notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# distributed-gpt2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start by loading the GPT-2 model re-implemented by Hugging Face in https://github.com/huggingface/transformers/blob/main/src/transformers/models/gpt2/modeling_gpt2.py."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import GPT2LMHeadModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`wte` stands for Weight Token Embedding. There are 50257 tokens in the GPT-2 vocabulary, and 768-dimensional embeddings are used for the lookup table.\n",
    "\n",
    "`wpe` stands for Weight Positional Embedding. The context has a length of 1024."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "transformer.wte.weight torch.Size([50257, 768])\n",
      "transformer.wpe.weight torch.Size([1024, 768])\n",
      "transformer.h.0.ln_1.weight torch.Size([768])\n",
      "transformer.h.0.ln_1.bias torch.Size([768])\n",
      "transformer.h.0.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.0.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.0.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.0.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.0.ln_2.weight torch.Size([768])\n",
      "transformer.h.0.ln_2.bias torch.Size([768])\n",
      "transformer.h.0.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.0.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.0.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.0.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.1.ln_1.weight torch.Size([768])\n",
      "transformer.h.1.ln_1.bias torch.Size([768])\n",
      "transformer.h.1.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.1.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.1.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.1.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.1.ln_2.weight torch.Size([768])\n",
      "transformer.h.1.ln_2.bias torch.Size([768])\n",
      "transformer.h.1.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.1.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.1.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.1.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.2.ln_1.weight torch.Size([768])\n",
      "transformer.h.2.ln_1.bias torch.Size([768])\n",
      "transformer.h.2.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.2.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.2.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.2.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.2.ln_2.weight torch.Size([768])\n",
      "transformer.h.2.ln_2.bias torch.Size([768])\n",
      "transformer.h.2.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.2.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.2.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.2.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.3.ln_1.weight torch.Size([768])\n",
      "transformer.h.3.ln_1.bias torch.Size([768])\n",
      "transformer.h.3.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.3.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.3.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.3.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.3.ln_2.weight torch.Size([768])\n",
      "transformer.h.3.ln_2.bias torch.Size([768])\n",
      "transformer.h.3.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.3.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.3.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.3.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.4.ln_1.weight torch.Size([768])\n",
      "transformer.h.4.ln_1.bias torch.Size([768])\n",
      "transformer.h.4.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.4.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.4.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.4.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.4.ln_2.weight torch.Size([768])\n",
      "transformer.h.4.ln_2.bias torch.Size([768])\n",
      "transformer.h.4.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.4.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.4.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.4.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.5.ln_1.weight torch.Size([768])\n",
      "transformer.h.5.ln_1.bias torch.Size([768])\n",
      "transformer.h.5.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.5.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.5.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.5.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.5.ln_2.weight torch.Size([768])\n",
      "transformer.h.5.ln_2.bias torch.Size([768])\n",
      "transformer.h.5.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.5.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.5.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.5.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.6.ln_1.weight torch.Size([768])\n",
      "transformer.h.6.ln_1.bias torch.Size([768])\n",
      "transformer.h.6.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.6.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.6.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.6.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.6.ln_2.weight torch.Size([768])\n",
      "transformer.h.6.ln_2.bias torch.Size([768])\n",
      "transformer.h.6.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.6.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.6.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.6.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.7.ln_1.weight torch.Size([768])\n",
      "transformer.h.7.ln_1.bias torch.Size([768])\n",
      "transformer.h.7.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.7.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.7.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.7.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.7.ln_2.weight torch.Size([768])\n",
      "transformer.h.7.ln_2.bias torch.Size([768])\n",
      "transformer.h.7.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.7.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.7.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.7.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.8.ln_1.weight torch.Size([768])\n",
      "transformer.h.8.ln_1.bias torch.Size([768])\n",
      "transformer.h.8.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.8.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.8.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.8.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.8.ln_2.weight torch.Size([768])\n",
      "transformer.h.8.ln_2.bias torch.Size([768])\n",
      "transformer.h.8.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.8.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.8.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.8.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.9.ln_1.weight torch.Size([768])\n",
      "transformer.h.9.ln_1.bias torch.Size([768])\n",
      "transformer.h.9.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.9.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.9.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.9.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.9.ln_2.weight torch.Size([768])\n",
      "transformer.h.9.ln_2.bias torch.Size([768])\n",
      "transformer.h.9.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.9.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.9.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.9.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.10.ln_1.weight torch.Size([768])\n",
      "transformer.h.10.ln_1.bias torch.Size([768])\n",
      "transformer.h.10.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.10.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.10.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.10.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.10.ln_2.weight torch.Size([768])\n",
      "transformer.h.10.ln_2.bias torch.Size([768])\n",
      "transformer.h.10.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.10.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.10.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.10.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.h.11.ln_1.weight torch.Size([768])\n",
      "transformer.h.11.ln_1.bias torch.Size([768])\n",
      "transformer.h.11.attn.c_attn.weight torch.Size([768, 2304])\n",
      "transformer.h.11.attn.c_attn.bias torch.Size([2304])\n",
      "transformer.h.11.attn.c_proj.weight torch.Size([768, 768])\n",
      "transformer.h.11.attn.c_proj.bias torch.Size([768])\n",
      "transformer.h.11.ln_2.weight torch.Size([768])\n",
      "transformer.h.11.ln_2.bias torch.Size([768])\n",
      "transformer.h.11.mlp.c_fc.weight torch.Size([768, 3072])\n",
      "transformer.h.11.mlp.c_fc.bias torch.Size([3072])\n",
      "transformer.h.11.mlp.c_proj.weight torch.Size([3072, 768])\n",
