Python statistics.variance函数代码示例

def run(data):
    f = open("analyzer.log", 'a+')
    c = costs(data)
    total = total_cost(data)
    f.write("\n############# COST #############\n")
    f.write("Total Cost : {0}\n".format(total))
    f.write("Total Cost Mean: {0}\n".format(mean(c)))
    f.write("Total Cost Median: {0}\n".format(median(c)))
    f.write("Total Cost Mode: {0}\n".format(mode(c)))
    f.write("Total Cost Variance: {0}\n".format(variance(c)))

    cost_action = action(data)
    f.write("Cost by Action: \n")
    for k, v in cost_action.iteritems():
        f.write("\t{0} -> {1} units\n".format(k, v))

    f.write("Percentage Cost by Action: \n")
    for k, v in cost_action.iteritems():
        f.write("\t{0} -> {1} %\n".format(k, round(((v * 100.) / total), 2)))

    f.write("Cost Variance by Action: \n")
    for k, v in cost_action.iteritems():
        c_action = costs_action(data, k)
        if len(c_action) > 1:
            f.write("\t{0} -> {1} units\n".format(k, round(variance(c_action), 2)))
        else:
            f.write("\t{0} -> {1} units\n".format(k, round(c_action[0], 2)))

    key_max, max_value = max_action_value(cost_action)
    f.write("More Expensive Action by value: {0} -> {1}\n".format(key_max[0], cost_action.get(key_max[0])))

    key_max, max_value = max_action_percentage(cost_action, total)
    f.write("More Expensive Action by percentage: {0} -> {1} %\n".format(key_max, round(max_value, 2)))

    f.close()

def model_analysis(x, x_matrix, y, line, y_hat, b):
    n = len(x) # number of samples
    s_x = stats.stdev(x) # standard deviation of x values
    s_y = stats.stdev(y) # standard deviation of y values
    s2_x = stats.variance(x) # variance of x values
    s2_y = stats.variance(y) # variance of y values
    s_xy = b * s2_x # covariance of VM
    
    mad_temp = 0
    SSE = 0
    for i in range(len(y)):
        temp = abs(y[i] - y_hat[i])
        mad_temp += temp
        SSE += temp**2 # sum of squares for error
    MAD = mad_temp / n    
    s_err = math.sqrt(SSE / (n - 2)) # standard error of estimate
    s_b = s_err / math.sqrt((n - 1) * s2_x)
    
    r = s_xy / (s_x * s_y) # sample coefficient of correlation
    R_2 = line.score(x_matrix, y) # coefficient of determination 
    R_2calc = s_xy**2 / (s2_x * s2_y)
    t = b / s_b # t-value for slope assuming true slope = 0
    
    f1.write('\nSkew = ' + str(b) + '\n')
    f1.write('Coefficient of correlation (r) = ' + str(r) + '\n')
    #f1.write('Coefficient of determination (R^2) via scikit = ' + str(R_2) + '\n')
    f1.write('Coefficient of determination (R^2) calculate = ' + str(R_2calc) + '\n')
    f1.write('Test statistic for clock skew (t) = ' + str(t) + '\n')
    f1.write('Mean Absolute Deviation (MAD) = ' + str(MAD) + '\n')
    f1.write('Sum of Squares for Forecast Error (SSE) = ' + str(SSE) + '\n')
    
    return

def main():

    statistics_file = open("statics.txt", "w")

    stats = {}

    final_statistics = {}

    entropy_kills = []
    entropy_dies = []

    initialize_server()

    stats = simulate_population()

    print(statistics)

    for key, val in stats.items() :
        entropy_kills += val[0]
        entropy_dies += val[1]

    print(str(entropy_kills))
    statistics_file.write(str(entropy_kills) + "\n")

    print(str(entropy_dies))
    statistics_file.write(str(entropy_dies) + "\n")


    final_statistics.update({"mean entropy kill " : statistics.mean(entropy_kills)})
    final_statistics.update({"variance entropy kill " : statistics.variance(entropy_kills)})
    final_statistics.update({"mean entropy_dies " : statistics.mean(entropy_dies) })
    final_statistics.update({"variance entropy_dies " : statistics.variance(entropy_dies) })

