def main():
S = np.linspace(0, 200, 200 * 8 + 1)
Sl = 0
Su = 250
N = 128 * T
K = 8 * (Su - Sl)
Sk = (K * (S - Sl) / (Su - Sl)).astype(int)
legend = []
label = "$\\Omega^v = [%i, 5]$"
model = FDEModel(N, dS, payoff)
fig = plt.figure()
fig.set_figwidth(1.8 * fig.get_figwidth())
ax = fig.add_subplot(1, 2, 1)
for i in range(1, 5):
Conv.times = [(i, 5)]
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
legend.append(label % i)
plt.xlabel("Stock Price")
plt.ylabel("Convertible Bond Price")
plt.legend(legend)
ax = fig.add_subplot(1, 2, 2, projection="3d")
plot_model(ax, dS, payoff)
def main():
S = np.linspace(0, 200, 200 * 8 + 1)
Sl = 0
Su = 250
N = 128 * T
K = 8 * (Su - Sl)
Sk = (K * (S - Sl) / (Su - Sl)).astype(int)
legend = []
label = "$\\Omega^v = [%i, 5]$"
model = FDEModel(N, dS, payoff)
fig = plt.figure()
fig.set_figwidth(1.8 * fig.get_figwidth())
ax = fig.add_subplot(1, 2, 1)
for i in range(1, 5):
Conv.times = [(i, 5)]
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
legend.append(label % i)
plt.xlabel("Stock Price")
plt.ylabel("Convertible Bond Price")
plt.legend(legend)
ax = fig.add_subplot(1, 2, 2, projection="3d")
plot_model(ax, dS, payoff)
def analyze(data, neuron, args=None, confs=None):
if args is None:
args = DEFAULT_ARGS
if confs is None:
confs = DEFAULT_CONFS
firing_rates = transform_spikes(neuron, filter_width=50)
data, neuron, firing_rates = remove_nans(data, neuron, firing_rates)
create_model = lambda: Model(PoissonRegressor(), spikes=neuron, n_folds=10)
create_model = lambda: Model(LinearRegression(), spikes=neuron, n_folds=10)
best_model = create_model()
subset = data.columns.to_list() # starting with all columns
data_ = transform_data(data[subset], args['bins'])
best_model(data_, firing_rates)
plot_model(data_, neuron, firing_rates, best_model, subset)
# return
bins = args['bins']
# naive estimation of SHAP values
# a real calculation will require 2^k (k=no. features) models, and to avg them with different weights (N choose k)
features_to_remove = get_one_dim_feature_names(
) + get_two_dim_feature_names()
for feature_to_remove in features_to_remove:
if isinstance(feature_to_remove, str): # 1D feature
feature_to_remove = [feature_to_remove]
model = create_model()
new_subset = [col for col in subset if col not in feature_to_remove]
print(new_subset, feature_to_remove)
new_bins = bins.copy()
[new_bins.pop(x) for x in feature_to_remove]
data_ = transform_data(data[new_subset], new_bins)
print(data_.shape)
model(data_, firing_rates)
if model > best_model:
subset = new_subset
best_model = model
bins = new_bins
get_avg_ll = lambda m: np.mean(
[x.results.likelihood for x in m.CVfolds])
print(feature_to_remove, get_avg_ll(model))
# shuffles_results = run_shuffles(best_model)
plot_model(data_, neuron, firing_rates, model, subset)
def main():
S = np.linspace(0, 200, 200 * 8 + 1)
Sl = 0
Su = 250
N = 128 * T
K = 8 * (Su - Sl)
Sk = (K * (S - Sl) / (Su - Sl)).astype(int)
model = FDEModel(N, dS, payoff)
fig = plt.figure()
fig.set_figwidth(1.8 * fig.get_figwidth())
ax = fig.add_subplot(1, 2, 1)
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
C.times = [2]
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
plt.xlabel("Stock Price")
plt.ylabel("Convertible Bond Price")
plt.legend(["$\\Omega^c = [2, 5)$", "$\\Omega^c = \\{2\\}$"])
ax = fig.add_subplot(1, 2, 2, projection="3d")
plot_model(ax, dS, payoff)
def main():
S = np.linspace(0, 200, 200 * 8 + 1)
Sl = 0
Su = 250
N = 128 * T
K = 8 * (Su - Sl)
Sk = (K * (S - Sl) / (Su - Sl)).astype(int)
model = FDEModel(N, dS, payoff)
fig = plt.figure()
fig.set_figwidth(1.8 * fig.get_figwidth())
ax = fig.add_subplot(1, 2, 1)
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
A.N = -A.C
ax.plot(S, model.price(Sl, Su, K).V[0][Sk])
plt.xlabel("Stock Price")
plt.ylabel("Convertible Bond Price")
plt.legend(["$R = 104$", "$R = 0$"])
ax = fig.add_subplot(1, 2, 2, projection="3d")
plot_model(ax, dS, payoff)
def main():
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")
plot_model(ax, dS, payoff)
def main():
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")
plot_model(ax, dS, payoff)
avg_reward_hist.append(r_total/t)
model_log.append(agent.model)
return avg_reward_hist, agent.model_hist
np.random.seed(10)
number_arms = 20
testbed = Arms(number_arms)
iters_test = 10000
print(testbed)
print('\nMODELS OF VARIOUS AGENTS:')
agent1 = Agent(number_arms, epsilon=0, Q_0=0) # -1 epsilon will never explore
reward1, model1 = test_agent(testbed, agent1, iters_test)
plot_model(model1, iters_test, testbed.arms, name="agent1")
agent2 = Agent(number_arms, epsilon=0.01, Q_0=0)
reward2, model2 = test_agent(testbed, agent2, iters_test)
plot_model(model2, iters_test, testbed.arms, name="agent2")
agent3 = Agent(number_arms, epsilon=0.1, Q_0=0)
reward3, model3 = test_agent(testbed, agent3, iters_test)
plot_model(model3, iters_test, testbed.arms, name="agent3")
agent4 = Agent(number_arms, epsilon=1)
reward4, model4 = test_agent(testbed, agent4, iters_test)
plot_model(model4, iters_test, testbed.arms, name="agent4")
compare_avg_performance([agent1, agent2, agent3, agent4], iters_test,
[reward1, reward2, reward3, reward4],
