shallow-training/scratchpad.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "physical-coating",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "tf.get_logger().setLevel('ERROR')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl\n",
    "import seaborn as sns\n",
    "import json\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "fig_dpi = 200\n",
    "\n",
    "%load_ext tensorboard"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "unable-security",
   "metadata": {},
   "source": [
    "# Scratchpad\n",
    "Testbed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "precise-invalid",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('features.csv', header=None).T\n",
    "data.columns = ['Clump thickness', 'Uniformity of cell size', 'Uniformity of cell shape', 'Marginal adhesion', 'Single epithelial cell size', 'Bare nuclei', 'Bland chomatin', 'Normal nucleoli', 'Mitoses']\n",
    "labels = pd.read_csv('targets.csv', header=None).T\n",
    "labels.columns = ['Benign', 'Malignant']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "equal-cooling",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train, data_test, labels_train, labels_test = train_test_split(data, labels, test_size=0.5, stratify=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "valuable-illinois",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_2 (Dense)              (None, 50)                500       \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 2)                 102       \n",
      "=================================================================\n",
      "Total params: 602\n",
      "Trainable params: 602\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "layers = [tf.keras.layers.InputLayer(input_shape=(9,)), \n",
    "          tf.keras.layers.Dense(50, activation='sigmoid'), \n",
    "          tf.keras.layers.Dense(2, activation='softmax')]\n",
    "\n",
    "model = tf.keras.models.Sequential(layers)\n",
    "model.compile('sgd', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "opened-terminology",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "10/10 [==============================] - 0s 16ms/step - loss: 0.6577 - accuracy: 0.6181 - val_loss: 0.6124 - val_accuracy: 0.6857\n"
     ]
    }
   ],
   "source": [
    "e=50\n",
    "history = model.fit(data_train, labels_train, epochs=e, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "forward-asthma",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.evaluate(data_test, \n",
    "               labels_test, \n",
    "               batch_size=128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "local-program",
   "metadata": {},
   "outputs": [],
   "source": [
    "json.loads(model.to_json())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "animated-raise",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'_self_setattr_tracking': True,\n",
       " '_is_model_for_instrumentation': True,\n",
       " '_instrumented_keras_api': True,\n",
       " '_instrumented_keras_layer_class': False,\n",
       " '_instrumented_keras_model_class': True,\n",
       " '_trainable': True,\n",
       " '_stateful': False,\n",
       " 'built': True,\n",
       " '_build_input_shape': TensorShape([None, 9]),\n",
       " '_saved_model_inputs_spec': TensorSpec(shape=(None, 9), dtype=tf.float32, name='input_1'),\n",
       " '_input_spec': None,\n",
       " '_supports_masking': True,\n",
       " '_name': 'sequential',\n",
       " '_activity_regularizer': None,\n",
       " '_trainable_weights': [],\n",
       " '_non_trainable_weights': [],\n",
       " '_updates': [],\n",
       " '_thread_local': <_thread._local at 0x1e9b3938c70>,\n",
       " '_callable_losses': [],\n",
       " '_losses': [],\n",
       " '_metrics': [],\n",
       " '_metrics_lock': <unlocked _thread.lock object at 0x000001E9CF471E40>,\n",
       " '_dtype_policy': <Policy \"float32\">,\n",
       " '_compute_dtype_object': tf.float32,\n",
       " '_autocast': False,\n",
       " '_layers': [<tensorflow.python.keras.engine.input_layer.InputLayer at 0x1e9cf471eb0>,\n",
       "  <tensorflow.python.keras.layers.core.Dense at 0x1e9cf4c8f10>,\n",
