def load_model(directory):
from keras.models import load_model as keras_load_model
model = keras_load_model(os.path.join(directory, BULK))
with open(os.path.join(directory, METADATA), "rb") as conf:
meta = json.load(conf)
model.metadata = meta
return model
def __init__(self, teacher_model):
self.train_model, self.born_again_model = None, None
self.temperature = args.temperature
self.teacher_model = keras_load_model(teacher_model)
for i in range(len(self.teacher_model.layers)):
self.teacher_model.layers[i].trainable = False
self.teacher_model.compile(optimizer="adam",
loss="categorical_crossentropy")
self.train_model, self.born_again_model = self.prepare()
self.train_model = convert_gpu_model(self.train_model)
def insert_deep_learning_model(pipeline_step, file_name):
# This is where we saved the random_name for this model
random_name = pipeline_step.model
# Load the Keras model here
keras_file_name = file_name[:-5] + random_name + '_keras_deep_learning_model.h5'
model = keras_load_model(keras_file_name)
# Put the model back in place so that we can still use it to get predictions without having to load it back in from disk
return model
def load_model(model_path, optimizer_='adam'):
'''
Load model
model_path: Path to the model file
'''
model = keras_load_model(model_path)
model.load_weights(model_path)
model = compile_model(model, optimizer_)
return model
def get_reconstructions(model_type, x_train, x_test, count):
params = get_model_parameters(model_type, x_train, x_test)
if exists(params["model_filename"]):
print("loading model : '%s'" % params["model_filename"])
m = keras_load_model(params["model_filename"])
else:
m = train_nn(params)
print("reconstruction prediction...")
Renc = m.predict(params["x_train"][:count, ...]) # reconstruction
return Renc
def load_object(root_folder, obj_descr_type, model_name, dataset,
refactoring_name):
if model_name == 'deep-learning' and obj_descr_type == 'model':
file_name = root_folder + "/" + obj_descr_type + "-" + model_name + "-" + dataset + "-" + refactoring_name.replace(
" ", "") + ".h5"
return keras_load_model(file_name,
custom_objects={
"binary_precision": binary_precision(),
"binary_recall": binary_recall()
})
else:
file_name = root_folder + "/" + obj_descr_type + "-" + model_name + "-" + dataset + "-" + refactoring_name.replace(
" ", "") + ".joblib"
return load(file_name)
def predict(model_file, sdf_file, id_col=None):
from keras.models import load_model as keras_load_model
import theano
import trainer
theano.config.warn.round = False
model = keras_load_model(model_file)
config = model.get_config()
fp_size = config[0]['config']['units']
data = trainer.read_sdf(sdf_file)
fps = trainer.generate_fps(data, fp_size=fp_size)
test_predictions = model.predict(fps)
return data, test_predictions
def load_model():
# load the pre-trained Keras model (here we are using a model
# pre-trained on ImageNet and provided by Keras, but you can
# substitute in your own networks just as easily)
# load as well the dictionary with labels to match with predictions
global session
session = tf.Session()
set_session(session)
global model
model_filepath = "models/export-full-model.hdf5"
model = keras_load_model(model_filepath)
model._make_predict_function()
global idx_to_class
with open("models/classes.json") as f:
idx_to_class = json.load(f)
def load_model(model_file, network_config):
# Import for loading legacy models.
from keras_contrib.layers import Deconvolution3D # noqa
model = keras_load_model(model_file)
# If necessary, wrap the loaded model to transpose the axes for both
# inputs and outputs.
if network_config.transpose:
inputs = []
perms = []
for old_input in model.input_layers:
input_shape = np.asarray(old_input.input_shape)[[3, 2, 1, 4]]
new_input = Input(shape=tuple(input_shape),
dtype=old_input.input_dtype,
name=old_input.name)
perm = Permute((3, 2, 1, 4),
input_shape=tuple(input_shape))(new_input)
inputs.append(new_input)
perms.append(perm)
old_outputs = model(perms)
if not isinstance(old_outputs, list):
old_outputs = [old_outputs]
outputs = []
for old_output in old_outputs:
new_output = Permute((3, 2, 1, 4))(old_output)
outputs.append(new_output)
new_model = Model(input=inputs, output=outputs)
# Monkeypatch the save to save just the underlying model.
func_type = type(model.save)
old_model = model
def new_save(_, *args, **kwargs):
old_model.save(*args, **kwargs)
new_model.save = func_type(new_save, new_model)
model = new_model
return model
def model_load(root_path):
"""
Notice that it is a calling function dedicated to OSARA Tool.