      "transformer.h.11.mlp.c_proj.bias torch.Size([768])\n",
      "transformer.ln_f.weight torch.Size([768])\n",
      "transformer.ln_f.bias torch.Size([768])\n",
      "lm_head.weight torch.Size([50257, 768])\n"
     ]
    }
   ],
   "source": [
    "model_hf = GPT2LMHeadModel.from_pretrained(\"gpt2\") # 124M parameters\n",
    "model_hf_state = model_hf.state_dict()\n",
    "\n",
    "for k, v in model_hf_state.items():\n",
    "    print(k, v.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([-0.1101, -0.0393,  0.0331,  0.1338, -0.0485, -0.0789, -0.2398, -0.0895,\n",
       "         0.0253, -0.1074, -0.1811, -0.0672,  0.0739, -0.0161,  0.0117,  0.1245,\n",
       "        -0.0020, -0.0815,  0.0338,  0.2365])"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_hf_state[\"transformer.wte.weight\"].view(-1)[:20]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With 768 dimensions for positional embeddings, tokens in the 1024-sized context are encoded using these dimensions.\n",
    "\n",
    "In the following, each row depicts an absolute position. A given token (columns) encodes some structure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7ffed99ceed0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.imshow(model_hf_state[\"transformer.wpe.weight\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's have a look at three random positions on the context to observe how channels react to token positions:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7ffed982ec90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9913c80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed992f110>,\n",
       " <matplotlib.lines.Line2D at 0x7ffee6137f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffee6137f50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffef0404590>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed986cfe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dd910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dda00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ddaf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ddbe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ddcd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dddc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ddeb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ddfa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed987c320>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98de9c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98deab0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98deba0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dec90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98ded80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dee70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98def60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df050>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df140>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df230>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df320>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df410>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df500>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df5f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df6e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98df7d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed986d730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98604a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfa10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfb00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfbf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfdd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dfec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed98dffb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97140e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97141d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97142c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97143b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97144a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714590>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714680>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714770>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714860>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714950>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714a40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714b30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714c20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714d10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714e00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714ef0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9714fe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97f9730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97150d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715280>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715370>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715460>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715550>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715640>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715820>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715a00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715af0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715be0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715cd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715dc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715eb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9715fa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716900>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97169f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716ae0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716bd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716cc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716db0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716ea0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9716f90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97178f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97179e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9717f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97300b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97301a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730a10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730b00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730bf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730ce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730dd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730ec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9730fb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97310a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731190>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731280>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731370>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731460>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731550>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731640>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731820>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731a00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731af0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731be0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731cd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731dc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731eb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9731fa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732900>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97329f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732ae0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732bd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732cc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732db0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732ea0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9732f90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97338f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97339e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9733f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97500b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97501a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750a10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750b00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9bdce30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750bf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9750f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751070>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751160>