       "  <tensorflow.python.keras.layers.core.Dense at 0x1e9cf518250>],\n",
       " '_inbound_nodes_value': [],\n",
       " '_outbound_nodes_value': [],\n",
       " '_expects_training_arg': True,\n",
       " '_default_training_arg': None,\n",
       " '_expects_mask_arg': True,\n",
       " '_dynamic': False,\n",
       " '_initial_weights': None,\n",
       " '_auto_track_sub_layers': False,\n",
       " '_preserve_input_structure_in_config': False,\n",
       " '_is_graph_network': True,\n",
       " 'inputs': [<KerasTensor: shape=(None, 9) dtype=float32 (created by layer 'input_1')>],\n",
       " 'outputs': [<KerasTensor: shape=(None, 2) dtype=float32 (created by layer 'dense_1')>],\n",
       " 'input_names': ['input_1'],\n",
       " 'output_names': ['dense_1'],\n",
       " '_compute_output_and_mask_jointly': True,\n",
       " '_distribution_strategy': None,\n",
       " 'predict_function': None,\n",
       " '_compiled_trainable_state': <WeakKeyDictionary at 0x1e9cf8449a0>,\n",
       " '_training_state': None,\n",
       " '_trackable_saver': <tensorflow.python.training.tracking.util.TrackableSaver at 0x1e9cf518640>,\n",
       " '_steps_per_execution': <tf.Variable 'Variable:0' shape=() dtype=int64, numpy=1>,\n",
       " '_train_counter': <tf.Variable 'Variable:0' shape=() dtype=int64, numpy=500>,\n",
       " '_test_counter': <tf.Variable 'Variable:0' shape=() dtype=int64, numpy=3>,\n",
       " '_predict_counter': <tf.Variable 'Variable:0' shape=() dtype=int64, numpy=0>,\n",
       " '_base_model_initialized': True,\n",
       " '_inferred_input_shape': None,\n",
       " '_has_explicit_input_shape': True,\n",
       " '_input_dtype': None,\n",
       " '_layer_call_argspecs': {<tensorflow.python.keras.engine.input_layer.InputLayer at 0x1e9cf471eb0>: FullArgSpec(args=['self', 'inputs'], varargs=None, varkw='kwargs', defaults=None, kwonlyargs=[], kwonlydefaults=None, annotations={}),\n",
       "  <tensorflow.python.keras.layers.core.Dense at 0x1e9cf4c8f10>: FullArgSpec(args=['self', 'inputs'], varargs=None, varkw=None, defaults=None, kwonlyargs=[], kwonlydefaults=None, annotations={}),\n",
       "  <tensorflow.python.keras.layers.core.Dense at 0x1e9cf518250>: FullArgSpec(args=['self', 'inputs'], varargs=None, varkw=None, defaults=None, kwonlyargs=[], kwonlydefaults=None, annotations={})},\n",
       " '_created_nodes': set(),\n",
       " '_graph_initialized': True,\n",
       " '_use_legacy_deferred_behavior': False,\n",
       " '_nested_inputs': <KerasTensor: shape=(None, 9) dtype=float32 (created by layer 'input_1')>,\n",
       " '_nested_outputs': <KerasTensor: shape=(None, 2) dtype=float32 (created by layer 'dense_1')>,\n",
       " '_enable_dict_to_input_mapping': True,\n",
       " '_input_layers': [<tensorflow.python.keras.engine.input_layer.InputLayer at 0x1e9cf471eb0>],\n",
       " '_output_layers': [<tensorflow.python.keras.layers.core.Dense at 0x1e9cf518250>],\n",
       " '_input_coordinates': [(<tensorflow.python.keras.engine.input_layer.InputLayer at 0x1e9cf471eb0>,\n",
       "   0,\n",
       "   0)],\n",
       " '_output_coordinates': [(<tensorflow.python.keras.layers.core.Dense at 0x1e9cf518250>,\n",
       "   0,\n",
       "   0)],\n",
       " '_output_mask_cache': {},\n",
       " '_output_tensor_cache': {},\n",
       " '_output_shape_cache': {},\n",
       " '_network_nodes': {'dense_1_ib-0', 'dense_ib-0', 'input_1_ib-0'},\n",
       " '_nodes_by_depth': defaultdict(list,\n",
       "             {0: [<tensorflow.python.keras.engine.node.Node at 0x1e9cf831f70>],\n",
       "              1: [<tensorflow.python.keras.engine.node.Node at 0x1e9cf518190>],\n",
       "              2: [<tensorflow.python.keras.engine.node.Node at 0x1e9cf4c8d30>]}),\n",
       " '_feed_input_names': ['input_1'],\n",
       " '_feed_inputs': [<KerasTensor: shape=(None, 9) dtype=float32 (created by layer 'input_1')>],\n",
       " '_feed_input_shapes': [(None, 9)],\n",
       " '_tensor_usage_count': Counter({'2103716908480': 1,\n",
       "          '2103720484432': 1,\n",
       "          '2103720530848': 1}),\n",
