making_functions_pure.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "making_functions_pure.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPBb/AzEzRItZZV4AzHZMnz",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/deterministic-algorithms-lab/Jax-Journey/blob/main/making_functions_pure.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lm3SRlDt4DRC"
      },
      "source": [
        "#Making Purified Versions of Impure Functions"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NEK2kZWa34cl"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f0XWWh8l0Ji7"
      },
      "source": [
        "In the class below, \n",
        "\n",
        "* ```params``` acts as a simple placeholders for what would be sent to ```pure_wrapped_user_fn``` later on. \n",
        "\n",
        "* ```self.params``` over-rides and hides the original class varibale ```params```.\n",
        "\n",
        "* The ```self.get_param``` is just a protection to ensure that the end user of the class always accesses the class' functionality by passing ```params``` within ```pure_wrapped_user_fn```. If the user would not make mistakes(i.e., not call ```impure_user_fn``` directly), the ```get_param``` thing can be entirely eliminated and we can directly access ```self.param['w']``` etc. in the ```impure_user_fn```. \n",
        "\n",
        "* Notice how ```self.params``` and ```self.is_running_pure``` are set only when we enter the pure function, and reset while leaving it. This behavior will be seen in Flax too, later on.. where you won't even be able to access submodules, variables etc. you register in the ```setup()``` function using calls  like ```print(model.variable)```, outside of the class. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FiSGdSigvYbD"
      },
      "source": [
        "class PurifyingRegressor():\n",
        "\n",
        "    # This is the function the user wants to write.\n",
        "    def impure_user_fn(self, x):\n",
        "        w = self.get_param('w')\n",
        "        b = self.get_param('b')\n",
        "        return w * x + b\n",
        "\n",
        "    # This is the wrapping/mode logic.\n",
        "    params = {}                                                                 \n",
        "    is_running_pure = False\n",
        "\n",
        "    def get_param(self, name):\n",
        "        if not self.is_running_pure:\n",
        "            raise Exception(\"We can only call this when wrapped!\")\n",
        "        else:\n",
        "            return self.params[name]\n",
        "\n",
        "    # This is the function that we want JAX to use!\n",
        "    # Note how it defers all calculation and logic to\n",
        "    # the user-written and just provides *context*!\n",
        "    def pure_wrapped_user_fn(self, params, *args):\n",
        "        self.is_running_pure = True\n",
        "        self.params = params                                        \n",
        "        result = self.impure_user_fn(*args)\n",
        "        self.is_running_pure = False\n",
        "        self.params = {}\n",
        "        return result"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rGiLpGY8viRi"
      },
      "source": [
        "pure_predict_fn = PurifyingRegressor().pure_wrapped_user_fn\n",
        "\n",
        "# Show that it's equal to this manually purified function:\n",
        "manually_pure_predict_fn = lambda params, x: params['w'] * x + params['b']\n",
        "\n",
        "args = ({'w': 13., 'b': 0.}, 42.)\n",
        "\n",
        "assert pure_predict_fn(*args) == manually_pure_predict_fn(*args)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j1KJGmZuvwvX"
      },
      "source": [
        "x = PurifyingRegressor()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XN2BGsjuw-Vb"
      },
      "source": [
        "x.pure_wrapped_user_fn(*args)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KrVoKjy1xMW1"
      },
      "source": [
        "print(x.params)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b2IbNzKNxN-Q"
      },
      "source": [
        "import jax\n",
        "f = jax.grad(pure_predict_fn)\n",
        "g = jax.grad(manually_pure_predict_fn)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XvPww1CRxune",
        "outputId": "46e3be83-ff4c-45a9-f394-21c0ad8477f2"
      },
      "source": [
        "print(f(*args))\n",
        "print(g(*args))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'b': array(1., dtype=float32), 'w': DeviceArray(42., dtype=float32)}\n",
            "{'b': array(1., dtype=float32), 'w': DeviceArray(42., dtype=float32)}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C5GX4RZF48ya"
      },
      "source": [
