lectures/09_Transformers_BERT.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to Python and Natural Language Technologies\n",
    "\n",
    "__Lecture 9, Transformers, BERT__\n",
    "\n",
    "__April 13, 2021__\n",
    "\n",
    "__Judit Ács__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gc\n",
    "from IPython.display import Image\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "\n",
    "from transformers import pipeline\n",
    "from transformers import AutoTokenizer, AutoModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Attention mechanism\n",
    "\n",
    "Attention:\n",
    "- emphasizes the important part of the input\n",
    "- and de-emphasizes the rest.\n",
    "- Mimics cognitive attention.\n",
    "\n",
    "Method:\n",
    "- It does this by assigning weights to the elements of the input sequence.\n",
    "- The weights depend on the current context in the decoder:\n",
    "    - the current decoder hidden state,\n",
    "    - the previous output.\n",
    "- The source vectors are multiplied by the weights and then summed -> **context vector**\n",
    "- The context vector is used for predicting the next output symbol."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"img/dl/attention_mechanism.jpg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problems\n",
    "\n",
    "Recall that we used recurrent neural cells, specifically LSTMs to encode and decode sequences.\n",
    "\n",
    "__Problem 1. No parallelism__\n",
    "\n",
    "LSTMs are recurrent, they rely on their left and right history (horizontal arrows), so the symbols need to be processed in order -> no parallelism.\n",
    "\n",
    "__Problem 2. Long-range dependencies__\n",
    "\n",
    "Long-range dependencies are not infrequent in NLP.\n",
    "\n",
    "\"The **people/person** who called and wanted to rent your house when you go away next year **are/is** from California\" -- Miller & Chomsky 1963\n",
    "\n",
    "LSTMs have a problem capturing these because there are too many backpropagation steps between the symbols."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Transformers\n",
    "\n",
    "Introduced in [Attention Is All You Need](https://papers.nips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf) by Vaswani et al., 2017\n",
    "\n",
    "Transformers solve Problem 1 by relying purely on attention instead of recurrence.\n",
    "\n",
    "Not having recurrent connections means that sequence position no longer matters.\n",
    "\n",
    "Recurrence is replaced by **self attention**.\n",
    "\n",
    "Each symbol is encoded the following way:\n",
    "\n",
    "__Step 1__: the encoder 'looks' at the other symbols in the input sequence\n",
    "    - In the example above: the representation of **are/is** depends on **people/person** more than any other word in the sentence, it should receive the highest attention weight."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"http://jalammar.github.io/images/t/transformer_self-attention_visualization.png\", embed=True)  # from Illustrated Transformers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Step 2__: the context vector is passed through a feed-forward network which is shared across all symbols."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"http://jalammar.github.io/images/t/encoder_with_tensors.png\", embed=True)  # from Illustrated Transformers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This visualization is available in the [Tensor2tensor notebook in Google Colab](https://colab.research.google.com/github/tensorflow/tensor2tensor/blob/master/tensor2tensor/notebooks/hello_t2t.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Other components\n",
    "\n",
    "__Residual connections__\n",
    "\n",
    "- Also called __skip connections__\n",
    "- The output of a module is added to the input\n",
    "\n",
    "$$\n",
    "\\text{output} = \\text{layer}(\\text{input}) + \\text{input}\n",
    "$$\n",
    "\n",
    "__Softmax__\n",
    "\n",
    "- Only used in the decoder\n",
    "- Maps the output vector to a probability distribution\n",
    "    - In other words it tells us how likely each symbol is.\n",
    "\n",
    "## Multiple heads and layers\n",
    "\n",
    "Transformers have a number of additional components summarized in this figure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"img/dl/transformer.png\")  # from Vaswani et al. 2018"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PyTorch support\n",
    "\n",
    "PyTorch has a `nn.Transformer` class and its encoder and decoder versions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.nn import TransformerEncoder, TransformerEncoderLayer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TransformerEncoderLayer(\n",
       "  (self_attn): MultiheadAttention(\n",
       "    (out_proj): _LinearWithBias(in_features=12, out_features=12, bias=True)\n",
       "  )\n",
       "  (linear1): Linear(in_features=12, out_features=7, bias=True)\n",
       "  (dropout): Dropout(p=0.2, inplace=False)\n",
       "  (linear2): Linear(in_features=7, out_features=12, bias=True)\n",
       "  (norm1): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "  (norm2): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "  (dropout1): Dropout(p=0.2, inplace=False)\n",
       "  (dropout2): Dropout(p=0.2, inplace=False)\n",