Load model(keras.models.Sequential format).
!Required: Function Name is "model_load"
:type root_path: str
:rtype: keras.models.Sequential
>>> root_path = r"C:\cygwin64\home\spc000\python\\new_project\model"
>>> model = model_load(root_path)
"""
root_path = Path(root_path)
model_path = root_path.joinpath("vggnet_model_acc_0.69.hdf5")
model = keras_load_model(model_path)
return model
def load_model():
# load the pre-trained Keras model (here we are using a model
global model
input_shape = (img_width, img_height, 3)
model = Sequential()
model.add(Conv2D(32, (3, 3), input_shape=input_shape))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3, 3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(64, (3, 3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Activation('hard_sigmoid'))
model = keras_load_model('model.h5')
def get_encodings(model_type, x_train, x_test):
from keras.models import Model
params = get_model_parameters(model_type, x_train, x_test)
if params is None:
m = train_optimal_nn(x_train, x_train, x_test, x_test, params)
else:
if exists(params["model_filename"]):
print("loading model : '%s'" % params["model_filename"])
m = keras_load_model(params["model_filename"])
else:
m = train_nn(params)
encoder = Model(m.input, m.get_layer('bottleneck').output)
print("bottleneck prediction...")
if params is None:
Zenc = encoder.predict(x_train)
else:
Zenc = encoder.predict(params["x_train"]) # bottleneck representation
print("Zenc.shape : %s" % str(Zenc.shape))
return Zenc.reshape((Zenc.shape[0], Zenc.shape[-1]))
def load_model(self, device = "cpu", optimizer = Adam(lr=1e-4), loss = "categorical_crossentropy",\
metrics = ['accuracy'] ):
if (device == "cpu"):
with tf.device("/cpu:0"):
if (loss == "jaccard"):
from jaccard_loss import jaccard_distance
self.model = keras_load_model(
self.model_path,
custom_objects={'jaccard_distance': jaccard_distance})
self.model.compile(optimizer,
loss=jaccard_distance,
metrics=metrics)
elif (loss == "fancy"):
from kerasfancyloss import fancy_loss
self.model = keras_load_model(
self.model_path,
custom_objects={'fancy_loss': fancy_loss})
self.model.compile(optimizer,
loss=fancy_loss,
metrics=metrics)
else:
self.model = keras_load_model(self.model_path)
self.model.compile(optimizer, loss, metrics)
elif (device == "gpu"):
if (loss == "jaccard"):
from jaccard_loss import jaccard_distance
self.model = keras_load_model(
self.model_path,
custom_objects={'jaccard_distance': jaccard_distance})
self.model.compile(optimizer,
loss=jaccard_distance,
metrics=metrics)
elif (loss == "fancy"):
from kerasfancyloss import fancy_loss
self.model = keras_load_model(
self.model_path, custom_objects={'fancy_loss': fancy_loss})
self.model.compile(optimizer, loss=fancy_loss, metrics=metrics)
else:
self.model = keras_load_model(self.model_path)
self.model.compile(optimizer, loss, metrics)
else:
print("Device not understood")
return None
def load_trigger_model(model_dir):
model = keras_load_model(os.path.join(model_dir, 'trigger.hdf'))
return model
def load_model(model_filename):
return keras_load_model(model_filename)
def __init__(self, encoder_name, encoder_file_path):
if encoder_name == "facenet_keras":
self.encoder_model = keras_load_model(encoder_file_path)
def load_model(model_path):
return keras_load_model(model_path, {
'DynamicKMaxPooling': DynamicKMaxPooling,
'KMaxPooling': KMaxPooling
})
def load_model(model_path):