       " '_obj_reference_counts_dict': ObjectIdentityDictionary({<_ObjectIdentityWrapper wrapping <tensorflow.python.keras.optimizer_v2.gradient_descent.SGD object at 0x000001E9CF518430>>: 1, <_ObjectIdentityWrapper wrapping <tensorflow.python.keras.engine.compile_utils.LossesContainer object at 0x000001E9CF83D8E0>>: 1, <_ObjectIdentityWrapper wrapping <tensorflow.python.keras.engine.compile_utils.MetricsContainer object at 0x000001E9CF844940>>: 1, <_ObjectIdentityWrapper wrapping True>: 1, <_ObjectIdentityWrapper wrapping 'categorical_crossentropy'>: 1, <_ObjectIdentityWrapper wrapping <tensorflow.python.eager.def_function.Function object at 0x000001E9D6A4AEE0>>: 1, <_ObjectIdentityWrapper wrapping <tensorflow.python.eager.def_function.Function object at 0x000001E9F00CBA90>>: 1, <_ObjectIdentityWrapper wrapping <tensorflow.python.keras.callbacks.History object at 0x000001E9F2BBFB50>>: 1}),\n",
       " '_run_eagerly': None,\n",
       " '_self_unconditional_checkpoint_dependencies': [TrackableReference(name='optimizer', ref=<tensorflow.python.keras.optimizer_v2.gradient_descent.SGD object at 0x000001E9CF518430>)],\n",
       " '_self_unconditional_dependency_names': {'optimizer': <tensorflow.python.keras.optimizer_v2.gradient_descent.SGD at 0x1e9cf518430>},\n",
       " '_self_unconditional_deferred_dependencies': {},\n",
       " '_self_update_uid': -1,\n",
       " '_self_name_based_restores': set(),\n",
       " '_self_saveable_object_factories': {},\n",
       " 'optimizer': <tensorflow.python.keras.optimizer_v2.gradient_descent.SGD at 0x1e9cf518430>,\n",
       " 'compiled_loss': <tensorflow.python.keras.engine.compile_utils.LossesContainer at 0x1e9cf83d8e0>,\n",
       " 'compiled_metrics': <tensorflow.python.keras.engine.compile_utils.MetricsContainer at 0x1e9cf844940>,\n",
       " '_is_compiled': True,\n",
       " 'loss': 'categorical_crossentropy',\n",
       " 'stop_training': False,\n",
       " 'train_function': <tensorflow.python.eager.def_function.Function at 0x1e9d6a4aee0>,\n",
       " 'history': <tensorflow.python.keras.callbacks.History at 0x1e9f2bbfb50>,\n",
       " 'test_function': <tensorflow.python.eager.def_function.Function at 0x1e9f00cba90>}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.__dict__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "third-accuracy",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.optimizer.get_config()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "synthetic-armstrong",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3000x1400 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(15,7))\n",
    "fig.set_dpi(fig_dpi)\n",
    "\n",
    "ax = axes[0]\n",
    "ax.set_title(\"Training vs Validation Loss\")\n",
    "ax.plot(history.history['loss'], label=\"train\", lw=2)\n",
    "ax.plot(history.history['val_loss'], label=\"validation\", lw=2, c=(1,0,0))\n",
    "ax.set_xlabel(\"Epochs\")\n",
    "ax.legend()\n",
    "\n",
    "ax = axes[1]\n",
    "ax.set_title(\"Training vs Validation Accuracy\")\n",
    "ax.plot(history.history['accuracy'], label=\"train\", lw=2)\n",
    "ax.plot(history.history['val_accuracy'], label=\"validation\", lw=2, c=(1,0,0))\n",
    "ax.set_xlabel(\"Epochs\")\n",
    "ax.set_ylim(0, 1)\n",
    "ax.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "coordinated-salvation",
   "metadata": {},
   "outputs": [],
   "source": [
    "tensor = model(data_test.to_numpy())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "roman-explorer",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(350, 2), dtype=float32, numpy=\n",
       "array([[0.9801853 , 0.01981472],\n",
       "       [0.9926891 , 0.00731087],\n",
       "       [0.9628417 , 0.03715829],\n",
       "       [0.03073526, 0.9692647 ],\n",
       "       [0.95234215, 0.04765787],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.10280664, 0.8971934 ],\n",
       "       [0.07686787, 0.9231321 ],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9831904 , 0.01680959],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.98902273, 0.01097724],\n",
       "       [0.98902273, 0.01097724],\n",