        "# ~~Storing~~ Initializing Parameters Inside Class"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lKh1iqgb79j4"
      },
      "source": [
        "* ```method``` is the user defined impure functions that accesses parameters, assuming they are members of the class.\n",
        "\n",
        "* Whenever ```initial_params()``` function will be called, with a ```method```, and the ```method```'s arguments, every time ```method``` tries to access a ```param```, using ```self.get_param```, the parameter will be initialized. The ```method``` can send in the shape argument as the shape of result of some of its previous computation or one of its ```args```'s shape. The initializers can be specified beforehand or as an argument. So all the parameters will be initialized as we go on doing calculations and calling ```get_params()``` inside  ```method()``` [**which is running inside ```initial_params()```**].\n",
        "\n",
        "* The ```purify_method``` function, returns a pure version of ```method``` that takes in ```params``` as its first argument, always. These are the same ```params``` which were returned by ```initial_params()```.\n",
        "\n",
        "* Notice, how in this case we can't replace all ```self.get_param()``` with simple ```self.param[]``` in whatever ```method``` maybe, because we want our class to handle all initialization, retrieval etc."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ysDKIWZ171C7"
      },
      "source": [
        "class FlatFancyNNModule():\n",
        "    def __init__(self, name):\n",
        "        self.name = name\n",
        "        self.params = None\n",
        "        self.are_we_pure = False\n",
        "        self.are_we_initializing = False\n",
        "\n",
        "    # This is where, depending on mode, we'll give the user-written\n",
        "    # impure function different things!\n",
        "    def get_param(self, name, shape, initializer):\n",
        "        if self.are_we_initializing:\n",
        "            if name not in self.params:\n",
        "                self.params[name] = initializer(shape)\n",
        "            return self.params[name]\n",
        "        elif self.are_we_pure:\n",
        "            return self.params[name]\n",
        "        else:\n",
        "            raise Exception(\"Can't access parameters outside of context!\")\n",
        "\n",
        "    # This function calls the initializers to give us an initial params dict.    \n",
        "    def initial_params(self, method, *args):\n",
        "        self.are_we_initializing = True\n",
        "        self.params = {}\n",
        "        method(*args)\n",
        "        self.are_we_initializing = False\n",
        "        x = self.params\n",
        "        self.params = {}\n",
        "        return x\n",
        "\n",
        "    # This function returns a pure version of the function.\n",
        "    def purify_method(self, method):\n",
        "        def pure_method(params, *args):\n",
        "            self.are_we_pure = True\n",
        "            self.params = params\n",
        "            result = method(*args)\n",
        "            self.are_we_pure = False\n",
        "            self.params = None\n",
        "            return result\n",
        "        return pure_method"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OR-CZZ-HAviL"
      },
      "source": [
        "In the below code, \n",
        "* The third argument to ```self.get_param``` is the initializer function.\n",
        "\n",
        "* Initializer functions usually take in shape and rng key, but here, since our base class only has functionality for shape, we send in a function that only takes ```shape``` as argument.\n",
        "\n",
        "* The initializers will assign 13 to ```w``` and 0 to ```b``` , irrespective of the shape of input. But their values as well as shapes can be made to depend on ```x``` 's value/shape. \n",
        "\n",
        "* Below, ```obj``` refers to an object of class ```FancyLinearRegressor```.\n",
        "\n",
        "* Note here that both the calls ```obj.initial_params(predict, x)``` and ```obj.initial_params(rms,xs,ys)``` will initialise parameters correctly. Similarly, both the functions can be purified. \n",
        "\n",
        "* Outside the class, calls will never be made to these function(```predict``` and ```rms```) directly. They will be made to ```obj.initial_params()```, ```obj.purify_method()``` and to the purified method returned by ```obj.initial_params()```.\n",
        "\n",
        "* ```obj.initial_params()``` will return ```params``` that must be stored in a variable outside class which must be used when calling the purified versions of ```rms()``` or ```predict()``` . \n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HdtK_MhaAFY4"