       ")"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedding_dim = 12\n",
    "# num_heads = 5  # embedding_dim must be divisible by the number of heads\n",
    "num_heads = 2\n",
    "hidden_size = 7\n",
    "dropout = 0.2\n",
    "TransformerEncoderLayer(embedding_dim, num_heads, hidden_size, dropout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TransformerEncoder(\n",
       "  (layers): ModuleList(\n",
       "    (0): TransformerEncoderLayer(\n",
       "      (self_attn): MultiheadAttention(\n",
       "        (out_proj): _LinearWithBias(in_features=12, out_features=12, bias=True)\n",
       "      )\n",
       "      (linear1): Linear(in_features=12, out_features=7, bias=True)\n",
       "      (dropout): Dropout(p=0.2, inplace=False)\n",
       "      (linear2): Linear(in_features=7, out_features=12, bias=True)\n",
       "      (norm1): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "      (norm2): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "      (dropout1): Dropout(p=0.2, inplace=False)\n",
       "      (dropout2): Dropout(p=0.2, inplace=False)\n",
       "    )\n",
       "    (1): TransformerEncoderLayer(\n",
       "      (self_attn): MultiheadAttention(\n",
       "        (out_proj): _LinearWithBias(in_features=12, out_features=12, bias=True)\n",
       "      )\n",
       "      (linear1): Linear(in_features=12, out_features=7, bias=True)\n",
       "      (dropout): Dropout(p=0.2, inplace=False)\n",
       "      (linear2): Linear(in_features=7, out_features=12, bias=True)\n",
       "      (norm1): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "      (norm2): LayerNorm((12,), eps=1e-05, elementwise_affine=True)\n",
       "      (dropout1): Dropout(p=0.2, inplace=False)\n",
       "      (dropout2): Dropout(p=0.2, inplace=False)\n",
       "    )\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer = TransformerEncoderLayer(embedding_dim, num_heads, hidden_size, dropout)\n",
    "TransformerEncoder(layer, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([9, 3, 12])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoder = TransformerEncoder(layer, 2)\n",
    "\n",
    "sequence_len = 9\n",
    "batch_size = 3\n",
    "X = torch.rand((sequence_len, batch_size, embedding_dim))\n",
    "y = encoder(X)\n",
    "y.size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Positional encoding\n",
    "\n",
    "Without recurrence word order information is lost.\n",
    "\n",
    "Positional information is important:\n",
    "\n",
    "    John loves Mary.\n",
    "    Mary loves John.\n",
    "\n",
    "Transformers apply positional encoding:\n",
    "\n",
    "$$\n",
    "\\text{PE}_{\\text{pos},2i} = \\sin(\\frac{\\text{pos}}{10000^{2i/d_{\\text{model}}}}), \\\\\n",
    "\\text{PE}_{\\text{pos},2i+1} = \\cos(\\frac{\\text{pos}}{10000^{2i/d_{\\text{model}}}}),\n",
    "$$\n",
    "\n",
    "where:\n",
    "- $d_{\\text{model}}$ is the input dimension to the Transformer, usually the embedding size\n",
    "- $\\text{pos}$ is the position of the symbol in the input sequence i.e. first word, second word etc.\n",
    "- $i$ is the coordinate index in the input vector."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a position encoder in PyTorch.\n",
    "\n",
    "For $\\text{pos}=0$, the sine values are 0, and the cosine values are 1:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([1.]), tensor([0.]))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = torch.FloatTensor([0.])\n",
    "torch.cos(t), torch.sin(t)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For large $i$ values, the denominator is close to 10000, so it's again close to 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([445, 109, 251,  47,  17,  42, 412, 231, 155, 230])\n",
      "tensor([0.0705, 0.0173, 0.0398, 0.0074, 0.0027, 0.0067, 0.0653, 0.0366, 0.0246,\n",
      "        0.0365])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(tensor([0.9975, 0.9999, 0.9992, 1.0000, 1.0000, 1.0000, 0.9979, 0.9993, 0.9997,\n",
       "         0.9993]),\n",
       " tensor([0.0705, 0.0173, 0.0398, 0.0074, 0.0027, 0.0067, 0.0653, 0.0366, 0.0246,\n",
       "         0.0364]))"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Pick a few random values for pos\n",
    "pos = torch.randint(512, size=(10, ))\n",
    "print(pos)\n",
    "\n",
    "# Divide by 10000^2*i/d_model. Make 2*i/d_model close to one (high 2*i values)\n",
    "t = pos / (10000 ** 0.95)\n",
    "print(t)\n",
    "torch.cos(t), torch.sin(t)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's generate the full grid. There are $\\text{maxlen} \\times d_\\text{model}$ values.\n",
    "\n",
    "__maxlen__ is the maximum position we allow. This has to be predefined.\n",
    "\n",
    "__d_model__ is the size of the input, which is embedding_dim in most cases."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.float32, torch.Size([20, 12]))"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "maxlen = 20\n",
    "d_model = 12\n",
    "\n",
    "pe = torch.zeros((maxlen, d_model))\n",
    "pe.dtype, pe.size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__pos__ are the indices of the sequence from 0 to $\\text{maxlen}-1$:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([ 0.,  1.,  2.,  3.,  4.,  5.,  6.,  7.,  8.,  9., 10., 11., 12., 13.,\n",