return keras_load_model(model_path)
def load_model(sufix=""):
return keras_load_model(FILENAME + sufix + EXTENSION)
def basic_run(agent, load_model, architecture, training, optimization, memory,
policy, testing):
TRAINED = False
env_name = "warehouse-v0"
env = gym.make(env_name)
# Overwrite the config variables with parse args
if args.gg_colab is not None:
training['visualize'], testing["render"] = (args.gg_colab, ) * 2
if args.do_testing is not None:
testing["do"] = args.do_testing
# make the manager available in the environment
env.manager = manager # used for metrics logging by the callbacks
# The model
nb_actions = env.action_space.n
input_shape = env.OS.input_shape()
if load_model is None:
model = getattr(models, architecture)(input_shape,
nb_actions,
use_bias=False)
else:
model = keras_load_model(manager.get_load_path(load_model, 'model.h5'))
TRAINED = True
model.build(input_shape)
print(model.summary())
# The memory
memory = getattr(rl_memory, memory["name"])(*memory["args"],
**memory["kwargs"])
# The policy
if policy['name'] == 'MaskingMaxBoltzmannQPolicy':
policy = MaskingMaxBoltzmannQPolicy(*policy["args"],
**policy["kwargs"])
elif policy['name'] == "LinearAnnealedPolicy":
policy = LinearAnnealedPolicy(
getattr(rl_policy, policy["sub_policy"])(), *policy["args"],
**policy["kwargs"])
else:
policy = getattr(rl_policy, policy["name"])(*policy["args"],
**policy["kwargs"])
# The agent
if agent['name'] == 'MaskingDQNAgent':
dqn = MaskingDQNAgent(model=model,
environment=env,
nb_actions=nb_actions,
memory=memory,
policy=policy,
*agent["args"],
**agent['kwargs'])
else:
dqn = agents[agent["name"]](model=model,
nb_actions=nb_actions,
memory=memory,
policy=policy,
*agent["args"],
**agent['kwargs'])
dqn.compile(getattr(optimizers,
optimization["optimizer"])(lr=optimization["lr"]),
metrics=['mae'])
if not TRAINED:
# # Training
history = dqn.fit(env, verbose=1,
**training) #callbacks=env.callbacks,
# TODO: Debug verbose=2, error with TrainEpisodeLogger
# Save history
manager.log_scalars("history",
[history.epoch, *history.history.values()],
["epoch", *history.history.keys()])
manager.log_scalars("reward", history.history["episode_reward"])
manager.log_scalars("avg_q", history.history["episode_avg_maxq"])
manager.log_scalars("avg_minq", history.history["episode_avg_minq"])
manager.log_scalars("n_step", history.history["nb_steps"])
manager.log_scalars("nb_episode_steps",
history.history["nb_episode_steps"])
manager.log_scalars("epoch", history.epoch)
# Post-training
# manager.save(dqn.model,"model")
dqn.model.save(os.path.join(manager.save_dir, "model.h5"),
overwrite=True)
# Finally, evaluate our algorithm for 5 episodes.
if testing["do"]:
print("\nTesting the agent")
dqn.test(env,
nb_episodes=testing["nb_episodes"],
visualize=testing["render"],
callbacks=env.callbacks)
def _load_model(self, filepath_model):
print('load model:', filepath_model)
self._model = keras_load_model(filepath_model)
def load_model(use_gmb):
return keras_load_model(get_model_file_name(use_gmb))
def run(self,
epochs=10,
load_model=False,
save_model=False,
train=True,
evaluate=True,
plot=True,
data_prep=True,
clear_data=False):
if load_model:
(train_images, train_labels), (test_images,
test_labels) = load_data()
if not os.path.exists('%s.h5' % self.model_name):
print("\nNO MODEL AVAILABLE\n")
load_model = False
else:
print("\nLOADING MODEL...\n")
self.model = keras_load_model('%s.h5' % self.model_name)