      },
      "source": [
        "class FancyLinearRegressor(FlatFancyNNModule):\n",
        "    def __init__(self, name=\"linreg\"):\n",
        "        super().__init__(name=name)\n",
        "    \n",
        "    def predict(self, x):\n",
        "        # Our \"initializers\" are a bit simple here.\n",
        "        w = self.get_param('w', (0,), lambda _shape: 13.)                       \n",
        "        b = self.get_param('b', (0,), lambda _shape: 0.)\n",
        "        return w * x + b\n",
        "\n",
        "    def rms(self, xs, ys):\n",
        "        # Same as before.\n",
        "        w = self.get_param('w', (0,), lambda _shape: 13.)\n",
        "        b = self.get_param('b', (0,), lambda _shape: 0.)\n",
        "        return jnp.sqrt(jnp.sum(jnp.square(w * xs + b - ys)))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PyJP0ZNJUnlN"
      },
      "source": [
        "my_regressor = FancyLinearRegressor()\n",
        "\n",
        "sample_input = 999.  # a placeholder, necessary so the method can be run completely\n",
        "params = my_regressor.initial_params(my_regressor.predict, sample_input)\n",
        "\n",
        "print(\"Params:\", params)\n",
        "\n",
        "pure_predict_fn = my_regressor.purify_method(my_regressor.predict)\n",
        "\n",
        "pure_predict_fn(params, 42.)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aBMQ2IHIU8zL"
      },
      "source": [
        "def update_combiner(param, grad, lr=0.002):\n",
        "    return param - lr * grad"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zVcyLNGxUvlU"
      },
      "source": [
        "pure_rms_fn = my_regressor.purify_method(my_regressor.rms)\n",
        "grad_rms_fn = jax.grad(pure_rms_fn)\n",
        "\n",
        "params = my_regressor.initial_params(my_regressor.predict, sample_input)\n",
        "\n",
        "for _ in range(15):\n",
        "    print(pure_rms_fn(params, xs, ys))\n",
        "    grads = grad_rms_fn(params, xs, ys)\n",
        "    params = jax.tree_multimap(update_combiner, params, grads)\n",
        "\n",
        "pure_predict_fn(params, 42.)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rjXrl6Q7VX3H"
      },
      "source": [
        "Suppose now, we now want to make a Neural Net, by composing together ```LinearRegressor```s. In the NN class, we will define various ```LinearRegressor```. \n",
        "\n",
        "\\\\\n",
        "**The parameters of these objects can be initialised only when ```initial_params``` of these objects called.** \n",
        "\n",
        "\\\\\n",
        "Suppose the ```predict``` function of NN, directly calls predict function of the ```LinearRegressor```s defined inside NN. \n",
        "\n",
        "\\\\\n",
        "So, now, when we try to run ```NN_obj.initial_params(NN_obj.predict, sample_input)```, it will call ```NN_obj.predict(sample_input)``` which will, in turn call ```LinearRegressor.predict(sample_input)``` which will in turn call ```get_param()``` which will raise an exception because, although ```are_we_initializing``` of the Neural Network object is ```True```, it is ```False``` for the ```LinearRegressor```. \n",
        "\n",
        "\\\\\n",
        "So, the correct way to proceed would be to initilaize and purify all submodules first,  store there parameters with yourself , then call your your own ```initial_param``` , and use only the purified versions of the member functions of submodules in your NN class. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vt5FzLA5bTOj"
      },
      "source": [
        "## Modifying NNModule to handle nested submodules"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kjZSXSrbbKU1"
      },
      "source": [
        "class FancyNNModule():\n",
        "    def __init__(self, name):\n",
        "        self.name = name\n",
        "        self.params = None\n",
        "        self.are_we_pure = False\n",
        "        self.are_we_initializing = False\n",
        "        # Implementations should register their submodules here!\n",
        "        self.submodules = {}\n",
        "\n",
        "    # This is where, depending on mode, we'll give the user-written\n",
        "    # impure function different things!\n",
        "    def get_param(self, name, shape, initializer):\n",
        "        if self.are_we_initializing:\n",
        "            if name not in self.params:\n",
        "                self.params[name] = initializer(shape)\n",
        "            return self.params[name]\n",
        "        elif self.are_we_pure:\n",
        "            return self.params[name]\n",
        "        else:\n",
        "            raise Exception(\"Can't access parameters outside of context!\")\n",