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751250>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751340>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751430>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751520>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751610>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751700>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97517f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97518e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97519d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751ac0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751bb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751ca0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751d90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751e80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9751f70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752060>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752150>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752240>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752330>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752420>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752510>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752600>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97526f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97527e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97528d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97529c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752ab0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752ba0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752c90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752d80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752e70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9752f60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753050>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753140>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753230>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753320>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753410>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753500>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97535f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97536e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97537d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97538c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97539b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753aa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753b90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753c80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753d70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753e60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9753f50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c8f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976c9e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976cad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976cbc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ccb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976cda0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ce90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976cf80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d070>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d160>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d250>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d340>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d430>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d520>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d610>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d700>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d7f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d8e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976d9d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976dac0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed987d3a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976dbb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976dd60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976de50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976df40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e030>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e120>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e210>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e300>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e3f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e4e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e5d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e6c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed989c170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e7b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976e960>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ea50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976eb40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ec30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ed20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ee10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ef00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976eff0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f0e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f1d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f2c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f3b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f4a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f590>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f680>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f770>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f860>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976f950>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976fa40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976fb30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed99b02f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976fc20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976fdd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976fec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed976ffb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c0e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c1d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c2c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c3b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c4a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c590>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c680>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c770>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c860>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978c950>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ca40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978cb30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978cc20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978cd10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ce00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978cef0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978cfe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d0d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d1c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d2b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d3a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d490>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d580>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d670>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d760>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d850>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978d940>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978da30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978db20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978dc10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978dd00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ddf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978dee0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978dfd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e0c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e1b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e2a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e390>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e480>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e570>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e660>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e750>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e840>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978e930>