if clear_data and not load_model:
while True:
user_input = input("\nCONFIRM DATA DELETION (y/n): ")
if user_input == 'y':
os.system("sh clear_data.sh")
break
elif user_input == 'n':
clear_data = False
break
if data_prep or clear_data:
print('\n')
while (True):
website = input("Scrape Google(0), Flickr(1), or Bing(2)? ")
if (website == '0'):
download_google_images(self.new_height, self.new_width,
load_model)
print('\n')
break
elif (website == '1'):
download_flickr_photos(self.new_height, self.new_width,
load_model)
print('\n')
break
elif (website == '2'):
download_bing_photos(self.new_height, self.new_width,
load_model)
print('\n')
break
if train and not load_model:
(train_images, train_labels), (test_images,
test_labels) = load_data()
train_images = train_images / 255.0
test_images = test_images / 255.0
self.model = Sequential()
self.model.add(
Conv2D(32, (5, 5),
activation='relu',
input_shape=(self.new_height, self.new_width, 3)))
self.model.add(MaxPooling2D(pool_size=(2, 2)))
self.model.add(Conv2D(32, (5, 5), activation='relu'))
self.model.add(MaxPooling2D(pool_size=(2, 2)))
self.model.add(Flatten())
self.model.add(Dense(1024, activation='relu'))
self.model.add(Dropout(0.5))
self.model.add(Dense(512, activation='relu'))
self.model.add(Dropout(0.5))
self.model.add(Dense(256, activation='relu'))
self.model.add(Dense(2, activation='softmax'))
start_time = time.time()
self.model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
hist = self.model.fit(train_images,
train_labels,
epochs=epochs,
validation_data=(test_images, test_labels))
end_time = time.time() - start_time
if end_time > 3600:
print('Total Training Time: %dhr %.1fmin\n\n' %
(int(end_time / 3600),
((end_time - int(end_time / 3600) * 3600) / 60)))
elif end_time > 60:
print("Total Training Time: %dmin %.2fs\n\n" %
((end_time / 60), (end_time - int(end_time / 60) * 60)))
else:
print("Total Training Time: %.2fs\n\n" % end_time)
if save_model and not load_model:
self.model.save('%s.h5' % self.model_name)
if evaluate:
print("TESTING")
self.model.evaluate(test_images, test_labels)[1]
print('\n')
if plot and not load_model:
f = plt.figure()
f.add_subplot(2, 1, 1)
plt.plot(hist.history['accuracy'], label='train accuracy')
plt.plot(hist.history['val_accuracy'], label='val accuracy')
plt.title('Model Accuracy (top) and Model Loss (bottom)')
plt.ylabel('Accuracy')
plt.legend(loc='lower left')
f.add_subplot(2, 1, 2)
plt.plot(hist.history['loss'], label='train loss')
plt.plot(hist.history['val_loss'], label='val loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.ylim([0, 2])
plt.legend(loc='lower left')
plt.show()
def load_trigger_modelfile(filepath):
return keras_load_model(str(filepath))
#return (commandString,commandVals)
return (commandVals)
if len(sys.argv) != 7:
print("Error with command line inputs")
sys.exit(0)
else:
FILE_NAME = sys.argv[1]
MODEL_PATH = sys.argv[2]
OUTPUT_PATH = sys.argv[3]
LOWER_BOUND = int(sys.argv[4])
UPPER_BOUND = int(sys.argv[5])
THREADS = int(sys.argv[6])
model = keras_load_model(MODEL_PATH)
nlf = pd.read_csv('NLF.csv', index_col=0)
conn = sqlite3.connect(OUTPUT_PATH)
c = conn.cursor()
c.execute("CREATE TABLE IF NOT EXISTS predictions(Sequence TEXT,\
Offset TEXT,\
Length TEXT,\
Protein_ID TEXT,\
PrecursorMZ float,\
Retention_Time float,\
Charge integer,\
b1 TEXT,\
b2 TEXT,\
bn1 TEXT,\
bn2 TEXT,\
def load_model(self, path):
"""Load Keras model from file."""