        "\n",
        "    def set_are_we_pure(self, ispure):\n",
        "        if ispure: self.params = {}\n",
        "        self.are_we_pure = ispure\n",
        "        for sm in self.submodules.values():\n",
        "            sm.set_are_we_pure(ispure)\n",
        "        if not ispure: self.params = None\n",
        "\n",
        "    def set_are_we_initializing(self, isinit):\n",
        "        if isinit:\n",
        "            self.params = {}\n",
        "        self.are_we_initializing = isinit\n",
        "        for sm in self.submodules.values():\n",
        "            sm.set_are_we_initializing(isinit)\n",
        "\n",
        "    # This method gathers params from this module and all submodules.\n",
        "    def gather_params(self):\n",
        "        params = self.params\n",
        "        for sm_name, sm in self.submodules.items():\n",
        "            for name, value in sm.gather_params().items():\n",
        "                params[sm_name + \"/\" + name] = value\n",
        "        return params\n",
        "\n",
        "    # This method spreads out params into self and all submodules.\n",
        "    def spread_params(self, params):\n",
        "        for name, value in params.items():\n",
        "            path = name.split(\"/\")\n",
        "            if len(path) == 1:\n",
        "                self.params[name] = value\n",
        "            else:\n",
        "                self.submodules[path[0]].spread_params({\"/\".join(path[1:]): value})\n",
        "\n",
        "    # This function calls the initializers to give us an initial params dict.    \n",
        "    def initial_params(self, method, *args):\n",
        "        self.set_are_we_initializing(True)                                      \n",
        "        method(*args)\n",
        "        self.set_are_we_initializing(False)\n",
        "        params = self.gather_params()\n",
        "        return params\n",
        "\n",
        "    # This function returns a pure version of the function.\n",
        "    def purify_method(self, method):\n",
        "        def pure_method(params, *args):\n",
        "            self.set_are_we_pure(True)\n",
        "            self.spread_params(params)\n",
        "            result = method(*args)\n",
        "            self.set_are_we_pure(False)\n",
        "            return result\n",
        "        return pure_method"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mESr8wWM5HFw"
      },
      "source": [
        "We will understand the code above and below together. Note that each module is now required to maintain a dictionary of its immediate submodules(see line 38 below).\n",
        "\n",
        "**Initialization Steps(Line 54-59 above)**\n",
        "\n",
        "* Set ```are_we_initializing=True``` for the module, and all its submodules recursively.\n",
        "\n",
        "* Call the ```method``` supplied in ```initial_params()``` using the args given. \n",
        "\n",
        "* Each call to ```get_param``` directly in the module(e.g., in ```FancyLSTMLM```) will lead to new parameters being assigned to the module's parent(```FancyNNModule```) class' ```params``` attribute.\n",
        "\n",
        "* Each call to a submodule(e.g. to ```self.cell()``` in ```FancyLSTMLM```), will lead to a call to a member function of the submodule; and the ```get_param()``` calls within it will assign new parameters to that object's ```params``` attribute(i.e., to ```self.cell.params``` in the below example). \n",
        "\n",
        "* Then we reset ```are_we_initialising``` of all submodules recursively and gather all parameters of all the nested modules as done in ```gather_params()``` and return the gathered parameters. \n",
        "\n",
        "* Each module's ```self.params``` attribute consists of a dictionary whose keys are their direct parameters and values are the values of those parameters.\n",
        "\n",
        "* ```gather_params()``` gathers together all those ```self.params``` into a ```params``` dict. The keys of this dicts are the paths(e.g. ```lstmcell/weight_ih``` would be a key in the ```param``` when calling ```gather_params()``` in ```FancyLSMTLM``` ) to each parameter in each of the submodules(nested as well as direct) and values are the values given to them after initialization. \n",
        "\n",
        "Similarly you can understand the ```purify_method()``` function's working. Observe how both ```spread_params()``` and ```get_params()``` need to recursively call **themselves** only. \n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qt_AAbtu6T62"
      },
      "source": [
        "random_key = jax.random.PRNGKey(0)\n",
        "\n",
