\nEpoch 30, Current Learning Rate: 5e-05\n","output_type":"stream"},{"name":"stderr","text":"100%|â–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆâ–ˆ| 126/126 [00:00<00:00, 495.71it/s]\n","output_type":"stream"},{"name":"stdout","text":"Epoch 29, Val Loss: 0.9472344348883376\nEpoch 29, Val Accuracy: 0.8159057637528767\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x500 with 3 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}
\ No newline at end of file