       "       [0.98237324, 0.01762678],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.01222708, 0.9877729 ],\n",
       "       [0.97519624, 0.02480372],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.08153479, 0.9184652 ],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.05978837, 0.94021165],\n",
       "       [0.9801853 , 0.01981472],\n",
       "       [0.09358412, 0.9064159 ],\n",
       "       [0.9801853 , 0.01981472],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.5003021 , 0.49969792],\n",
       "       [0.9815221 , 0.01847788],\n",
       "       [0.11047691, 0.889523  ],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.00534083, 0.9946591 ],\n",
       "       [0.9825487 , 0.0174513 ],\n",
       "       [0.97305536, 0.02694468],\n",
       "       [0.9633045 , 0.03669543],\n",
       "       [0.9898519 , 0.01014804],\n",
       "       [0.95956725, 0.04043271],\n",
       "       [0.11385527, 0.88614476],\n",
       "       [0.0526755 , 0.9473245 ],\n",
       "       [0.08191174, 0.91808826],\n",
       "       [0.975471  , 0.02452892],\n",
       "       [0.96881   , 0.03119001],\n",
       "       [0.9862664 , 0.01373362],\n",
       "       [0.02994935, 0.97005063],\n",
       "       [0.9898519 , 0.01014804],\n",
       "       [0.16885298, 0.831147  ],\n",
       "       [0.97960424, 0.02039572],\n",
       "       [0.9884575 , 0.01154249],\n",
       "       [0.11157885, 0.8884212 ],\n",
       "       [0.9038618 , 0.09613817],\n",
       "       [0.9811632 , 0.01883673],\n",
       "       [0.71255827, 0.28744167],\n",
       "       [0.01894411, 0.9810559 ],\n",
       "       [0.9862664 , 0.01373362],\n",
       "       [0.02488227, 0.9751177 ],\n",
       "       [0.9899156 , 0.01008447],\n",
       "       [0.08436204, 0.91563797],\n",
       "       [0.97465366, 0.02534635],\n",
       "       [0.9788304 , 0.02116962],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9596988 , 0.04030127],\n",
       "       [0.06126701, 0.938733  ],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.06121646, 0.9387835 ],\n",
       "       [0.34550568, 0.65449435],\n",
       "       [0.99108255, 0.00891742],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.00280253, 0.9971975 ],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.99204755, 0.00795248],\n",
       "       [0.29036736, 0.7096326 ],\n",
       "       [0.98586327, 0.01413671],\n",
       "       [0.97960806, 0.0203919 ],\n",
       "       [0.98411894, 0.01588103],\n",
       "       [0.04055419, 0.9594458 ],\n",
       "       [0.9881726 , 0.01182744],\n",
       "       [0.1958921 , 0.8041079 ],\n",
       "       [0.02864912, 0.9713509 ],\n",
       "       [0.00602777, 0.99397224],\n",
       "       [0.08019029, 0.91980964],\n",
       "       [0.11171903, 0.8882809 ],\n",
       "       [0.9881726 , 0.01182744],\n",
       "       [0.17073832, 0.82926166],\n",
       "       [0.22782806, 0.7721719 ],\n",
       "       [0.44395986, 0.5560401 ],\n",
       "       [0.6307267 , 0.36927328],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.9818465 , 0.01815345],\n",
       "       [0.78953874, 0.21046132],\n",
       "       [0.00406485, 0.9959351 ],\n",
       "       [0.9814414 , 0.01855859],\n",
       "       [0.00271158, 0.99728847],\n",
       "       [0.96431845, 0.03568152],\n",
       "       [0.02134761, 0.9786524 ],\n",
       "       [0.96153224, 0.03846784],\n",
       "       [0.05516699, 0.944833  ],\n",
       "       [0.9775564 , 0.02244362],\n",
       "       [0.97136647, 0.02863356],\n",
       "       [0.98825955, 0.01174045],\n",
       "       [0.97474086, 0.02525917],\n",
       "       [0.9838537 , 0.01614636],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.04362229, 0.95637774],\n",
       "       [0.00622424, 0.9937757 ],\n",
       "       [0.9775203 , 0.02247973],\n",
       "       [0.3481092 , 0.6518908 ],\n",
       "       [0.01587938, 0.98412067],\n",
       "       [0.9884575 , 0.01154249],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9825487 , 0.0174513 ],\n",
       "       [0.980631  , 0.01936896],\n",