    for key, val in final_statistics.items():
        print(str(key) + " : " + str(val))
        statistics_file.write(str(key) + " : " + str(val) + "\n")

    plt.figure(1)

    plt.subplot(211)

    plt.xlabel('id bot')

    plt.ylabel('kill')

    plt.boxplot(entropy_kills, labels=list('K'))

    plt.subplot(212)

    plt.xlabel('id bot')

    plt.ylabel('dies')

    plt.boxplot(entropy_dies, labels=list('D'))

    plt.show()

    statistics_file.close()

def countXYZ(data):
	x=[]
	y=[]
	z=[]
	for tmp in data:
		x.append(float(int(tmp[0:3],16))*12.0/4096.0)
		y.append(float(int(tmp[3:6],16))*14.0/4096.0)
		z.append(float(int(tmp[6:9],16))*36.0/4096.0)
	return [6.0-sum(x)/len(x),7.0-sum(y)/len(y),sum(z)/len(z),\
	statistics.variance(x),statistics.variance(y),statistics.variance(z)]

def caculation(data):
    trt_1 = data[data['trt'] == 1]
    trt_0 = data[data['trt'] == 0]
    
    medi = statistics.median(trt_1['y']) - statistics.median(trt_0['y'])
    mean = statistics.mean(trt_1['y']) - statistics.mean(trt_0['y'])
    peop = len(trt_1) + len(trt_0)
    vari = statistics.variance(trt_1['y']) + statistics.variance(trt_0['y'])
    z_stat, p_val = stats.ranksums(trt_0['y'], trt_1['y']) 
    return [medi, mean, peop, p_val]

    def adjRating(self, ratings, VERBOSE=False):
        new_ratings = {}
        change = False
        max_change = 0.0
        for team in self.teams:
            # Set up arrays for ORating and DRating
            ODiff = []
            DDiff = []
            ct = self.teams[team]

            if VERBOSE:
                print("%s" % ct.name)
                print(" Home Games:")
            for game in ct.home_games:
                ODiff.append(game.hs + ratings[game.at.name][1])
                DDiff.append(game.aws - ratings[game.at.name][0])
                if VERBOSE:
                    print("  %s: "
                          "ODiff Entry=%.2f, "
                          "DDiff Entry=%.2f" % (game, ODiff[len(ODiff)-1],
                                              DDiff[len(DDiff)-1]))

            if VERBOSE:
                print(" Away Games:")
            for game in ct.away_games:
                ODiff.append(game.aws + ratings[game.ht.name][1])
                DDiff.append(game.hs - ratings[game.ht.name][0])
                if VERBOSE:
                    print("  %s: "
                          "ODiff Entry=%.2f, "
                          "DDiff Entry=%.2f" % (game, ODiff[len(ODiff)-1],
                                              DDiff[len(DDiff)-1]))

            temp_AdjO = (sum(ODiff)/float(len(ODiff))) - self.average
            temp_AdjD = self.average - (sum(DDiff)/float(len(DDiff)))
            ct.AOV = statistics.variance(ODiff)/len(ODiff)
            ct.ADV = statistics.variance(DDiff)/len(DDiff)
            new_ratings.update({ct.name: [temp_AdjO, temp_AdjD]})
            ch_th_O = abs((temp_AdjO-ratings[ct.name][0])/ratings[ct.name][0])
            ch_th_D = abs((temp_AdjD-ratings[ct.name][1])/ratings[ct.name][1])

            max_change = max(ch_th_D, ch_th_O, max_change)
            if ch_th_O > .0025 or ch_th_D > .0025:
                change = True
            if VERBOSE:
                print(" Team Stats")
                print("  AdjPointsScored: %s" % ODiff)
                print("  AdjPointsAllowed: %s" % DDiff)
                print("  AdjO=%.2f, AdjD=%.2f, "
                      "ChO=%.4f, ChD=%.4f" %
                      (temp_AdjO, temp_AdjD, ch_th_O, ch_th_D))
                print("  AOV=%.4f, ADV=%.4f" % (ct.AOV, ct.ADV))
        # print("%.4f" % max_change)
        return change, new_ratings

    def get_statistics(self):
        """
        Returns various statistics about the benchmark run.
        :return: See description.
        """
        ret_val = {}
        not_applicable = 'N/A'