self._model = keras_load_model(path)
def load_model(path):
return keras_load_model(path)
for i in zip(array1, array2):
print(i[0], " | ", i[1])
if len(sys.argv) != 2:
print("Error with command line inputs")
sys.exit(0)
else:
INPUT_PATH = sys.argv[1]
df = pd.read_pickle(INPUT_PATH + "dfTest.pkl").reindex()
xTest = np.load(INPUT_PATH + "XTest.npy")
yTest = np.load(INPUT_PATH + "YTest.npy")
#model = keras_load_model(INPUT_PATH + "model.h5", custom_objects={'cosine_similarity': cosine_similarity})
model = keras_load_model(INPUT_PATH + "model.h5")
predictions = model.predict(xTest)
ionList = [
"b1", "b2", "bn1", "bn2", "bo1", "bo2", "y1", "y2", "yn1", "yn2", "yo1",
"yo2"
]
"""
for i, val in enumerate(predictions):
printResults(yTest[i], val)
"""
print(xTest.shape)
print(yTest.shape)
print(df.shape)
def fit(self, X, y):
global keras_imported, KerasRegressor, KerasClassifier, EarlyStopping, ModelCheckpoint, TerminateOnNaN, keras_load_model
self.model_name = get_name_from_model(self.model)
X_fit = X
if self.model_name[:12] == 'DeepLearning' or self.model_name in [
'BayesianRidge', 'LassoLars', 'OrthogonalMatchingPursuit',
'ARDRegression', 'Perceptron', 'PassiveAggressiveClassifier',
'SGDClassifier', 'RidgeClassifier', 'LogisticRegression'
]:
if scipy.sparse.issparse(X_fit):
X_fit = X_fit.todense()
if self.model_name[:12] == 'DeepLearning':
if keras_imported == False:
# Suppress some level of logs
os.environ['TF_CPP_MIN_VLOG_LEVEL'] = '3'
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
from keras.callbacks import EarlyStopping, ModelCheckpoint, TerminateOnNaN
from keras.models import load_model as keras_load_model
from keras.wrappers.scikit_learn import KerasRegressor, KerasClassifier
keras_imported = True
# For Keras, we need to tell it how many input nodes to expect, which is our num_cols
num_cols = X_fit.shape[1]
model_params = self.model.get_params()
del model_params['build_fn']
try:
del model_params['feature_learning']
except:
pass
try:
del model_params['num_cols']
except:
pass
if self.type_of_estimator == 'regressor':
self.model = KerasRegressor(
build_fn=utils_models.make_deep_learning_model,
num_cols=num_cols,
feature_learning=self.feature_learning,
**model_params)
elif self.type_of_estimator == 'classifier':
self.model = KerasClassifier(
build_fn=utils_models.make_deep_learning_classifier,
num_cols=num_cols,
feature_learning=self.feature_learning,
**model_params)
if self.model_name[:12] == 'DeepLearning':
try:
if self.is_hp_search == True:
patience = 5
verbose = 0
else:
patience = 25
verbose = 2
X_fit, y, X_test, y_test = self.get_X_test(X_fit, y)
try:
X_test = X_test.toarray()
except AttributeError as e:
pass
if not self.is_hp_search:
print(
'\nWe will stop training early if we have not seen an improvement in validation accuracy in {} epochs'
.format(patience))
print(
'To measure validation accuracy, we will split off a random 10 percent of your training data set'
)
early_stopping = EarlyStopping(monitor='val_loss',
patience=patience,
verbose=verbose)
terminate_on_nan = TerminateOnNaN()
now_time = datetime.datetime.now()
time_string = str(now_time.year) + '_' + str(
now_time.month) + '_' + str(now_time.day) + '_' + str(
now_time.hour) + '_' + str(now_time.minute)
temp_file_name = 'tmp_dl_model_checkpoint_' + time_string + str(
random.random()) + '.h5'
model_checkpoint = ModelCheckpoint(temp_file_name,
monitor='val_loss',
save_best_only=True,
mode='min',
period=1)
callbacks = [early_stopping, terminate_on_nan]
if not self.is_hp_search:
callbacks.append(model_checkpoint)
self.model.fit(X_fit,
y,
callbacks=callbacks,
validation_data=(X_test, y_test),
verbose=verbose)