       "        14., 15., 16., 17., 18., 19.])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos = torch.arange(maxlen, dtype=torch.float)\n",
    "pos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reminder:\n",
    "\n",
    "$$\n",
    "\\text{PE}_{\\text{pos},2i} = \\sin(\\frac{\\text{pos}}{10000^{2i/d_{\\text{model}}}}), \\\\\n",
    "\\text{PE}_{\\text{pos},2i+1} = \\cos(\\frac{\\text{pos}}{10000^{2i/d_{\\text{model}}}}),\n",
    "$$\n",
    "\n",
    "Let's define the denominator:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([1.0000e+00, 4.6416e+00, 2.1544e+01, 1.0000e+02, 4.6416e+02, 2.1544e+03])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "divterm = 10000 ** (torch.arange(0, d_model, step=2) / float(d_model))\n",
    "divterm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([20, 6])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor([[0.0000e+00, 0.0000e+00, 0.0000e+00, 0.0000e+00, 0.0000e+00, 0.0000e+00],\n",
       "        [1.0000e+00, 2.1544e-01, 4.6416e-02, 1.0000e-02, 2.1544e-03, 4.6416e-04],\n",
       "        [2.0000e+00, 4.3089e-01, 9.2832e-02, 2.0000e-02, 4.3089e-03, 9.2832e-04],\n",
       "        [3.0000e+00, 6.4633e-01, 1.3925e-01, 3.0000e-02, 6.4633e-03, 1.3925e-03],\n",
       "        [4.0000e+00, 8.6177e-01, 1.8566e-01, 4.0000e-02, 8.6177e-03, 1.8566e-03],\n",
       "        [5.0000e+00, 1.0772e+00, 2.3208e-01, 5.0000e-02, 1.0772e-02, 2.3208e-03],\n",
       "        [6.0000e+00, 1.2927e+00, 2.7850e-01, 6.0000e-02, 1.2927e-02, 2.7850e-03],\n",
       "        [7.0000e+00, 1.5081e+00, 3.2491e-01, 7.0000e-02, 1.5081e-02, 3.2491e-03],\n",
       "        [8.0000e+00, 1.7235e+00, 3.7133e-01, 8.0000e-02, 1.7235e-02, 3.7133e-03],\n",
       "        [9.0000e+00, 1.9390e+00, 4.1774e-01, 9.0000e-02, 1.9390e-02, 4.1774e-03],\n",
       "        [1.0000e+01, 2.1544e+00, 4.6416e-01, 1.0000e-01, 2.1544e-02, 4.6416e-03],\n",
       "        [1.1000e+01, 2.3699e+00, 5.1057e-01, 1.1000e-01, 2.3699e-02, 5.1057e-03],\n",
       "        [1.2000e+01, 2.5853e+00, 5.5699e-01, 1.2000e-01, 2.5853e-02, 5.5699e-03],\n",
       "        [1.3000e+01, 2.8008e+00, 6.0341e-01, 1.3000e-01, 2.8008e-02, 6.0341e-03],\n",
       "        [1.4000e+01, 3.0162e+00, 6.4982e-01, 1.4000e-01, 3.0162e-02, 6.4982e-03],\n",
       "        [1.5000e+01, 3.2317e+00, 6.9624e-01, 1.5000e-01, 3.2317e-02, 6.9624e-03],\n",
       "        [1.6000e+01, 3.4471e+00, 7.4265e-01, 1.6000e-01, 3.4471e-02, 7.4265e-03],\n",
       "        [1.7000e+01, 3.6625e+00, 7.8907e-01, 1.7000e-01, 3.6625e-02, 7.8907e-03],\n",
       "        [1.8000e+01, 3.8780e+00, 8.3549e-01, 1.8000e-01, 3.8780e-02, 8.3549e-03],\n",
       "        [1.9000e+01, 4.0934e+00, 8.8190e-01, 1.9000e-01, 4.0934e-02, 8.8190e-03]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos.size(), divterm.size()\n",
    "print((pos[:, None] / divterm).size())\n",
    "pos[:, None] / divterm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f56dc1858e0>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pe[:, ::2] = torch.sin(pos[:, None] / divterm)\n",
    "sns.heatmap(pe, cmap='RdBu', center=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.0000, 0.0000],\n",
       "        [0.0005, 0.0000],\n",
       "        [0.0009, 0.0000],\n",
       "        [0.0014, 0.0000],\n",
       "        [0.0019, 0.0000],\n",
       "        [0.0023, 0.0000],\n",
       "        [0.0028, 0.0000],\n",
       "        [0.0032, 0.0000],\n",
       "        [0.0037, 0.0000],\n",
       "        [0.0042, 0.0000],\n",
       "        [0.0046, 0.0000],\n",
       "        [0.0051, 0.0000],\n",
       "        [0.0056, 0.0000],\n",
       "        [0.0060, 0.0000],\n",
       "        [0.0065, 0.0000],\n",
       "        [0.0070, 0.0000],\n",
       "        [0.0074, 0.0000],\n",
       "        [0.0079, 0.0000],\n",
       "        [0.0084, 0.0000],\n",
       "        [0.0088, 0.0000]])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pe[:, 10:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f56dc05f340>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pe[:, 1::2] = torch.cos(pos[:, None] / divterm)\n",
    "sns.heatmap(pe, cmap='RdBu', center=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Combining it in a `nn.Module`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# took inspiration from here: https://pytorch.org/tutorials/beginner/transformer_tutorial.html\n",
    "class PositionalEncoding(nn.Module):\n",
    "\n",
    "    def __init__(self, d_model, dropout=0.1, maxlen=50):\n",
    "        super(PositionalEncoding, self).__init__()\n",
    "        self.dropout = nn.Dropout(p=dropout)\n",
    "        \n",
    "        pe = torch.zeros(maxlen, d_model)\n",
    "        pos = torch.arange(maxlen, dtype=torch.float)\n",
    "        divterm = 10000 ** (torch.arange(0, d_model, step=2) / float(d_model))\n",
    "        pe[:, ::2] = torch.sin(pos[:, None] / divterm)\n",
    "        pe[:, 1::2] = torch.cos(pos[:, None] / divterm)\n",
    "        \n",
    "        # Since pe is a constant value not a parameter of the module, we register it as a buffer.\n",
    "        # Buffers are part of the state dictionary of the module along with parameters.\n",
    "        # Docs: https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.register_buffer\n",
    "        self.register_buffer('pe', pe)\n",
    "\n",
    "    def forward(self, x):\n",