        if len(self.rollovers) > 1:  # Skip the last rollover...it could've been smaller than chunk_size
            rollover_times = []
            last_time = self.started
            for i in xrange(len(self.rollovers) - 1):
                rollover_times.append(
                    (self.rollovers[i] - last_time).total_seconds()
                )
                last_time = self.rollovers[i]

            ret_val['rollover_mean'] = self.num_to_seconds(mean(rollover_times))
            ret_val['rollover_stdev'] = self.num_to_seconds(stdev(rollover_times))
            ret_val['rollover_variance'] = self.num_to_seconds(variance(rollover_times))
        else:
            ret_val['rollover_mean'] = not_applicable
            ret_val['rollover_stdev'] = not_applicable
            ret_val['rollover_variance'] = not_applicable

        if self.initial_mem_usage is None:
            ret_val['initial_mem_usage'] = not_applicable
        else:
            ret_val['initial_mem_usage'] = self.num_to_megabytes(self.initial_mem_usage)

        if len(self.resources) > 0:
            cpu_util_list = [x[1] for x in self.resources]
            ret_val['cpu_util_mean'] = self.num_to_percent(mean(cpu_util_list))
            ret_val['cpu_util_stdev'] = self.num_to_percent(stdev(cpu_util_list))
            if len(cpu_util_list) > 1:
                ret_val['cpu_util_variance'] = self.num_to_percent(variance(cpu_util_list))
            else:
                ret_val['cpu_util_variance'] = not_applicable

            mem_usage_list = [x[2] for x in self.resources]
            ret_val['mem_usage_mean'] = self.num_to_megabytes(mean(mem_usage_list))
        else:
            ret_val['cpu_util_mean'] = not_applicable
            ret_val['cpu_util_stdev'] = not_applicable
            ret_val['cpu_util_variance'] = not_applicable

            ret_val['mem_usage_mean'] = not_applicable

        return ret_val

def DemoStatistic(dataParam):

	#calculate mean
	mean = statistics.mean(dataParam);

	#calculate median
	median = statistics.median(dataParam);

	#calculate standard deviation
	stdv = statistics.variance(dataParam);

	#count values outside 3 sigma range
	noiseCount = 0;
	for value in dataParam:
		if value < (-3* stdv) + mean or value > (3*stdv) + mean:
			#print(" %.4f" %(value));
			noiseCount += 1;

			
	print("-----------------Simple------------------------");
	print("Data length: %d" %(len(dataParam)) );
	print("Values outside 3sigma: %d" %(noiseCount) );
	print("Mean: %.7f" %(mean));
	print("Median: %.7f"%(median));
	print("Standard deviation: %.7f"%(stdv));
	print("------------------------------------------------");

def analyze(graphs):
    """summary stats for the graphs:
    >>> graphs = [{'win': ['a', 'b', 'philosophy']}, {'win': ['c', 'd', 'philosophy']}, {'fail': ['e', 'f', 'g', 'h', 'i', 'j', 'k']}]
    >>> analyze(graphs)
    ... {'min': 2, 'max': 2, 'mean': 2.0, 'median': 2.0, 'var': 0.0}
    """
    win_path_lengths = []
    fail_path_lengths = []

    for graph in graphs:
        if graph.get('win'):
            win_path_lengths.append(len(graph['win']) - 1)
        if graph.get('fail'):
            fail_path_lengths.append(len(graph['fail']) - 1)

    #stats
    win_perc = sum(win_path_lengths)/sum([sum(win_path_lengths), sum(fail_path_lengths)])
    min_path_length = min(win_path_lengths)
    max_path_length = max(win_path_lengths)
    mean_path_length = mean(win_path_lengths)
    median_path_length = median(win_path_lengths)
    var_path_length = variance(win_path_lengths)

    print('Cache is enabled by default, turning it off will affect the distributions')
    print('Percentage of pages leading to Philosophy: {}'.format(win_perc))
    print('Distribution of paths leading to Philosophy: min {}, max {}, mean {}, median {}, var {}'.format(
           min_path_length, max_path_length, mean_path_length, median_path_length, var_path_length))

    return dict(min=min_path_length,
                max=max_path_length,
                mean=mean_path_length,
                median=median_path_length,
                var=var_path_length)

def discard_spurious_lines(lines, expected):
    """ Discards the discordant line(s).

