def main(input_data_dump, num_frames_per_video, batch_size, labels, model_file):
# Get our data.
X_train, X_test, y_train, y_test = get_data(input_data_dump, num_frames_per_video, labels, True)
num_classes = len(labels)
size_of_each_frame = X_train.shape[2]
# Get our network.
net = get_network_wide(num_frames_per_video, size_of_each_frame, num_classes)
# Train the model.
try:
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/' + model_file)
print("\nModel already exists! Loading it")
print("Model Loaded")
except Exception:
model = tflearn.DNN(net, tensorboard_verbose=0)
print("\nNo previous checkpoints of %s exist" % (model_file))
model.fit(X_train, y_train, validation_set=(X_test, y_test),
show_metric=True, batch_size=batch_size, snapshot_step=100,
n_epoch=100)
# Save it.
x = input("Do you wanna save the model and overwrite? y or n: ")
if(x.strip().lower() == "y"):
model.save('checkpoints/' + model_file)
def main(filename, frames, batch_size, num_classes, input_length):
# Get our data.
X_train, X_test, y_train, y_test = get_data(
filename, frames, num_classes, input_length, True)
# Get sizes.
print ("y_train[0] : ", y_train)
num_classes = len(y_train[0])
print ("Num classes : - ", num_classes)
# print "Y train : ", y_train[0]
# Get our network.
# net = get_network_deep(frames, input_length, num_classes)
net = get_network_wide(frames, input_length, num_classes)
# Train the model.
dir = os.path.join(os.getcwd(), "checkpoints")
if not os.path.exists(dir):
os.makedirs(dir)
if os.path.exists('checkpoints/rnn.tflearn'):
print ("Model already exists! Loading it")
model.load('checkpoints/rnn.tflearn',weights_only="true")
print ("Model Loaded")
else:
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X_train, y_train, validation_set=(X_test, y_test),
show_metric=True, batch_size=batch_size, snapshot_step=100,
n_epoch=10)
# Save it.
print ("Do you wanna save the model and overwrite? y or n")
x = input()
if(x == "y"):
model.save('checkpoints/rnn.tflearn')
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, X_test, y_train, y_test = get_data(filename, frames, num_classes,
input_length, True)
# Get sizes.
print "y_train[0] : ", y_train
num_classes = len(y_train[0])
print "Num classes : - ", num_classes
# print "Y train : ", y_train[0]
# Get our network.
# net = get_network_deep(frames, input_length, num_classes)
net = get_network_wide(frames, input_length, num_classes)
# Train the model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/rnn.tflearn')
print "Model Loaded"
model.fit(X_train,
y_train,
validation_set=(X_test, y_test),
show_metric=True,
batch_size=batch_size,
snapshot_step=100,
n_epoch=10)
# Save it.
print "Do you wanna save the model and overwrite? y or n"
x = raw_input()
if (x == "y"):
model.save('checkpoints/rnn.tflearn')
def main(filename, frames, batch_size, num_classes, input_length):
# Get our data.
X, Y = get_data(input_data_dump, num_frames_per_video, labels, False)
num_classes = len(labels)
size_of_each_frame = X.shape[2]
# Get our network.
net = get_network_wide(num_frames_per_video, size_of_each_frame, num_classes)
# Train the model.
model = tflearn.DNN(net, tensorboard_verbose=0)
try:
model.load('checkpoints/' + model_file)
print("\nModel Exists! Loading it")
print("Model Loaded")
except Exception:
print("\nNo previous checkpoints of %s exist" % (model_file))
print("Exiting..")
sys.exit()
predictions = model.predict(X)
predictions = np.array([np.argmax(pred) for pred in predictions])
Y = np.array([np.argmax(each) for each in Y])
# Writing predictions and gold labels to file
rev_labels = dict(zip(list(labels.values()), list(labels.keys())))
print(rev_labels)
with open("result.txt", "w") as f:
f.write("gold, pred\n")
for a, b in zip(Y, predictions):
f.write("%s %s\n" % (rev_labels[a], rev_labels[b]))
acc = 100 * np.sum(predictions == Y) / len(Y)
print("Accuracy: ", acc)
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, _, y_train, _ = get_data(filename, frames, num_classes,
input_length)
# Get sizes.
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Get our model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load(RNN_MODEL)
print(f"Y: {y_train}")
# Evaluate.
#res = model.evaluate(X_train, y_train) #Evaluates model accuracy.
res = model.predict(X_train) #Predicts probabilities of each class
print(f"RESULTADO: {res}")
print(f"RESULTADO tipo: {type(res)}")
print(f"RESULTADO shape: {res.shape}")
res = model.predict_label(
X_train
) #Predicts bitmask of classed (1 if it is that class, 0 if not)
print(f"RESULTADO LABEL: {res}")
def main(filename, frames, batch_size, num_classes, input_length,
output_model):
"""From the blog post linked above."""
