def demo():
inputSentence = input("Enter your sentence: ")
print("============" + p + "===================")
print(type(p))
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
evaluate(ptrNet, inputSentence)
def processSingle(path):
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
ptrNet.load_state_dict(torch.load(path))
nlp = spacy.load("en_core_web_sm")
sentences = [
"this is definitely a difficult sentence", "your plan seems excellent",
"it is really working", "let me start with a poem",
"he is sure to pass the exam", "we meet every wednesday",
"do plan to we it tomorrow", "i like working on brain",
"development good is a this", "i am going to canada",
"The script that I am describing here mainly for training and validation "
]
if True:
for origLine in sentences:
print("")
inputLine, outputLine = processSentence(nlp, ptrNet, origLine)
print("Input Sentence: ", inputLine)
print("Outpt Sentence: ", outputLine)
print("Orig Sentence: ", origLine)
else:
lines = open("../data/englishSentences_test.dat").read().splitlines()
for i in range(7):
print("\n")
origLine = random.choice(lines)
inputLine, outputLine = processSentence(nlp, ptrNet, origLine)
print("Input Sentence: ", inputLine)
print("Outpt Sentence: ", outputLine)
print("Orig Sentence: ", origLine)
def modelEvaluate(path):
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
ptrNet.load_state_dict(torch.load(path))
evaluateWordSort(ptrNet, 1)
def modelEvaluate(accuracyStats, path):
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
ptrNet.load_state_dict(torch.load(path))
for epoch in range(EPOCHS):
evaluateWordSort(accuracyStats, ptrNet, 1)
printAccStats(accuracyStats)
def modelEvaluate(accuracyStats, path):
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
ptrNet.load_state_dict(torch.load(path))
for sentenceLength in range(4, 11):
print(sentenceLength)
evaluateWordSort(accuracyStats, ptrNet, sentenceLength)
printAccStats(accuracyStats)
def modelTrainAndSave(path):
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
optimizer = optim.Adam(ptrNet.parameters())
program_starts = time.time()
for epoch in range(EPOCHS):
train(ptrNet, optimizer, epoch + 1)
evaluateWordSort(ptrNet, epoch + 1)
torch.save(ptrNet.state_dict(), path)
now = time.time()
print("It has been {0} seconds since the loop started".format(
now - program_starts))
for i in range(out.size(0)):
print("=============================================")
print(
"yref",
y_val[i],
out[i],
y_val[i] - out[i],
)
xv = convertToWordSingle(x_val[i])
print("orig", xv)
v = out[i].numpy()
print("[", end="")
for index in v:
print(xv[index] + ", ", end="")
print("]")
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
optimizer = torch.optim.Adam(ptrNet.parameters())
program_starts = time.time()
for epoch in range(EPOCHS):
train(ptrNet, optimizer, epoch + 1)
evaluateWordSort(ptrNet, epoch + 1)
now = time.time()
print("It has been {0} seconds since the loop started".format(now -
program_starts))
out, _ = model(x_val, y_val, teacher_force_ratio=0.)
out = out.permute(1, 0)
for i in range(out.size(0)):
print("=============================================")
print("yref", y_val[i], out[i], y_val[i] - out[i])
print("orig", text_val[i])
v = torch.Tensor.cpu(out[i]).numpy()
print("[", end="")
for index in v:
print(text_val[i][index] + " ", end="")
print("]")
if config.GPU == True:
ptrNet = PointerNetwork(config.HIDDEN_SIZE).cuda()
else:
ptrNet = PointerNetwork(config.HIDDEN_SIZE)
optimizer = optim.Adam(ptrNet.parameters())
program_starts = time.time()
for epoch in range(EPOCHS):
train(ptrNet, optimizer, epoch + 1)
evaluateWordSort(ptrNet, epoch + 1)
now = time.time()
print("It has been {0} seconds since the loop started".format(now -
program_starts))