def fetch_internal(net, val, perfix, epoch, is_rnn=False):
def verify(name):
exclude = [
'weight', 'bias', 'gamma', 'beta', 'blockgrad', 'data', 'label'
]
for _ in exclude:
if _ in name:
print 'Abandoned:', name
return False
# if name.startswith('_'):
# return False
for _ in ['c', 'h']:
if name == _:
print 'Abandoned:', name
return False
return True
net = net.get_internals()
print '\n', net.list_outputs(), '\n'
features = [
net[i] for i in range(len(net.list_outputs())) if verify(net[i].name)
]
names = [_.name for _ in features]
net = mx.sym.Group(features)
from mxnet.model import load_checkpoint
sym, arg, aux = load_checkpoint(perfix, epoch)
if not is_rnn:
model = mx.model.FeedForward(net,
ctx=mu.gpu(1),
num_epoch=1,
begin_epoch=0)
else:
from HeartDeepLearning.RNN import rnn_feed
model = rnn_feed.Feed(net, ctx=mu.gpu(1), num_epoch=1, begin_epoch=0)
shape = OrderedDict(val.provide_data + val.provide_label)
model._init_params(shape)
model.arg_params.update(arg)
model.aux_params.update(aux)
print '\nStart Predict'
outputs, img, label = mu.predict_draw(model, val)
outputs = dict(zip(names, outputs))
arg = {k: v.asnumpy() for k, v in arg.items()}
aux = {k: v.asnumpy() for k, v in aux.items()}
for _ in ['outputs', 'arg', 'aux']:
print '\nKey of %s' % _
o = locals()[_]
for k in o:
print k, o[k].shape, o[k].mean(), o[k].std()
print 'Done'
return outputs, img, label, arg, aux
def fetch_internal(net, val, perfix, epoch, is_rnn=False):
def verify(name):
exclude = ['weight', 'bias', 'gamma',
'beta', 'blockgrad', 'data', 'label']
for _ in exclude:
if _ in name:
print 'Abandoned:', name
return False
# if name.startswith('_'):
# return False
for _ in ['c', 'h']:
if name == _:
print 'Abandoned:', name
return False
return True
net = net.get_internals()
print '\n', net.list_outputs(), '\n'
features = [net[i]
for i in range(len(net.list_outputs())) if verify(net[i].name)]
names = [_.name for _ in features]
net = mx.sym.Group(features)
from mxnet.model import load_checkpoint
sym, arg, aux = load_checkpoint(perfix, epoch)
if not is_rnn:
model = mx.model.FeedForward(
net, ctx=mu.gpu(1), num_epoch=1, begin_epoch=0)
else:
from HeartDeepLearning.RNN import rnn_feed
model = rnn_feed.Feed(net, ctx=mu.gpu(1), num_epoch=1, begin_epoch=0)
shape = OrderedDict(val.provide_data + val.provide_label)
model._init_params(shape)
model.arg_params.update(arg)
model.aux_params.update(aux)
print '\nStart Predict'
outputs, img, label = mu.predict_draw(model, val)
outputs = dict(zip(names, outputs))
arg = { k:v.asnumpy() for k,v in arg.items() }
aux = { k:v.asnumpy() for k,v in aux.items() }
for _ in ['outputs', 'arg', 'aux']:
print '\nKey of %s' % _
o = locals()[_]
for k in o:
print k, o[k].shape, o[k].mean(), o[k].std()
print 'Done'
return outputs, img, label, arg, aux
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
"""Create a model from previously saved checkpoint.
Parameters
----------
prefix : str
path prefix of saved model files. You should have
"prefix-symbol.json", "prefix-xxxx.params", and
optionally "prefix-xxxx.states", where xxxx is the
epoch number.
epoch : int
epoch to load.
load_optimizer_states : bool
whether to load optimizer states. Checkpoint needs
to have been made with save_optimizer_states=True.
data_names : list of str
Default is `('data')` for a typical model used in image classification.
label_names : list of str
Default is `('softmax_label')` for a typical model used in image
classification.
logger : Logger
Default is `logging`.
context : Context or list of Context
Default is `cpu()`.
work_load_list : list of number
Default `None`, indicating uniform workload.
fixed_param_names: list of str
Default `None`, indicating no network parameters are fixed.