# Get our data.
X_train, X_test, y_train, y_test = get_data(filename, frames, num_classes,
input_length)
# Get sizes.
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Train the model.
print(f"Y_TEST[0]: {y_test[0]}")
print(f"Y_TEST[0] tipo: {type(y_test[0])}, shape: {y_test[0].shape}")
input()
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X_train,
y_train,
validation_set=(X_test, y_test),
show_metric=True,
batch_size=batch_size,
snapshot_step=100,
n_epoch=4)
# Save it.
model.save(output_model)
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, X_test, y_train, y_test = get_data(filename, frames, num_classes,
input_length)
# Get sizes.
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Train the model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X_train,
y_train,
validation_set=(X_test, y_test),
show_metric=True,
batch_size=batch_size,
snapshot_step=100,
n_epoch=4)
# Save it.
model.save('checkpoints/rnn.tflearn')
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/rnn.tflearn')
# Evaluate. Note that although we're passing in "train" data,
# this is actually our holdout dataset, so we never actually
# used it to train the model. Bad variable semantics.
print(model.evaluate(X_train, y_train))
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, y_train = get_data(
filename, frames, num_classes, input_length, False)
# print X_train
# print y_train
# Get sizes.
print ("Y train :- ", y_train[0])
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
dir = os.path.join(os.getcwd(), "checkpoints")
if not os.path.exists(dir):
os.makedirs(dir)
print ("HELLO1")
# Get our model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/rnn.tflearn',weights_only="true")
print ("HELLO2")
# Evaluate.
hc = model.predict(X_train)
print ("HELLO3")
hc = [np.argmax(every) for every in hc]
print ("HELLO4")
aadi = [np.argmax(every) for every in y_train]
print ("HELLO5")
print ("l1 :", len(aadi))
print ("HELLO6")
print ("l2 ", len(hc))
print ("HELLO7")
answer = []
for i in range(0, len(hc)):
answer.append([aadi[i], hc[i]])
print ("HELLO8")
answer.sort()
print ("HELLO9")
f = open("results.txt", "wb")
print ("HELLO10")
for x in answer:
print (x[0], x[1])
f.write(str(x[0])+" "+str(x[1])+"\n")
print ("HELLO11")
print(model.evaluate(X_train, y_train))
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, _, y_train, _ = get_data(filename, frames, num_classes, input_length)
# Get sizes.
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Get our model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/rnn.tflearn')
# Evaluate.
print(model.evaluate(X_train, y_train))
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, X_test, y_train, y_test = get_data(filename, frames, num_classes, input_length)
# Get sizes.
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Train the model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X_train, y_train, validation_set=(X_test, y_test),
show_metric=True, batch_size=batch_size, snapshot_step=100,
n_epoch=4)
# Save it.
model.save('checkpoints/rnn.tflearn')
def main(filename, frames, batch_size, num_classes, input_length):
"""From the blog post linked above."""
# Get our data.
X_train, y_train = get_data(filename, frames, num_classes, input_length,
False)
# print X_train
# print y_train
# Get sizes.
# print ("Y train :- ", y_train)
print("X_train:-", X_train[0])
num_classes = len(y_train[0])
# Get our network.
net = get_network_wide(frames, input_length, num_classes)
# Get our model.
model = tflearn.DNN(net, tensorboard_verbose=0)
model.load('checkpoints/rnn.tflearn')
# model.load('checkpoints_backup/rnn.tflearn')
# Evaluate.
hc = model.predict(X_train)
hc = [np.argmax(every) for every in hc]
print(hc)
aadi = [np.argmax(every) for every in y_train]
print("l1 :", len(aadi))
print("l2 ", len(hc))
answer = []
for i in range(0, len(hc)):
answer.append([aadi[i], hc[i]])
answer.sort()
print(answer)
f = open("results.txt", "w")
for x in answer:
print(x[0], x[1])
f.write(str(x[0]) + " " + str(x[1]) + "\n")
print(model.evaluate(X_train, y_train))