        "def unif_initializer(shape):\n",
        "    global random_key\n",
        "    sample_key, random_key = jax.random.split(random_key)\n",
        "    return jax.random.uniform(sample_key, shape)\n",
        "\n",
        "class FancyLSTMCell(FancyNNModule):\n",
        "    def __init__(self, hid_dim, name=\"lstmcell\"):\n",
        "        super().__init__(name=name)\n",
        "        self.hid_dim = hid_dim\n",
        "\n",
        "    def __call__(self, inputs, h, c):\n",
        "        weight_ih = self.get_param(\"weight_ih\",\n",
        "                (4*self.hid_dim, self.hid_dim), unif_initializer)\n",
        "        weight_hh = self.get_param(\"weight_hh\",\n",
        "                (4*self.hid_dim, self.hid_dim), unif_initializer)\n",
        "        bias = self.get_param(\"bias\",\n",
        "                (4*self.hid_dim,), lambda shape: jnp.zeros(shape))\n",
        "\n",
        "        ifgo = weight_ih @ inputs + weight_hh @ h + bias\n",
        "        i, f, g, o = jnp.split(ifgo, indices_or_sections=4, axis=-1)\n",
        "        i = jax.nn.sigmoid(i)\n",
        "        f = jax.nn.sigmoid(f)\n",
        "        g = jnp.tanh(g)\n",
        "        o = jax.nn.sigmoid(o)\n",
        "        new_c = f * c + i * g\n",
        "        new_h = o * jnp.tanh(new_c)\n",
        "        return (new_h, new_c)\n",
        "\n",
        "class FancyLSTMLM(FancyNNModule):\n",
        "    def __init__(self, vocab_size, dim, name=\"lstmlm\"):\n",
        "        super().__init__(name=name)\n",
        "        self.vocab_size = vocab_size\n",
        "        self.dim = dim\n",
        "        # Now create a submodule and register it!\n",
        "        self.cell = FancyLSTMCell(dim)\n",
        "        self.submodules[self.cell.name] = self.cell\n",
        "    \n",
        "    @property\n",
        "    def hc_0(self):\n",
        "        _c_0 = self.get_param(\"c_0\",\n",
        "                (self.dim,), lambda shape: jnp.zeros(shape))\n",
        "        return (jnp.tanh(_c_0), _c_0)\n",
        "\n",
        "    def forward(self, seq, hc):\n",
        "        loss = 0.\n",
        "        embeddings = self.get_param(\"embeddings\",\n",
        "                (self.vocab_size, self.dim), unif_initializer)\n",
        "        for idx in seq:\n",
        "            loss -= jax.nn.log_softmax(embeddings @ hc[0])[idx]\n",
        "            hc = self.cell(embeddings[idx,:], *hc)\n",
        "        return loss, hc\n",
        "\n",
        "    def greedy_argmax(self, hc, length=6):\n",
        "        idxs = []\n",
        "        embeddings = self.get_param(\"embeddings\",\n",
        "                (self.vocab_size, self.dim), unif_initializer)\n",
        "        for i in range(length):\n",
        "            idx = jnp.argmax(embeddings @ hc[0])\n",
        "            idxs.append(int(idx))\n",
        "            hc = self.cell(embeddings[idx,:], *hc)\n",
        "return idxs"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LDZQ_LpF6n0M"
      },
      "source": [
        "lm = FancyLSTMLM(vocab_size, 17)\n",
        "\n",
        "# Since jitting and our mutable-state-galore \"tracer\" don't play nice, only\n",
        "# call jit on the outermost level: the actually pure function!\n",
        "pure_sample_fn = lm.purify_method(\n",
        "    lambda: lm.greedy_argmax(lm.hc_0))\n",
        "pure_forward_fn = jax.jit(lm.purify_method(\n",
        "    lambda seq, hc: lm.forward(seq, hc if hc else lm.hc_0)))\n",
        "grad_loss_fn = jax.jit(jax.grad(lm.purify_method(\n",
        "    lambda seq, hc: lm.forward(seq, hc if hc else lm.hc_0)[0])))\n",
        "\n",
        "params = lm.initial_params(lambda: lm.forward(jnp.array([0]), lm.hc_0))\n",
        "print(\"All parameters, recursively found:\", list(params.keys()))\n",
        "\n",
        "print(\"Sample before:\", pure_sample_fn(params))\n",
        "\n",
        "bptt_length = 3\n",
        "for epoch in range(101):\n",
        "    totalloss = 0.\n",
        "    hc = None\n",
        "    for start in range(0, len(training_data), bptt_length):\n",
        "        batch = training_data[start:start+bptt_length]\n",
        "        loss, new_hc = pure_forward_fn(params, batch, hc)\n",
        "        if epoch % 50 == 0:\n",
        "            totalloss += loss.item()\n",
        "        grads = grad_loss_fn(params, batch, hc)\n",
        "        params = jax.tree_multimap(update_combiner_01, params, grads)\n",
        "        hc = new_hc\n",
        "    if totalloss:\n",
        "        print(\"Loss:\", totalloss)\n",
        "\n",
        "print(\"Sample after:\", pure_sample_fn(params))"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}