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ea20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978eb10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ec00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ecf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ede0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978eed0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978efc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f0b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f1a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978f920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fa10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fb00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fbf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fdd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978fec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed978ffb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a40e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a41d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a42c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a43b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a44a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4590>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4680>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4770>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4860>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4950>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4a40>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4b30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4c20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4d10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4e00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4ef0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a4fe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a50d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a51c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a52b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a53a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5490>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5580>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5670>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5760>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5850>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5940>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5a30>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5b20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5c10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5d00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5df0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5ee0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a5fd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a60c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a61b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a62a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6390>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6480>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6570>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6660>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6750>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6840>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6930>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6a20>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6b10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6c00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6cf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6de0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6ed0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a6fc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a70b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a71a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9bdccb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a78f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a79e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97a7f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc0b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc1a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bc920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bca10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcb00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcbf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcdd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bcfb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd0a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd190>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd280>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd370>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd460>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd550>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd640>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd820>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bd910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bda00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bdaf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bdbe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bdcd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bddc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bdeb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bdfa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be900>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97be9f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97beae0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bebd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97becc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bedb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97beea0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bef90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf8f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bf9e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bfad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bfbc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bfcb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bfda0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bfe90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97bff80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e00b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e01a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0a10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0b00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0bf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0ce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0dd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0ec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e0fb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e10a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1190>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1280>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1370>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1460>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1550>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1640>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1820>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1a00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1af0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1be0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1cd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1dc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1eb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e1fa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2900>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e29f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2ae0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2bd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2cc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2db0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2ea0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e2f90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e38f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e39e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed97e3f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc0b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc1a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc290>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc380>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc470>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc560>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc650>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc740>