    "        # The input sequence may be shorter than maxlen\n",
    "        seqlen = x.size(0)\n",
    "        \n",
    "        # The middle dimension is the batch size.\n",
    "        # We add it as a dummy dimension.\n",
    "        x = x + self.pe[:seqlen, None, :]\n",
    "        return self.dropout(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([10, 7, 12])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d_model = 12\n",
    "maxlen = 20\n",
    "batch_size = 7\n",
    "seqlen = 10\n",
    "pos_enc = PositionalEncoding(d_model=d_model, dropout=0., maxlen=maxlen)\n",
    "x = torch.rand(size=(seqlen, batch_size, d_model))\n",
    "x_pe = pos_enc(x)\n",
    "x_pe.size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Contextual embeddings\n",
    "\n",
    "In GloVe and Word2vec representations, words have static representations, in other words, the same vector is assigned for every occurrence of the word.\n",
    "But words can have different meaning in different contexts, e.g. the word 'stick':\n",
    "\n",
    "1. Find some dry sticks and we'll make a campfire.\n",
    "2. Let's stick with glove embeddings.\n",
    "\n",
    "![elmo](http://jalammar.github.io/images/elmo-embedding-robin-williams.png)\n",
    "\n",
    "_(Peters et. al., 2018 in the ELMo paper)_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ELMo\n",
    "\n",
    "**E**mbeddings from **L**anguage **Mo**dels\n",
    "\n",
    "Word representations are functions of the full sentences instead of the word alone.\n",
    "\n",
    "Two bidirectional LSTM layers are linearly combined.\n",
    "\n",
    "[Deep contextualized word representations](https://arxiv.org/abs/1802.05365) by Peters et al., 2018, 6300 citations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# BERT\n",
    "\n",
    "[BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://www.aclweb.org/anthology/N19-1423/)\n",
    "by Devlin et al. 2018, 17500 citations\n",
    "\n",
    "[BERTology](https://huggingface.co/transformers/bertology.html) is the nickname for the growing amount of BERT-related research.\n",
    "\n",
    "Trained on two tasks:\n",
    "\n",
    "1. Masked language model:\n",
    "\n",
    "    1. 15% of the <s>tokens</s>wordpieces are selected at the beginning.\n",
    "    2. 80% of those are replaced with `[MASK]`,\n",
    "    3. 10% are replaced with a random token,\n",
    "    4. 10% are kept intact.\n",
    "    \n",
    "2. Next sentence prediction:\n",
    "    - Are sentences A and B consecutive sentences?\n",
    "    - Generate 50-50%.\n",
    "    - Binary classification task."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Embedding layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"img/dl/bert_embedding.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transformer layers\n",
    "\n",
    "\n",
    "## Finetuning\n",
    "\n",
    "1. Take a trained BERT model.\n",
    "2. Add a small classification layer on top (typically a 2-layer MLP).\n",
    "3. Train BERT along with the classification layer on an annotated dataset.\n",
    "    - Much smaller than the data BERT was trained on\n",
    "\n",
    "Another option: freeze BERT and train the classification layer only.\n",
    "- Easier training regime.\n",
    "- Smaller memory footprint.\n",
    "- Worse performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(\"img/dl/bert_encoding_finetuning.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BERT pretrained checkpoints\n",
    "\n",
    "### BERT-Base\n",
    "\n",
    "- 12 layers\n",
    "- 12 attention heads per layer\n",
    "- 768 hidden size\n",
    "- 110M parameters\n",
    "\n",
    "### BERT-Large\n",
    "\n",
    "- 24 layers\n",
    "- 16 attention heads per layer\n",
    "- 1024 hidden size\n",
    "- 340M parameters\n",
    "\n",
    "### Cased and uncased\n",
    "\n",
    "Uncased: everything is lowercased. Diacritics are removed.\n",
    "\n",
    "### Multilingual BERT - mBERT\n",
    "\n",
    "104 language version trained on the 100 largest Wikipedia."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BERT implementations\n",
    "\n",
    "[Original Tensorflow implementation](https://github.com/google-research/bert)\n",
    "\n",
    "[Huggingface Transformers](https://huggingface.co/transformers/)\n",
    "- PyTorch implementation originally for BERT-only\n",
    "- Now it supports dozens of other models\n",
    "- Hundreds of other model checkpoints from the community"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# BERT tokenization\n",
    "\n",
    "## WordPiece tokenizer\n",
    "\n",
    "BERT's input **must** be tokenized with BERT's own tokenizer.\n",
    "\n",
    "A middle ground between word and character tokenization.\n",
    "\n",
    "Static vocabulary:\n",
    "- Byte-pair encoding: simple frequency-based tokenization method\n",
    "- Continuation symbols (\\#\\#symbol)\n",
    "- Special tokens: `[CLS]`, `[SEP]`, `[MASK]`, `[UNK]`\n",
    "- It tokenizes everything, falling back to characters and `[UNK]` if necessary\n",
    "\n",
    "`AutoTokenizer` is a factory class for pretrained tokenizers. ng id. `from_pretrained` instantiates the corresponding class and loads the weights:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'transformers.tokenization_bert.BertTokenizer'>\n",