       "       [0.9746114 , 0.02538859],\n",
       "       [0.9820693 , 0.01793065],\n",
       "       [0.88999826, 0.11000182],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.9899156 , 0.01008447],\n",
       "       [0.18822296, 0.811777  ],\n",
       "       [0.98365664, 0.01634333],\n",
       "       [0.9775203 , 0.02247973],\n",
       "       [0.02279323, 0.9772068 ],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9893377 , 0.0106623 ],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.9917625 , 0.00823749],\n",
       "       [0.9827151 , 0.01728498],\n",
       "       [0.04650781, 0.9534922 ],\n",
       "       [0.9721344 , 0.02786556],\n",
       "       [0.04326608, 0.95673394],\n",
       "       [0.9898519 , 0.01014804],\n",
       "       [0.9811445 , 0.0188555 ],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9820363 , 0.01796375],\n",
       "       [0.5693888 , 0.4306112 ],\n",
       "       [0.9842153 , 0.01578473],\n",
       "       [0.00773495, 0.9922651 ],\n",
       "       [0.98602563, 0.01397439],\n",
       "       [0.9855597 , 0.01444028],\n",
       "       [0.9381649 , 0.06183509],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.02490397, 0.97509605],\n",
       "       [0.09574957, 0.9042504 ],\n",
       "       [0.13992904, 0.860071  ],\n",
       "       [0.03468701, 0.96531296],\n",
       "       [0.9717713 , 0.02822872],\n",
       "       [0.9858747 , 0.01412521],\n",
       "       [0.9581099 , 0.04189004],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.00402759, 0.99597245],\n",
       "       [0.04732035, 0.95267963],\n",
       "       [0.05004258, 0.9499575 ],\n",
       "       [0.97465366, 0.02534635],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9893835 , 0.01061649],\n",
       "       [0.3334639 , 0.66653615],\n",
       "       [0.90755075, 0.09244923],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.98903185, 0.01096814],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.96168005, 0.03831991],\n",
       "       [0.981965  , 0.01803497],\n",
       "       [0.98902273, 0.01097724],\n",
       "       [0.00236953, 0.9976305 ],\n",
       "       [0.08982091, 0.9101791 ],\n",
       "       [0.1135387 , 0.8864613 ],\n",
       "       [0.9938671 , 0.0061329 ],\n",
       "       [0.13269322, 0.8673067 ],\n",
       "       [0.61057794, 0.38942203],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9801853 , 0.01981472],\n",
       "       [0.9903584 , 0.00964163],\n",
       "       [0.97204506, 0.02795494],\n",
       "       [0.9853139 , 0.01468608],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.9853094 , 0.01469058],\n",
       "       [0.16317306, 0.836827  ],\n",
       "       [0.02048163, 0.97951835],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.97727305, 0.02272695],\n",
       "       [0.00605206, 0.99394786],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.14910528, 0.85089475],\n",
       "       [0.01027166, 0.98972833],\n",
       "       [0.9404712 , 0.0595288 ],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9151222 , 0.08487778],\n",
       "       [0.17249718, 0.82750285],\n",
       "       [0.7364639 , 0.2635361 ],\n",
       "       [0.9560006 , 0.04399942],\n",
       "       [0.9898519 , 0.01014804],\n",
       "       [0.14065848, 0.8593415 ],\n",
       "       [0.06515042, 0.93484956],\n",
       "       [0.00246285, 0.99753714],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.99190474, 0.00809523],\n",
       "       [0.86330867, 0.13669129],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.9825487 , 0.0174513 ],\n",
       "       [0.16500047, 0.8349995 ],\n",
       "       [0.53791696, 0.4620831 ],\n",
       "       [0.9898519 , 0.01014804],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9708858 , 0.02911417],\n",
       "       [0.98892224, 0.01107774],\n",
       "       [0.5222743 , 0.47772565],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.00319345, 0.9968066 ],\n",
       "       [0.98514056, 0.01485936],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.9548213 , 0.04517876],\n",
       "       [0.98882884, 0.01117116],\n",