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc830>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fc920>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fca10>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcb00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcbf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcce0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcdd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcec0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fcfb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd0a0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd190>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd280>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd370>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd460>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd550>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd640>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd730>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd820>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fd910>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fda00>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fdaf0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fdbe0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fdcd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fddc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fdeb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fdfa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe090>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe180>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe270>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe360>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe450>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe540>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe630>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe720>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe810>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe900>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fe9f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95feae0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95febd0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fecc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fedb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95feea0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fef90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed987c770>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ff8f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ffaa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ffb90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ffc80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ffd70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95ffe60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed95fff50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96188f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96189e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9618f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619070>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619160>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619250>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619340>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619430>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619520>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619610>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619700>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96197f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96198e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96199d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619ac0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619bb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619ca0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619d90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619e80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9619f70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a060>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a150>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a240>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a330>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a420>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a510>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a600>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a6f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a7e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a8d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961a9c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961aab0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961aba0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961ac90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961ad80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961ae70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961af60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b050>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b140>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b230>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b320>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b410>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b500>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b5f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b6e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b7d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b8c0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961b9b0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961baa0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961bb90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961bc80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961bd70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961be60>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed961bf50>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638080>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638170>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638260>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638350>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638440>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638530>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638620>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638710>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638800>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96388f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96389e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638ad0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638bc0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638cb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638da0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638e90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9638f80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639070>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639160>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639250>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639340>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639430>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639520>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639610>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639700>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96397f0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96398e0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed96399d0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639ac0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639bb0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639ca0>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639d90>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639e80>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed9639f70>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed963a060>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed963a150>,\n",
       " <matplotlib.lines.Line2D at 0x7ffed963a240>]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(model_hf_state[\"transformer.wpe.weight\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Device set to use cpu\n",
      "Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precisely by providing a specific strategy to `truncation`.\n",
      "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[{'generated_text': \"Hello world, but what I'm really curious about is who I actually met on my wedding night in 2005, and what I was up to for the rest of my life?\\n\\nI think my answer will be a fairly straightforward one: it's\"},\n",