      "30522\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['my', 'bea', '##gle', \"'\", 's', 'name', 'is', 'tun', '##der', '##ke', '.']"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = AutoTokenizer.from_pretrained('bert-base-uncased')\n",
    "print(type(t))\n",
    "print(len(t.get_vocab()))\n",
    "\n",
    "t.tokenize(\"My beagle's name is Tündérke.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['р', '##у', '##с', '##с', '##к', '##ии']"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"Русский\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Cased** models keep diacritics:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28996\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['My',\n",
       " 'be',\n",
       " '##ag',\n",
       " '##le',\n",
       " \"'\",\n",
       " 's',\n",
       " 'name',\n",
       " 'is',\n",
       " 'T',\n",
       " '##ü',\n",
       " '##nd',\n",
       " '##é',\n",
       " '##rk',\n",
       " '##e',\n",
       " '.']"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = AutoTokenizer.from_pretrained('bert-base-cased')\n",
    "print(len(t.get_vocab()))\n",
    "\n",
    "t.tokenize(\"My beagle's name is Tündérke.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It character tokenizes Chinese and Japanese but doesn't know all the characters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['日', '本', '[UNK]']"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"日本語\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Korean is missing from this version:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['한', '[UNK]']"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"한 한국어\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## mBERT tokenization\n",
    "\n",
    "104 languages, 1 vocabulary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "t = AutoTokenizer.from_pretrained('bert-base-multilingual-cased')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "119547"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(t.get_vocab())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['My',\n",
       " 'pu',\n",
       " '##ppy',\n",
       " \"'\",\n",
       " 's',\n",
       " 'name',\n",
       " 'is',\n",
       " 'T',\n",
       " '##ün',\n",
       " '##dé',\n",
       " '##rke',\n",
       " '.']"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"My puppy's name is Tündérke.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['한', '한국', '##어']"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"한 한국어\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['日', '本', '語']"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.tokenize(\"日本語\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using BERT\n",
    "\n",
    "## Using `BertModel` directly\n",
    "\n",
    "`AutoModel`\n",
    "- each pretrained checkpoint has a string id. `from_pretrained` instantiates the corresponding class and loads the weights:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(transformers.modeling_bert.BertModel,\n",
       " transformers.tokenization_bert.BertTokenizer)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer = AutoTokenizer.from_pretrained('bert-base-cased')\n",
    "model = AutoModel.from_pretrained('bert-base-cased')\n",
    "type(model), type(tokenizer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['There', 'are', 'black', 'cats', 'and', 'black', 'dogs', '.']"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer.tokenize(\"There are black cats and black dogs.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`__call__` return a dictionary of BERT's encoding:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_ids': [101, 1247, 1132, 1602, 11771, 1105, 1602, 6363, 119, 102], 'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer(\"There are black cats and black dogs.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be used for pairs of sentences. Note the values of `token_type_ids`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_ids': [101, 1247, 1132, 1602, 11771, 1105, 1602, 6363, 119, 102, 2543, 5650, 119, 102], 'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer(\"There are black cats and black dogs.\", \"Another sentence.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be used for multiple sentences:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_ids': [[101, 1247, 1132, 1602, 11771, 1105, 1602, 6363, 119, 102], [101, 1247, 1132, 1160, 1653, 11771, 119, 102]], 'token_type_ids': [[0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0, 0, 0]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1]]}"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer([\"There are black cats and black dogs.\", \"There are two white cats.\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1],\n",
       "        [1, 1, 1, 1, 1, 1, 1, 1, 0, 0]])"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded = tokenizer([\"There are black cats and black dogs.\", \"There are two white cats.\"], return_tensors='pt', padding=True)\n",
    "encoded['attention_mask']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need tensors as inputs for BERT:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 10])"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded = tokenizer(\"There are black cats and black dogs.\", return_tensors='pt')\n",