       "       [0.582204  , 0.41779602],\n",
       "       [0.95525944, 0.04474052],\n",
       "       [0.9811445 , 0.0188555 ],\n",
       "       [0.06356736, 0.93643266],\n",
       "       [0.99108255, 0.00891742],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.97405064, 0.02594929],\n",
       "       [0.9923038 , 0.0076962 ],\n",
       "       [0.99204755, 0.00795248],\n",
       "       [0.980293  , 0.01970703],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.975471  , 0.02452892],\n",
       "       [0.97204506, 0.02795494],\n",
       "       [0.040151  , 0.959849  ],\n",
       "       [0.99301505, 0.00698495],\n",
       "       [0.52752644, 0.47247356],\n",
       "       [0.03305454, 0.96694547],\n",
       "       [0.9675697 , 0.03243022],\n",
       "       [0.02140073, 0.9785993 ],\n",
       "       [0.01063837, 0.9893616 ],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.97960806, 0.0203919 ],\n",
       "       [0.10082608, 0.89917386],\n",
       "       [0.0673348 , 0.93266517],\n",
       "       [0.9447868 , 0.05521324],\n",
       "       [0.01348197, 0.98651797],\n",
       "       [0.99204755, 0.00795248],\n",
       "       [0.00669446, 0.9933055 ],\n",
       "       [0.00779293, 0.9922071 ],\n",
       "       [0.00551592, 0.994484  ],\n",
       "       [0.9771452 , 0.02285476],\n",
       "       [0.0136543 , 0.9863457 ],\n",
       "       [0.98627186, 0.01372816],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.46391317, 0.53608686],\n",
       "       [0.9801853 , 0.01981472],\n",
       "       [0.72362626, 0.27637374],\n",
       "       [0.09471217, 0.90528786],\n",
       "       [0.0470202 , 0.95297986],\n",
       "       [0.00357349, 0.9964265 ],\n",
       "       [0.04937639, 0.95062363],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.9836901 , 0.01630987],\n",
       "       [0.07471903, 0.9252809 ],\n",
       "       [0.98411894, 0.01588103],\n",
       "       [0.05152437, 0.94847566],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.9871589 , 0.01284102],\n",
       "       [0.95937496, 0.04062498],\n",
       "       [0.03402388, 0.9659761 ],\n",
       "       [0.10793284, 0.8920672 ],\n",
       "       [0.11131216, 0.88868785],\n",
       "       [0.9865871 , 0.01341282],\n",
       "       [0.9855597 , 0.01444028],\n",
       "       [0.01075125, 0.98924875],\n",
       "       [0.98955685, 0.01044314],\n",
       "       [0.99108255, 0.00891742],\n",
       "       [0.01742107, 0.9825789 ],\n",
       "       [0.9616439 , 0.03835611],\n",
       "       [0.9217987 , 0.07820132],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.01513559, 0.9848644 ],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.981965  , 0.01803497],\n",
       "       [0.9776144 , 0.02238566],\n",
       "       [0.05140878, 0.9485912 ],\n",
       "       [0.84267443, 0.15732561],\n",
       "       [0.9522025 , 0.04779752],\n",
       "       [0.9712332 , 0.02876678],\n",
       "       [0.99301505, 0.00698495],\n",
       "       [0.10411921, 0.89588076],\n",
       "       [0.9889004 , 0.01109953],\n",
       "       [0.01602884, 0.9839712 ],\n",
       "       [0.99204755, 0.00795248],\n",
       "       [0.9881715 , 0.01182842],\n",
       "       [0.01760418, 0.9823958 ],\n",
       "       [0.9705625 , 0.0294375 ],\n",
       "       [0.00573339, 0.99426657],\n",
       "       [0.9855597 , 0.01444028],\n",
       "       [0.20140044, 0.79859954],\n",
       "       [0.96860284, 0.03139716],\n",
       "       [0.8577237 , 0.14227627],\n",
       "       [0.98384744, 0.01615255],\n",
       "       [0.04443884, 0.9555612 ],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.36669564, 0.63330436],\n",
       "       [0.06703284, 0.9329671 ],\n",
       "       [0.17048047, 0.8295195 ],\n",
       "       [0.9775203 , 0.02247973],\n",
       "       [0.93597096, 0.06402898],\n",
       "       [0.9475726 , 0.05242743],\n",
       "       [0.9791864 , 0.02081357],\n",
       "       [0.9817647 , 0.01823526],\n",
       "       [0.99108255, 0.00891742],\n",
       "       [0.9872952 , 0.01270485],\n",
