import utils
import utils_blocks as ub
import theano
from theano import tensor as tsr
from blocks import algorithms, main_loop
import blocks.extensions as be
import blocks.extensions.monitoring as bm
import theano.tensor.shared_randomstreams
import numpy as np


class ClippedParams(algorithms.StepRule):
    """A rule to maintain parameters within a specified range"""
    def __init__(self, min_value, max_value):
        self.min_value = min_value
        self.max_value = max_value

    def compute_step(self, parameter, previous_step):
        min_clipped = tsr.switch(parameter - previous_step < self.min_value,
                                 0, previous_step)
        return tsr.switch(parameter - previous_step > self.max_value,
                          0, min_clipped), []


def create_vi_model(n_nodes, n_samp=100):
    """return variational inference theano computation graph"""
    def aux(a, b):
        rand = a + b * np.random.normal(size=(n_nodes, n_nodes))
        return np.clip(rand, 1e-3, 1 - 1e-3).astype(theano.config.floatX)

    x = tsr.matrix(name='x', dtype='int8')
    s = tsr.matrix(name='s', dtype='int8')
    mu = theano.shared(value=aux(.5, .1), name='mu1')
    sig = theano.shared(value=aux(.5, .1), name='sig1')
    mu0 = theano.shared(value=aux(.5, .1), name='mu0')
    sig0 = theano.shared(value=aux(.5, .1), name='sig0')

    srng = tsr.shared_randomstreams.RandomStreams(seed=123)
    theta = (srng.normal((n_samp, n_nodes, n_nodes)) * sig[None, :, :]
             + mu[None, :, :])
    y = tsr.maximum(tsr.dot(x, theta), 1e-3)
    infect = tsr.log(1. - tsr.exp(-y[0:-1])).dimshuffle(1, 0, 2)
    lkl_pos = tsr.sum(infect * (x[1:] & s[1:])) / n_samp
    lkl_neg = tsr.sum(-y[0:-1].dimshuffle(1, 0, 2) * (~x[1:] & s[1:])) / n_samp
    lkl = lkl_pos + lkl_neg
    kl = tsr.sum(tsr.log(sig0 / sig) + (sig**2 + (mu0 - mu)**2)/(2*sig0)**2)
    cost = - lkl + kl
    cost.name = 'cost'

    return x, s, mu, sig, cost


if __name__ == "__main__":
    batch_size = 100
    frequency = 10
    n_samples = 50
    graph = utils.create_random_graph(n_nodes=10)
    print('GRAPH:\n', graph, '\n-------------\n')

    x, s, mu, sig, cost = create_vi_model(len(graph), n_samples)
    rmse = ub.rmse_error(graph, mu)

    step_rules = algorithms.CompositeRule([algorithms.AdaDelta(),
                                           ClippedParams(1e-3, 1 - 1e-3)])

    alg = algorithms.GradientDescent(cost=cost, parameters=[mu, sig],
                                     step_rule=step_rules)
    data_stream = ub.dynamic_data_stream(graph, batch_size)
    loop = main_loop.MainLoop(
        alg, data_stream,
        log_backend="sqlite",
        extensions=[
            be.FinishAfter(after_n_batches=10**4),
            bm.TrainingDataMonitoring([cost, rmse, mu], every_n_batches=frequency),
            be.Printing(every_n_batches=frequency, after_epoch=False),
            ub.JSONDump("logs/tmp.json", every_n_batches=10),
            #ub.ActiveLearning(dataset=data_stream.dataset, params=graph)
        ]
    )
    loop.run()