def component4d():
print("-----Component 4d (Check results folder)-----")
model1 = MarcovModel("franz_kafka_collected_works.txt", "word", 3)
model2 = MarcovModel("practice_phrenology_simplified.txt", "word", 3)
model3 = MarcovModel("william_shakespeare_collected_works.txt", "word", 3)
models = [model1, model2, model3]
f = open("results/component4d.txt", "a")
length_of_test_input = 60
for model in models:
f.write(f"\n\n-----Estimation Results for {model.corpus_filename}-----\n")
for estimator_model in models:
f.write(f"\n-----When estimated by: {estimator_model.corpus_filename}-----\n")
estimates = []
text = ""
nextPos = 0
while nextPos < (len(model.testset) - length_of_test_input):
text = model.testset[nextPos:nextPos + length_of_test_input]
estimate = estimator_model.estimate(text)
# print(estimate)
estimates.append(estimate)
text = ""
nextPos += length_of_test_input
f.write(f"{statistics.mean(estimates)}")
def component4f():
# intentionally chose Franz Kafka and Edgar Allan Poe because we thought they were similar
print("-----Component 4f (Check results folder)-----")
f = open("results/component4f.txt", "a")
for order in range(1, 4):
f.write("\n Level: Word\n")
f.write(f"\n Order: {order}\n")
model1 = MarcovModel("franz_kafka_collected_works.txt", "word", 1)
model2 = MarcovModel("edgar_allan_poe_collected_works.txt", "word", 1)
models = [model1, model2]
length_of_test_input = 60
for model in models:
f.write(f"\n\n-----Estimation Results for {model.corpus_filename}-----\n")
for estimator_model in models:
f.write(f"\n-----When estimated by: {estimator_model.corpus_filename}-----\n")
estimates = []
text = ""
nextPos = 0
while nextPos < (len(model.testset) - length_of_test_input):
text = model.testset[nextPos:nextPos + length_of_test_input]
estimate = estimator_model.estimate(text)
# print(estimate)
estimates.append(estimate)
text = ""
nextPos += length_of_test_input
f.write(f"{statistics.mean(estimates)}")
def component2f():
print("-----Component 2f (Check results folder)-----")
f = open("results/component2f.txt", "a")
for order in range(1, 13):
f.write(f"\n Order: {order} \n")
model = MarcovModel("harry_potter_1.txt", "word", order)
f.write(model.generate(40))
def component2b():
print("-----Component 2b (Check results folder)-----")
f = open("results/component2b.txt", "a")
for order in range(1, 6):
f.write(f"\n-----Level: Word / Order: {order}-----\n")
model = MarcovModel("franz_kafka_collected_works.txt", "word", order)
f.write(model.generate(20, "The"))
def component4():
# Basic idea for the hybrid generation is to prioritize meaning over syntax. In other words, we generate a text using the original generate method. Then you correct any jarring grammar mistakes with the pos generation scheme.
print("-----working on component4 (Check 'results' folder)-----")
f = open("results/component4.txt", "a")
model = MarcovModel("arthur_conan_doyle_collected_works.txt",
level="word",
order=3,
hybrid=True)
for i in range(3):
f.write("-----Hybrid Generate Engaged, Level: word, Order: 3-----\n")
f.write(model.generate(50) + "\n")
def component3b():
print("-----working on component3b (Check 'results' folder)-----")
f = open("results/component3b.txt", "a")
f.write("\n-----POS ENGAGED-----\n\n")
for order in range(1, 8):
model = MarcovModel("alexander_dumas_collected_works.txt",
level="word",
order=order,
pos=True)
f.write(
f"-----Results for {model.corpus_filename}, Order: {order} -----\n"
)
f.write(model.generate(30) + "\n")
def component2e():
print("-----Component 2e (Check results folder)-----")
model1 = MarcovModel("franz_kafka_collected_works.txt", "character", 5)
model2 = MarcovModel("arthur_conan_doyle_collected_works.txt", "word", 3)
model3 = MarcovModel("edgar_allan_poe_collected_works.txt", "word", 2)
models = [model1, model2, model3]
f = open("results/component2e.txt", "a")
for model in models:
f.write(
f"\n-----Results for {model.corpus_filename} / Level: {model.level} / Order: {model.order}-----\n"
)
f.write(model.generate(30))
def component3b():
print("-----Component 3b (Check results folder)-----")
f = open("results/component3b.txt", "a")
# Sue
model1 = MarcovModel("harry_potter_1.txt", "word", 5)
# Yemi
model2 = MarcovModel("franz_kafka_collected_works.txt", "word", 5)
# Nicole
model3 = MarcovModel("origin_of_species.txt", "word", 5)
# Maanya
model4 = MarcovModel("practice_phrenology_simplified.txt", "word", 5)
models = [model1, model2, model3, model4]
for model in random.sample(models, len(models)):
f.write("\nmodel: \n")
f.write(model.generate(30))
def component3c():
print("-----Component 3c (Check results folder)-----")
f = open("results/component3c.txt", "a")
# Sue
model1 = MarcovModel("harry_potter_1.txt", "word", 4)
# Yemi
model2 = MarcovModel("franz_kafka_collected_works.txt", "word", 4)
# Nicole
model3 = MarcovModel("origin_of_species.txt", "word", 4)
# Maanya
model4 = MarcovModel("practice_phrenology_simplified.txt", "word", 4)
models = [model1, model2, model3, model4]
for model in random.sample(models, len(models)):
f.write(f"\n-----{model.corpus_filename}/ Level: Word / Order: 4-----\n")
f.write(model.generate(35))
def component2d():
print("-----Component 2d (Check results folder)-----")
f = open("results/component2d.txt", "a")
f.write(f"\n-----Mystery Model / Level: Character / Order: 3-----\n")
model1 = MarcovModel("franz_kafka_collected_works.txt", "character", 3)
model2 = MarcovModel("arthur_conan_doyle_collected_works.txt", "character",
3)
f.write("\n Results for Mystery Model \n")
f.write(model1.generate(50, "The"))
f.write("\n\n Results for Mystery Model \n")
f.write(model2.generate(50, "The"))
def component3c():
print("-----working on component3c (Check 'results' folder)-----")
f = open("results/component3c.txt", "a")
f.write("-----without the POS-----")
model = MarcovModel("alexander_dumas_collected_works.txt",
level="word",
order=5,
pos=False)
f.write(model.generate(30) + "\n")
f.write("-----with the POS-----")
model2 = MarcovModel("alexander_dumas_collected_works.txt",
level="word",
order=5,
pos=True)
f.write(model2.generate(30) + "\n")
def component2c():
print("-----Component 2c (Check results folder)-----")
f = open("results/component2c.txt", "a")
for order in range(1, 6):
f.write(f"\n-----Mystery Model / Level: Word / Order: {order}-----\n")
model1 = MarcovModel("franz_kafka_collected_works.txt", "word", order)
model2 = MarcovModel("arthur_conan_doyle_collected_works.txt", "word",
order)
f.write("\n Level: word \n")
f.write(f"\n Order: {order} \n")
f.write("\n Results for Mystery Model \n")
f.write(model1.generate(50, "The"))
f.write("\n Results for Mystery Model \n")
f.write(model2.generate(50, "The"))