test.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PyTorch\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "\n",
    "\n",
    "def create_torch_tensors(device):\n",
    "    x = torch.rand((10000, 10000), dtype=torch.float32)\n",
    "    y = torch.rand((10000, 10000), dtype=torch.float32)\n",
    "    x = x.to(device)\n",
    "    y = y.to(device)\n",
    "\n",
    "    return x, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device(\"cpu\")\n",
    "x, y = create_torch_tensors(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14.4 ms ± 131 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "x * y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device(\"mps\")\n",
    "x, y = create_torch_tensors(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.72 ms ± 19.4 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "x * y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TensorFlow\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "\n",
    "def create_tf_tensors():\n",
    "    x = tf.random.uniform((10000, 10000), dtype=tf.float32)\n",
    "    y = tf.random.uniform((10000, 10000), dtype=tf.float32)\n",
    "\n",
    "    return x, y\n",
    "\n",
    "\n",
    "x, y = create_tf_tensors()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15.3 ms ± 106 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "\n",
    "with tf.device(\"/CPU:0\"):\n",
    "    x * y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.71 ms ± 25.6 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "\n",
    "with tf.device(\"/GPU:0\"):\n",
    "    x * y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## JAX\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "os.environ[\"JAX_PLATFORMS\"] = \"cpu\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Platform 'METAL' is experimental and not all JAX functionality may be correctly supported!\n",
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "W0000 00:00:1716385928.837884  732948 mps_client.cc:510] WARNING: JAX Apple GPU support is experimental and not all JAX functionality is correctly supported!\n",
      "I0000 00:00:1716385928.845272  732948 service.cc:145] XLA service 0x12da04c10 initialized for platform METAL (this does not guarantee that XLA will be used). Devices:\n",
      "I0000 00:00:1716385928.845282  732948 service.cc:153]   StreamExecutor device (0): Metal, <undefined>\n",
      "I0000 00:00:1716385928.847533  732948 mps_client.cc:406] Using Simple allocator.\n",
      "I0000 00:00:1716385928.847540  732948 mps_client.cc:384] XLA backend will use up to 103077429248 bytes on device 0 for SimpleAllocator.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M3 Max\n",
      "\n",
      "systemMemory: 128.00 GB\n",
      "maxCacheSize: 48.00 GB\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import jax\n",
    "import jax.numpy as jnp\n",
    "\n",
    "\n",
    "def create_jax_tensors():\n",
    "    x = jax.random.uniform(jax.random.PRNGKey(0), (10000, 10000), dtype=jnp.float32)\n",
    "    y = jax.random.uniform(jax.random.PRNGKey(1), (10000, 10000), dtype=jnp.float32)\n",
    "\n",
    "    return x, y\n",
    "\n",
    "\n",
    "x, y = create_jax_tensors()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.72 ms ± 39.8 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "x * y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}