# TODO: give some kind of logging on how the model did here! best epoch, best accuracy, etc.
if self.is_hp_search is False:
self.model = keras_load_model(temp_file_name)
try:
os.remove(temp_file_name)
except OSError as e:
pass
except KeyboardInterrupt as e:
print(
'Stopping training at this point because we heard a KeyboardInterrupt'
)
print(
'If the deep learning model is functional at this point, we will output the model in its latest form'
)
print(
'Note that this feature is an unofficial beta-release feature that is known to fail on occasion'
)
if self.is_hp_search is False:
self.model = keras_load_model(temp_file_name)
try:
os.remove(temp_file_name)
except OSError as e:
pass
elif self.model_name[:4] == 'LGBM':
X_fit = X.toarray()
X_fit, y, X_test, y_test = self.get_X_test(X_fit, y)
try:
X_test = X_test.toarray()
except AttributeError as e:
pass
if self.type_of_estimator == 'regressor':
eval_metric = 'rmse'
elif self.type_of_estimator == 'classifier':
if len(set(y_test)) > 2:
eval_metric = 'multi_logloss'
else:
eval_metric = 'binary_logloss'
verbose = True
if self.is_hp_search == True:
verbose = False
if self.X_test is not None:
eval_name = 'X_test_the_user_passed_in'
else:
eval_name = 'random_holdout_set_from_training_data'
cat_feature_indices = self.get_categorical_feature_indices()
if cat_feature_indices is None:
self.model.fit(X_fit,
y,
eval_set=[(X_test, y_test)],
early_stopping_rounds=100,
eval_metric=eval_metric,
eval_names=[eval_name],
verbose=verbose)
else:
self.model.fit(X_fit,
y,
eval_set=[(X_test, y_test)],
early_stopping_rounds=100,
eval_metric=eval_metric,
eval_names=[eval_name],
categorical_feature=cat_feature_indices,
verbose=verbose)
elif self.model_name[:8] == 'CatBoost':
X_fit = X_fit.toarray()
if self.type_of_estimator == 'classifier' and len(
pd.Series(y).unique()) > 2:
# TODO: we might have to modify the format of the y values, converting them all to ints, then back again (sklearn has a useful inverse_transform on some preprocessing classes)
self.model.set_params(loss_function='MultiClass')
cat_feature_indices = self.get_categorical_feature_indices()
self.model.fit(X_fit, y, cat_features=cat_feature_indices)
elif self.model_name[:16] == 'GradientBoosting':
if not sklearn_version > '0.18.1':
X_fit = X_fit.toarray()
patience = 20
best_val_loss = -10000000000
num_worse_rounds = 0
best_model = deepcopy(self.model)
X_fit, y, X_test, y_test = self.get_X_test(X_fit, y)
# Add a variable number of trees each time, depending how far into the process we are
if os.environ.get('is_test_suite', False) == 'True':
num_iters = list(range(1, 50, 1)) + list(range(
50, 100, 2)) + list(range(100, 250, 3))
else:
num_iters = list(range(
1, 50, 1)) + list(range(50, 100, 2)) + list(
range(100, 250, 3)) + list(range(250, 500, 5)) + list(
range(500, 1000, 10)) + list(range(
1000, 2000, 20)) + list(range(
2000, 10000, 100))
# TODO: get n_estimators from the model itself, and reduce this list to only those values that come under the value from the model
try:
for num_iter in num_iters:
warm_start = True
if num_iter == 1:
warm_start = False
self.model.set_params(n_estimators=num_iter,
warm_start=warm_start)
self.model.fit(X_fit, y)
if self.training_prediction_intervals == True:
val_loss = self.model.score(X_test, y_test)
else:
try:
val_loss = self._scorer.score(self, X_test, y_test)
except Exception as e:
val_loss = self.model.score(X_test, y_test)
if val_loss - self.min_step_improvement > best_val_loss:
best_val_loss = val_loss
num_worse_rounds = 0
best_model = deepcopy(self.model)
else:
num_worse_rounds += 1
print(
'[' + str(num_iter) +
'] random_holdout_set_from_training_data\'s score is: '
+ str(round(val_loss, 3)))
if num_worse_rounds >= patience:
break
except KeyboardInterrupt:
print(
'Heard KeyboardInterrupt. Stopping training, and using the best checkpointed GradientBoosting model'
)