       " {'generated_text': \"Hello world, this is not the best world-view for real world use, and most often I find it when I'm using a VR headset on low resolution. There are a few areas where VR looks cool but it's only for single play sessions\"},\n",
       " {'generated_text': \"Hello world, if you'll follow me on twitter :P #crowdfunding\\n\\nThe Crowdfunding team was founded in 2017 after a year of service, in which there is no need to worry about the money involved in the project. We have\"},\n",
       " {'generated_text': \"Hello world, you've lost your soul.\\n\\nOh my, you're my mother! Why will your soul ever return?\\n\\nThis isn't like how some other children might be\\n\\nThat's so nice! My body hurts, you\"},\n",
       " {'generated_text': \"Hello world, I've lost all of the fun stuff. I've fallen in love with my dog and I've taken it up on this trip to Los Angeles. Now I don't know how I did it, but I had the whole experience,\"}]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import pipeline, set_seed\n",
    "\n",
    "generator = pipeline(\"text-generation\", model=\"gpt2\")\n",
    "set_seed(42)\n",
    "generator(\"Hello world,\", max_length=50, num_return_sequences=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First Citizen:\n",
      "Before we proceed any further, hear me speak.\n",
      "\n",
      "All:\n",
      "Speak, speak.\n",
      "\n",
      "First Citizen:\n",
      "You are all resolved rather to die than to famish?\n",
      "\n",
      "All:\n",
      "Resolved. resolved.\n",
      "\n",
      "First Citizen:\n",
      "First, you know Caius Marcius is chief enemy to the people.\n",
      "\n",
      "All:\n",
      "We know't, we know't.\n",
      "\n",
      "First Citizen:\n",
      "Let us kill him, and we'll have corn at our own price.\n",
      "Is't a verdict?\n",
      "\n",
      "All:\n",
      "No more talking on't; let it be done: away, away!\n",
      "\n",
      "Second Citizen:\n",
      "One word, good citizens.\n",
      "\n",
      "First Citizen:\n",
      "We are accounted poor citizens, the patricians good.\n",
      "What authority surfeits on would relieve us: if they\n",
      "would yield us but the superfluity, while it were\n",
      "wholesome, we might guess they relieved us humanely;\n",
      "but they think we are too dear: the leanness that\n",
      "afflicts us, the object of our misery, is as an\n",
      "inventory to particularise their abundance; our\n",
      "sufferance is a gain to them Let us revenge this with\n",
      "our pikes, ere we become rakes: for the gods know I\n",
      "speak this in hunger for bread, not in thirst for revenge.\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "with open(\"input.txt\", \"r\") as f:\n",
    "    text = f.read()\n",
    "\n",
    "data = text[:1000] # first 1000 characters\n",
    "print(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "GPT-2 has a compression ratio of ~3:1. 1000 characters correspond to ~300 tokens."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[5962, 22307, 25, 198, 8421, 356, 5120, 597, 2252, 11, 3285, 502, 2740, 13, 198, 198, 3237, 25, 198, 5248, 461, 11, 2740, 13, 198, 198, 5962, 22307, 25, 198, 1639, 389, 477, 12939, 2138, 284, 4656, 621, 284, 1145, 680, 30, 198, 198, 3237, 25, 198, 4965, 5634, 13, 12939, 13, 198, 198, 5962, 22307, 25, 198, 5962, 11, 345, 760, 327, 1872, 385, 1526, 28599, 318, 4039, 4472, 284, 262, 661, 13, 198, 198, 3237, 25, 198, 1135, 760, 470, 11, 356, 760, 470, 13, 198, 198, 5962, 22307, 25, 198, 5756, 514, 1494, 683, 11, 290, 356, 1183, 423, 11676, 379, 674, 898, 2756, 13, 198, 3792, 470, 257, 15593, 30, 198, 198, 3237, 25, 198, 2949, 517, 3375, 319, 470, 26, 1309, 340, 307, 1760, 25, 1497, 11, 1497, 0, 198, 198, 12211, 22307, 25, 198, 3198, 1573, 11, 922, 4290, 13, 198, 198, 5962, 22307, 25, 198, 1135, 389, 17830, 3595, 4290, 11, 262, 1458, 1173, 1547, 922, 13, 198, 2061, 4934, 969, 5036, 896, 319, 561, 26958, 514, 25, 611, 484, 198, 19188, 7800, 514, 475, 262, 48713, 414, 11, 981, 340, 547, 198, 1929, 4316, 462, 11, 356, 1244, 4724, 484, 22598, 514, 31533, 306, 26, 198, 4360, 484, 892, 356, 389, 1165, 13674, 25, 262, 10904, 1108, 326, 198, 2001, 42267, 514, 11, 262, 2134, 286, 674, 24672, 11, 318, 355, 281, 198, 24807, 284, 1948, 786, 511, 20038, 26, 674, 198, 82, 13712, 590, 318, 257, 4461, 284, 606, 3914, 514, 15827, 428, 351, 198, 454, 279, 7938, 11, 304, 260, 356, 1716, 374, 1124, 25, 329, 262, 11858, 760, 314, 198, 47350, 428, 287, 16460, 329, 8509, 11, 407, 287, 24613, 329, 15827, 13, 628]\n"
     ]
    }
   ],
   "source": [
    "import tiktoken\n",
    "\n",
    "enc = tiktoken.get_encoding(\"gpt2\")\n",
    "tokens = enc.encode(data)\n",
    "print(tokens)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We want to feed a transformer with this sequence of tokens, so they need to be re-aranged to build `idx` of shape (B, T)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[ 5962, 22307,    25,   198,  8421,   356],\n",
      "        [ 5120,   597,  2252,    11,  3285,   502],\n",
      "        [ 2740,    13,   198,   198,  3237,    25],\n",
      "        [  198,  5248,   461,    11,  2740,    13]])\n",
      "tensor([[22307,    25,   198,  8421,   356,  5120],\n",
      "        [  597,  2252,    11,  3285,   502,  2740],\n",
      "        [   13,   198,   198,  3237,    25,   198],\n",
      "        [ 5248,   461,    11,  2740,    13,   198]])\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "\n",
    "buf = torch.tensor(tokens[:24 + 1])\n",
    "x = buf[:-1].view(4, 6)\n",
    "y = buf[1:].view(4, 6)\n",
    "print(x)\n",
    "print(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A data loader could look as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[ 5962, 22307,    25,   198,  8421,   356,  5120,   597,  2252,    11,\n",
       "           3285,   502,  2740,    13,   198,   198,  3237,    25,   198,  5248,\n",
       "            461,    11,  2740,    13,   198,   198,  5962, 22307,    25,   198,\n",
       "           1639,   389],\n",
       "         [  477, 12939,  2138,   284,  4656,   621,   284,  1145,   680,    30,\n",
       "            198,   198,  3237,    25,   198,  4965,  5634,    13, 12939,    13,\n",
       "            198,   198,  5962, 22307,    25,   198,  5962,    11,   345,   760,\n",
       "            327,  1872],\n",
       "         [  385,  1526, 28599,   318,  4039,  4472,   284,   262,   661,    13,\n",
       "            198,   198,  3237,    25,   198,  1135,   760,   470,    11,   356,\n",
       "            760,   470,    13,   198,   198,  5962, 22307,    25,   198,  5756,\n",
       "            514,  1494],\n",
       "         [  683,    11,   290,   356,  1183,   423, 11676,   379,   674,   898,\n",
       "           2756,    13,   198,  3792,   470,   257, 15593,    30,   198,   198,\n",
       "           3237,    25,   198,  2949,   517,  3375,   319,   470,    26,  1309,\n",
       "            340,   307]]),\n",
       " tensor([[22307,    25,   198,  8421,   356,  5120,   597,  2252,    11,  3285,\n",
       "            502,  2740,    13,   198,   198,  3237,    25,   198,  5248,   461,\n",
       "             11,  2740,    13,   198,   198,  5962, 22307,    25,   198,  1639,\n",
       "            389,   477],\n",
       "         [12939,  2138,   284,  4656,   621,   284,  1145,   680,    30,   198,\n",
       "            198,  3237,    25,   198,  4965,  5634,    13, 12939,    13,   198,\n",
       "            198,  5962, 22307,    25,   198,  5962,    11,   345,   760,   327,\n",
       "           1872,   385],\n",
       "         [ 1526, 28599,   318,  4039,  4472,   284,   262,   661,    13,   198,\n",
       "            198,  3237,    25,   198,  1135,   760,   470,    11,   356,   760,\n",
       "            470,    13,   198,   198,  5962, 22307,    25,   198,  5756,   514,\n",
       "           1494,   683],\n",
       "         [   11,   290,   356,  1183,   423, 11676,   379,   674,   898,  2756,\n",
       "             13,   198,  3792,   470,   257, 15593,    30,   198,   198,  3237,\n",
       "             25,   198,  2949,   517,  3375,   319,   470,    26,  1309,   340,\n",
       "            307,  1760]]))"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load some input data and prefix tokens\n",
    "def load_data():\n",
    "    import tiktoken\n",
    "    enc = tiktoken.get_encoding(\"gpt2\")\n",
    "\n",
    "    with open(\"input.txt\", \"r\") as f:\n",
    "        data = f.read()\n",
    "\n",
    "    data = data[:1000]\n",
    "    tokens = enc.encode(data)\n",
    "    B, T = 4, 32 # T = block_size = bs\n",
    "    buf = torch.tensor(tokens[:B * T + 1])\n",
    "\n",
    "    x = buf[:-1].view(B, T)\n",
    "    y = buf[1:].view(B, T)\n",
    "\n",
    "    return x, y\n",
    "\n",
    "load_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}