    "encoded['input_ids'].size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_ids': tensor([[  101,  1247,  1132,  1602, 11771,  1105,  1602,  6363,   119,   102]]), 'token_type_ids': tensor([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])}"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "odict_keys(['last_hidden_state', 'pooler_output'])"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output = model(**encoded, return_dict=True)\n",
    "output.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([1, 10, 768]), torch.Size([1, 768]))"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output['last_hidden_state'].size(), output['pooler_output'].size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting all layers:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "odict_keys(['last_hidden_state', 'pooler_output', 'hidden_states'])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output = model(**encoded, output_hidden_states=True, return_dict=True)\n",
    "output.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13, torch.Size([1, 10, 768]))"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(output['hidden_states']), output['hidden_states'][0].size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Remove variable from the global namespace, run the garbage collector:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "398"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del model\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BERT applications\n",
    "\n",
    "### Sequence classification\n",
    "\n",
    "Pretrained model for sentiment analysis.\n",
    "\n",
    "Base model: `distilbert-base-uncased`\n",
    "\n",
    "Finetuned on the [Stanford Sentiment Treebank](https://nlp.stanford.edu/sentiment/index.html) or SST-2, a popular sentiment analysis dataset.\n",
    "\n",
    "Model id: `distilbert-base-uncased-finetuned-sst-2-english`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'label': 'POSITIVE', 'score': 0.9998814463615417}]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nlp = pipeline(\"sentiment-analysis\")\n",
    "nlp(\"This is an amazing class.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'label': 'POSITIVE', 'score': 0.9302219152450562}]"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nlp(\"This is not a good class but it's not too bad either.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'label': 'NEGATIVE', 'score': 0.9993655681610107}]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nlp(\"This is not a class.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "287"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del nlp\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sequence tagging/labeling: Named entity recognition\n",
    "\n",
    "Base model: `bert-large-cased`\n",
    "\n",
    "Finetuned on [CoNLL-2003 NER](https://www.clips.uantwerpen.be/conll2003/ner/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = pipeline(\"ner\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'word': 'Jupiter',\n",
       "  'score': 0.917730450630188,\n",
       "  'entity': 'I-LOC',\n",
       "  'index': 1},\n",
       " {'word': 'James', 'score': 0.5041977167129517, 'entity': 'I-LOC', 'index': 8},\n",
       " {'word': 'Universe',\n",
       "  'score': 0.5249300599098206,\n",
       "  'entity': 'I-LOC',\n",
       "  'index': 13}]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = nlp(\"Jupiter is a planet that orbits around James the center of the Universe\")\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'word': 'George',\n",
       "  'score': 0.9989229440689087,\n",
       "  'entity': 'I-PER',\n",
       "  'index': 1},\n",
       " {'word': 'C', 'score': 0.9973233938217163, 'entity': 'I-PER', 'index': 2},\n",
       " {'word': '##lo', 'score': 0.9540075063705444, 'entity': 'I-PER', 'index': 3},\n",
       " {'word': '##oney',\n",
       "  'score': 0.9897923469543457,\n",
       "  'entity': 'I-PER',\n",
       "  'index': 4},\n",
       " {'word': 'E', 'score': 0.9577357769012451, 'entity': 'I-PER', 'index': 10},\n",
       " {'word': '##ste',\n",
       "  'score': 0.9451481103897095,\n",
       "  'entity': 'I-PER',\n",
       "  'index': 11},\n",
       " {'word': '##lla', 'score': 0.865635097026825, 'entity': 'I-PER', 'index': 12}]"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = nlp(\"George Clooney has a pet pig named Estella.\")\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1706"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del nlp\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Machine translation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of T5Model were not initialized from the model checkpoint at t5-base and are newly initialized: ['encoder.embed_tokens.weight', 'decoder.embed_tokens.weight']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'translation_text': 'Hugging Face est une entreprise technologique basée à New York et à Paris.'}]\n"
     ]
    }
   ],
   "source": [
    "nlp = pipeline(\"translation_en_to_fr\")\n",
    "print(nlp(\"Hugging Face is a technology company based in New York and Paris\", max_length=40))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Even the [blessé - blessed false cognate](https://frenchtogether.com/french-english-false-friends/) is handled correctly:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'translation_text': \"J'ai été béni par Dieu après avoir blessé ma tête.\"}]"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nlp(\"I was blessed by God after I injured my head.\", max_length=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1082"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del nlp\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Masked language modeling\n",