    Discards the line or lines that minimizes the variance
    of the distances between the set of the other lines
    (of size expected).

    The hypothesis is that, when more lines than expected
    were detected, the variance of the distances bewteen
    lines will be minimized when the spurious lines are
    out of the set of selected lines.

    Be aware of the number of possible combinations before
    calling this function. Having just one or two extra lines
    should generally be ok.

    """
    best_variance = math.inf
    for combination in itertools.combinations(lines, expected):
        diffs = [b[0] - a[0]
                 for a, b in zip(combination[:-1], combination[1:])]
        variance = statistics.variance(diffs)
        if variance < best_variance:
            best_combination = combination
            best_variance = variance
    return best_combination

def makeCharactOutOfEndDifferencesList(listdiffs, type=STAT_MEAN):
    """
    type=0, mean
    1, variance
    2, sigma
    """
    import copy
    import statistics
    import scipy.stats as scps

    lst = copy.deepcopy(listdiffs)

    func = {
        -1: statistics.median,
        0: statistics.mean,
        1: statistics.variance,
        2: lambda x: math.sqrt(statistics.variance(x)),
        3: scps.skewtest,
        4: scps.kurtosistest,
    }

    for j in range(len(lst)):
        for k in range(len(lst[j])):
            lst[j][k] = func[type](lst[j][k])

    return lst

 def compute(self, ydata):
     y = ydata.T
     var_y = statistics.variance(y)
     self.summary = self.summary+var_y
     self.var_y.append(var_y)
     return { 'summary': self.summary,
              'var_y': self.var_y }

    def async_update(self):
        """Get the latest data and updates the states."""
        if self._max_age is not None:
            self._purge_old()

        if not self.is_binary:
            try:  # require only one data point
                self.mean = round(statistics.mean(self.states), 2)
                self.median = round(statistics.median(self.states), 2)
            except statistics.StatisticsError as err:
                _LOGGER.error(err)
                self.mean = self.median = STATE_UNKNOWN

            try:  # require at least two data points
                self.stdev = round(statistics.stdev(self.states), 2)
                self.variance = round(statistics.variance(self.states), 2)
            except statistics.StatisticsError as err:
                _LOGGER.error(err)
                self.stdev = self.variance = STATE_UNKNOWN

            if self.states:
                self.count = len(self.states)
                self.total = round(sum(self.states), 2)
                self.min = min(self.states)
                self.max = max(self.states)
                self.change = self.states[-1] - self.states[0]
                self.average_change = self.change
                if len(self.states) > 1:
                    self.average_change /= len(self.states) - 1
                if self._max_age is not None:
                    self.max_age = max(self.ages)
                    self.min_age = min(self.ages)
            else:
                self.min = self.max = self.total = STATE_UNKNOWN
                self.average_change = self.change = STATE_UNKNOWN

def getVariances(optimalParameters):
    params = optimalParameters[0].keys()
    variances = {}
    for parameter in params:
        allValues = [x[parameter] for x in optimalParameters]
        variances[parameter] = statistics.variance(allValues)
    return variances

 def get(self):
     values = sorted(self.reservoir.values)
     count = len(values)
     # instead of failing return empty / subset so that json2insert & co
     # don't fail
     if count == 0:
         return dict(n=0)
     elif count == 1:
         return dict(min=values[0], max=values[0], mean=values[0], n=count)
     percentiles = [percentile(values, p) for p in self.plevels]
     min_ = values[0]
     max_ = values[-1]
     stdev = statistics.stdev(values)
     return dict(
         min=min_,
         max=max_,
         mean=statistics.mean(values),
         median=statistics.median(values),
         variance=statistics.variance(values),
         error_margin=error_margin(95, stdev, self.reservoir.count),
         stdev=stdev,
         # replace . with _ so that the output can be inserted into crate
         # crate doesn't allow dots in column names
         percentile={str(i[0]).replace('.', '_'): i[1] for i in
                     zip(self.plevels, percentiles)},
         n=self.reservoir.count,
         samples=self.reservoir.values
     )