pass
self.model = best_model
print(
'The number of estimators that were the best for this training dataset: '
+ str(self.model.get_params()['n_estimators']))
print('The best score on the holdout set: ' + str(best_val_loss))
else:
self.model.fit(X_fit, y)
if self.X_test is not None:
del self.X_test
del self.y_test
return self
sys.exit(0)
else:
INPUT_PATH = sys.argv[1]
PRED_TYPE = 0 #Ms2
#PRED_TYPE = 1 #Retention Time
df = pd.read_pickle(INPUT_PATH + "dfTest.pkl").reindex()
xTest = np.load(INPUT_PATH + "XTest.npy")
pepOut = open(INPUT_PATH + "prositInput.csv", "w")
prositCompare = open(INPUT_PATH + "prositCompare.csv", "w")
if PRED_TYPE == 0:
yTest = np.load(INPUT_PATH + "YTestMs2.npy")
#model = keras_load_model(INPUT_PATH + "model.h5", custom_objects={'cosine_similarity': cosine_similarity})
model = keras_load_model(INPUT_PATH + "modelMs2.h5")
predictions = model.predict(xTest)
ionList = [
"b1", "b2", "bn1", "bn2", "bo1", "bo2", "y1", "y2", "yn1", "yn2",
"yo1", "yo2"
]
jsonArray = {}
tmpPred = {}
jsonArrayExp = {}
tmpPredExp = {}
for i in range(len(yTest)):
#for i in range(1000):
def load_model(model_filename, models_folder=MODELS_FOLDER):
return keras_load_model(os.path.join(models_folder, model_filename))
def perform_model_inference(self, _path_infer_this_image,
_objects_found_info_arr):
print(
f"\n\nExecuting model inference on image_to_infer : {_path_infer_this_image}\n"
)
## load the keras pretained model if not already loaded
if self.reloaded_yolo_model is None:
print(f"\n\n LOADED KERAS MODEL \n\n")
saved_model_location = r'/home/rohit/PyWDUbuntu/thesis/saved_keras_model/yolov3_coco80.saved.model'
self.reloaded_yolo_model = keras_load_model(saved_model_location)
## set some parameters for network
net_h, net_w = 416, 416
obj_thresh, nms_thresh = 0.5, 0.45
anchors = [[116, 90, 156, 198, 373, 326], [30, 61, 62, 45, 59, 119],
[10, 13, 16, 30, 33, 23]]
labels = ["person", "bicycle", "car", "motorbike", "aeroplane", "bus", "train", "truck", \
"boat", "traffic light", "fire hydrant", "stop sign", "parking meter", "bench", \
"bird", "cat", "dog", "horse", "sheep", "cow", "elephant", "bear", "zebra", "giraffe", \
"backpack", "umbrella", "handbag", "tie", "suitcase", "frisbee", "skis", "snowboard", \
"sports ball", "kite", "baseball bat", "baseball glove", "skateboard", "surfboard", \
"tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl", "banana", \
"apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut", "cake", \
"chair", "sofa", "pottedplant", "bed", "diningtable", "toilet", "tvmonitor", "laptop", "mouse", \
"remote", "keyboard", "cell phone", "microwave", "oven", "toaster", "sink", "refrigerator", \
"book", "clock", "vase", "scissors", "teddy bear", "hair drier", "toothbrush"]
image_to_infer_cv2 = cv2.imread(_path_infer_this_image)
image_h, image_w, _ = image_to_infer_cv2.shape
try:
image_to_infer_preprocessed = self.preprocess_input(
image_to_infer_cv2, net_h, net_w)
except Exception as error_inference_preprocess_image:
print(
f"\nFATAL ERROR: Problem reading the input file.\nError message: {error_inference_preprocess_image}\nExit RC=400"
)
exit(400)
## run the prediction
yolos = self.reloaded_yolo_model.predict(image_to_infer_preprocessed)
boxes = []
for i in range(len(yolos)):
## decode the output of the network
boxes += self.decode_netout(yolos[i][0], anchors[i], obj_thresh,
nms_thresh, net_h, net_w)
## correct the sizes of the bounding boxes
self.correct_yolo_boxes(boxes, image_h, image_w, net_h, net_w)
## suppress non-maximal boxes
self.do_nms(boxes, nms_thresh)
## draw bounding boxes into the image
self.draw_boxes(image_to_infer_cv2, boxes, labels, obj_thresh,
_objects_found_info_arr)
## save the image as intermediate file -- see later whether to return and processing is possible
cv2.imwrite(r'./intermediate_file_inferenece_image.jpg',
(image_to_infer_cv2).astype('uint8'))