    "\n",
    "Uses `distilroberta-base`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of RobertaForMaskedLM were not initialized from the model checkpoint at distilroberta-base and are newly initialized: ['lm_head.decoder.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "nlp = pipeline(\"fill-mask\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Twitter is a bad idea /s> [MASK] /s> /s> /s> /s> /s> /s> END END END END'"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = \"Twitter is a bad idea /s> [MASK]\"\n",
    "\n",
    "for n in range(10):\n",
    "    result = nlp(f\"{prompt} {nlp.tokenizer.mask_token}\")\n",
    "    token = result[0]['token_str'][1:]\n",
    "    prompt += \" \" + token\n",
    "    \n",
    "prompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'score': 0.17927514016628265,\n",
      "  'sequence': '<s>HuggingFace is creating a tool that the community uses to '\n",
      "              'solve NLP tasks.</s>',\n",
      "  'token': 3944,\n",
      "  'token_str': 'Ġtool'},\n",
      " {'score': 0.11349447071552277,\n",
      "  'sequence': '<s>HuggingFace is creating a framework that the community uses '\n",
      "              'to solve NLP tasks.</s>',\n",
      "  'token': 7208,\n",
      "  'token_str': 'Ġframework'},\n",
      " {'score': 0.052435461431741714,\n",
      "  'sequence': '<s>HuggingFace is creating a library that the community uses to '\n",
      "              'solve NLP tasks.</s>',\n",
      "  'token': 5560,\n",
      "  'token_str': 'Ġlibrary'},\n",
      " {'score': 0.03493537753820419,\n",
      "  'sequence': '<s>HuggingFace is creating a database that the community uses '\n",
      "              'to solve NLP tasks.</s>',\n",
      "  'token': 8503,\n",
      "  'token_str': 'Ġdatabase'},\n",
      " {'score': 0.028602318838238716,\n",
      "  'sequence': '<s>HuggingFace is creating a prototype that the community uses '\n",
      "              'to solve NLP tasks.</s>',\n",
      "  'token': 17715,\n",
      "  'token_str': 'Ġprototype'}]\n"
     ]
    }
   ],
   "source": [
    "from pprint import pprint\n",
    "pred = nlp(f\"HuggingFace is creating a {nlp.tokenizer.mask_token} that the community uses to solve NLP tasks.\")\n",
    "pprint(pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'score': 0.03921409323811531,\n",
      "  'sequence': '<s>This is a very good idea.</s>',\n",
      "  'token': 713,\n",
      "  'token_str': 'This'},\n",
      " {'score': 0.02041138894855976,\n",
      "  'sequence': '<s>That is a very good idea.</s>',\n",
      "  'token': 1711,\n",
      "  'token_str': 'That'},\n",
      " {'score': 0.018834417685866356,\n",
      "  'sequence': '<s>It is a very good idea.</s>',\n",
      "  'token': 243,\n",
      "  'token_str': 'It'},\n",
      " {'score': 0.010761899873614311,\n",
      "  'sequence': '<s>Which is a very good idea.</s>',\n",
      "  'token': 32251,\n",
      "  'token_str': 'Which'},\n",
      " {'score': 0.006242118775844574,\n",
      "  'sequence': '<s>Bitcoin is a very good idea.</s>',\n",
      "  'token': 42817,\n",
      "  'token_str': 'Bitcoin'}]\n"
     ]
    }
   ],
   "source": [
    "pred = nlp(f\"{nlp.tokenizer.mask_token} is a very good idea.\")\n",
    "pprint(pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'score': 0.03035598248243332,\n",
      "  'sequence': '<s>This is a bad idea.</s>',\n",
      "  'token': 713,\n",
      "  'token_str': 'This'},\n",
      " {'score': 0.017567547038197517,\n",
      "  'sequence': '<s>That is a bad idea.</s>',\n",
      "  'token': 1711,\n",
      "  'token_str': 'That'},\n",
      " {'score': 0.011682895012199879,\n",
      "  'sequence': '<s>It is a bad idea.</s>',\n",
      "  'token': 243,\n",
      "  'token_str': 'It'},\n",
      " {'score': 0.010721716098487377,\n",
      "  'sequence': '<s>Which is a bad idea.</s>',\n",
      "  'token': 32251,\n",
      "  'token_str': 'Which'},\n",
      " {'score': 0.008407930843532085,\n",
      "  'sequence': '<s>Twitter is a bad idea.</s>',\n",
      "  'token': 22838,\n",
      "  'token_str': 'Twitter'}]\n"
     ]
    }
   ],
   "source": [
    "pred = nlp(f\"{nlp.tokenizer.mask_token} is a bad idea.\")\n",
    "pprint(pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "337"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del nlp\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Other models\n",
    "\n",
    "## Pretrained models\n",
    "\n",
    "RoBERTa: identical model, larger training data, different training objective\n",
    "\n",
    "DistilBERT: smaller version of BERT. It was _distilled_ or compressed from BERT with a student-teacher setup.\n",
    "\n",
    "ALBERT: smaller BERT\n",
    "\n",
    "XLM-RoBERTa: multilingual version of RoBERTa\n",
    "\n",
    "Distil-mBERT: distilled multilingual BERT"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Community models\n",
    "\n",
    "[Over 1000 community contributions](https://huggingface.co/models)\n",
    "\n",
    "## huBERT\n",