       "       [0.97686416, 0.02313586],\n",
       "       [0.9803022 , 0.01969773],\n",
       "       [0.9791864 , 0.02081357],\n",
       "       [0.9478458 , 0.05215417],\n",
       "       [0.00895109, 0.99104893],\n",
       "       [0.8635009 , 0.13649909],\n",
       "       [0.98365664, 0.01634333],\n",
       "       [0.03919939, 0.96080065],\n",
       "       [0.63790035, 0.36209962],\n",
       "       [0.0029722 , 0.9970278 ],\n",
       "       [0.98825574, 0.01174434],\n",
       "       [0.96775603, 0.03224399],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.9372071 , 0.06279288],\n",
       "       [0.12397458, 0.8760254 ],\n",
       "       [0.93621886, 0.06378116],\n",
       "       [0.94735366, 0.05264638],\n",
       "       [0.99078685, 0.00921312],\n",
       "       [0.971693  , 0.02830696],\n",
       "       [0.19744301, 0.802557  ],\n",
       "       [0.99301505, 0.00698495],\n",
       "       [0.48893696, 0.51106304],\n",
       "       [0.00531539, 0.99468464],\n",
       "       [0.00586492, 0.9941351 ],\n",
       "       [0.97135025, 0.02864975],\n",
       "       [0.9771452 , 0.02285476],\n",
       "       [0.0034244 , 0.99657553],\n",
       "       [0.9855597 , 0.01444028],\n",
       "       [0.00339224, 0.9966078 ],\n",
       "       [0.9928901 , 0.00710991],\n",
       "       [0.0600323 , 0.9399677 ],\n",
       "       [0.01615115, 0.9838488 ],\n",
       "       [0.9857479 , 0.01425208],\n",
       "       [0.41190255, 0.58809745],\n",
       "       [0.9492881 , 0.0507119 ],\n",
       "       [0.01772507, 0.9822749 ]], dtype=float32)>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "caring-assets",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(350, 2), dtype=float32, numpy=\n",
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.]], dtype=float32)>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.round(tensor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "controversial-modern",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2,), dtype=int32, numpy=array([2, 0])>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.constant([1, 0]) + tf.constant([1, 0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "correct-lodging",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2,), dtype=bool, numpy=array([False, False])>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.constant([1, 0]) == tf.constant([0, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "suited-standard",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor([1 0], shape=(2,), dtype=int64)\n",
      "tf.Tensor([ True  True], shape=(2,), dtype=bool)\n",
      "tf.Tensor(0, shape=(), dtype=int64)\n",
      "\n",
      "tf.Tensor([1 0], shape=(2,), dtype=int64)\n",
      "tf.Tensor([ True  True], shape=(2,), dtype=bool)\n",
      "tf.Tensor(0, shape=(), dtype=int64)\n",
      "\n",
      "tf.Tensor([1 0], shape=(2,), dtype=int64)\n",
      "tf.Tensor([ True  True], shape=(2,), dtype=bool)\n",
      "tf.Tensor(0, shape=(), dtype=int64)\n",
      "\n",
      "tf.Tensor([0 1], shape=(2,), dtype=int64)\n",
      "tf.Tensor([False False], shape=(2,), dtype=bool)\n",
      "tf.Tensor(1, shape=(), dtype=int64)\n",
      "\n",
      "tf.Tensor([1 0], shape=(2,), dtype=int64)\n",
      "tf.Tensor([ True  True], shape=(2,), dtype=bool)\n",
      "tf.Tensor(0, shape=(), dtype=int64)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for c in tf.constant(labels_test)[:5]:\n",
    "    print(c)\n",
    "    print(c == tf.constant([1, 0], dtype='int64'))\n",
    "    print(tf.math.argmax(c))\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "greater-publisher",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(2,), dtype=int32, numpy=array([0, 2])>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.math.reduce_sum(tf.constant([[0, 1], [0, 1]]), axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "considerable-fluid",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5.])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.linspace(5, 150, num=1)"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}