    "\n",
    "The first Hungarian-only model and the only one registered on Huggingface.\n",
    "Other models are available at https://hilanco.github.io/.\n",
    "\n",
    "BERT base, trained on Webcorpus 2.0, a version of CommonCrawl.\n",
    "\n",
    "Its tokenizer works much better for Hungarian than mBERT's:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "hubert_tokenizer = AutoTokenizer.from_pretrained('SZTAKI-HLT/hubert-base-cc')\n",
    "# hubert = AutoModel.from_pretrained('SZTAKI-HLT/hubert-base-cc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['George',\n",
       " 'Cl',\n",
       " '##oo',\n",
       " '##ney',\n",
       " 'Magyarországról',\n",
       " 'szóló',\n",
       " ',',\n",
       " 'az',\n",
       " 'Orbán',\n",
       " '-',\n",
       " 'kormányt',\n",
       " 'krit',\n",
       " '##izáló',\n",
       " 'level',\n",
       " '##ére',\n",
       " 'miniszteri',\n",
       " 'és',\n",
       " 'államtitkár',\n",
       " '##i',\n",
       " 'szinten',\n",
       " 'is',\n",
       " 'reagált',\n",
       " 'a',\n",
       " 'magyar',\n",
       " 'kormány',\n",
       " '.']"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent = (\"George Clooney Magyarországról szóló, az Orbán-kormányt kritizáló levelére miniszteri és \"\n",
    "        \"államtitkári szinten is reagált a magyar kormány.\")\n",
    "hubert_tokenizer.tokenize(sent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "bert_tokenizer = AutoTokenizer.from_pretrained('bert-base-multilingual-cased')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['George',\n",
       " 'Cl',\n",
       " '##oon',\n",
       " '##ey',\n",
       " 'Magyarország',\n",
       " '##ról',\n",
       " 'szóló',\n",
       " ',',\n",
       " 'az',\n",
       " 'Or',\n",
       " '##bán',\n",
       " '-',\n",
       " 'kormány',\n",
       " '##t',\n",
       " 'kr',\n",
       " '##iti',\n",
       " '##zá',\n",
       " '##ló',\n",
       " 'level',\n",
       " '##ére',\n",
       " 'mini',\n",
       " '##szter',\n",
       " '##i',\n",
       " 'és',\n",
       " 'állam',\n",
       " '##tit',\n",
       " '##kár',\n",
       " '##i',\n",
       " 'szinte',\n",
       " '##n',\n",
       " 'is',\n",
       " 're',\n",
       " '##ag',\n",
       " '##ált',\n",
       " 'a',\n",
       " 'magyar',\n",
       " 'kormány',\n",
       " '.']"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bert_tokenizer.tokenize(sent)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GPT-2 text generation\n",
    "\n",
    "Causal language modeliing is when the $i^{th}$ token is modeled based on all the previous tokens as opposed to masked language modeling where both left and right context are used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of GPT2Model were not initialized from the model checkpoint at gpt2 and are newly initialized: ['h.0.attn.masked_bias', 'h.1.attn.masked_bias', 'h.2.attn.masked_bias', 'h.3.attn.masked_bias', 'h.4.attn.masked_bias', 'h.5.attn.masked_bias', 'h.6.attn.masked_bias', 'h.7.attn.masked_bias', 'h.8.attn.masked_bias', 'h.9.attn.masked_bias', 'h.10.attn.masked_bias', 'h.11.attn.masked_bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "text_generator = pipeline(\"text-generation\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is a serious issue we should address,\" he said.\n",
      "\n",
      "\"We need to be able to do that. We need to be able to do that in a way that is fair and equitable.\"\n",
      "\n",
      "Topics: government-and-politics\n"
     ]
    }
   ],
   "source": [
    "print(text_generator(\"This is a serious issue we should address\", max_length=50, do_sample=False)[0]['generated_text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Twitter is a bad idea, Jack Dorsey had a bad day when he came up with it.\n",
      "\n",
      "The idea of a \"social media\" is a bad idea.\n",
      "\n",
      "The idea of a \"social media\" is a bad idea.\n",
      "\n",
      "The idea of a \"social media\" is a bad idea.\n",
      "\n",
      "The idea of a \"social media\" is a bad idea.\n",
      "\n",
      "The idea of a \"social media\" is a bad idea.\n",
      "\n",
      "The idea of\n"
     ]
    }
   ],
   "source": [
    "print(text_generator(\"Twitter is a bad idea, Jack Dorsey had a bad day when he came up with it\", max_length=100, do_sample=False)[0]['generated_text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "742"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "del text_generator\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Further information\n",
    "\n",
    "[Official PyTorch Transformer tutorial](https://pytorch.org/tutorials/beginner/transformer_tutorial.html)\n",
    "\n",
    "[Illustrated Transformer](http://jalammar.github.io/illustrated-transformer/)\n",
    "- Famous blog post with a detailed gentle introduction to Transformers\n",
    "\n",
    "[The Annotated Transformer](http://nlp.seas.harvard.edu/2018/04/03/attention.html)\n",
    "- A walkthrough of original Transformer paper with code and detailed illustration\n",
    "\n",
    "[Huggingface Transformers - Summary of tasks](https://huggingface.co/transformers/task_summary.html)\n",
    "\n",
    "[My blog post about mBERT's tokenizer](http://juditacs.github.io/2019/02/19/bert-tokenization-stats.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}