"""
sym, args, auxs = load_checkpoint(prefix, epoch)
# if loading old trained model
new_args = copy.deepcopy(args)
for key in args.keys():
if "i2r" in key:
new_key = key.replace("i2r_", "")
print(key, new_key)
new_args.pop(key)
new_args[new_key] = args[key]
mod = Module(symbol=sym, **kwargs)
mod._arg_params = new_args
mod._aux_params = auxs
mod.params_initialized = True
if load_optimizer_states:
mod._preload_opt_states = "%s-%04d.states" % (prefix, epoch)
return mod
def load(prefix,
epoch,
load_optimizer_states=False,
symbol=None,
**kwargs):
"""Creates a model from previously saved checkpoint.
Parameters
----------
prefix : str
path prefix of saved model files. You should have
"prefix-symbol.json", "prefix-xxxx.params", and
optionally "prefix-xxxx.states", where xxxx is the
epoch number.
epoch : int
epoch to load.
load_optimizer_states : bool
whether to load optimizer states. Checkpoint needs
to have been made with save_optimizer_states=True.
data_names : list of str
Default is `('data')` for a typical model used in image classification.
label_names : list of str
Default is `('softmax_label')` for a typical model used in image
classification.
logger : Logger
Default is `logging`.
context : Context or list of Context
Default is ``cpu()``.
work_load_list : list of number
Default ``None``, indicating uniform workload.
fixed_param_names: list of str
Default ``None``, indicating no network parameters are fixed.
"""
sym, args, auxs = load_checkpoint(prefix, epoch)
sym = sym if symbol is None else symbol
mod = CustomModule(symbol=sym, **kwargs)
mod._arg_params = args
mod._aux_params = auxs
mod.params_initialized = True
if load_optimizer_states:
mod._preload_opt_states = '%s-%04d.states' % (prefix, epoch)
return mod
def load(prefix, epoch, load_optimizer_states=False, symbol=None, **kwargs):
"""Creates a model from previously saved checkpoint.
Parameters
----------
prefix : str
path prefix of saved model files. You should have
"prefix-symbol.json", "prefix-xxxx.params", and
optionally "prefix-xxxx.states", where xxxx is the
epoch number.
epoch : int
epoch to load.
load_optimizer_states : bool
whether to load optimizer states. Checkpoint needs
to have been made with save_optimizer_states=True.
data_names : list of str
Default is `('data')` for a typical model used in image classification.
label_names : list of str
Default is `('softmax_label')` for a typical model used in image
classification.
logger : Logger
Default is `logging`.
context : Context or list of Context
Default is ``cpu()``.
work_load_list : list of number
Default ``None``, indicating uniform workload.
fixed_param_names: list of str
Default ``None``, indicating no network parameters are fixed.
"""
sym, args, auxs = load_checkpoint(prefix, epoch)
sym = sym if symbol is None else symbol
mod = CustomModule(symbol=sym, **kwargs)
mod._arg_params = args
mod._aux_params = auxs
mod.params_initialized = True
if load_optimizer_states:
mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
return mod
download(model_url, model_file_path)
download(image_url, test_image_path)
download(inference_symbol_url, inference_symbol_path)
zip_ref = zipfile.ZipFile(model_file_path, 'r')
zip_ref.extractall(dir)
zip_ref.close()
zip_ref = zipfile.ZipFile(inference_symbol_path)
zip_ref.extractall(dir)
zip_ref.close()
######################################################################
# Convert and compile model with NNVM or Relay for CPU.
sym = mx.sym.load("%s/%s/ssd_resnet50_inference.json" % (dir, inference_symbol_folder))
_, arg_params, aux_params = load_checkpoint("%s/%s" % (dir, model_name), 0)
import argparse
parser = argparse.ArgumentParser()
parser.add_argument(
"-f", "--frontend",
help="Frontend for compilation, nnvm or relay",
type=str,
default="nnvm")
args = parser.parse_args()
if args.frontend == "relay":
net, params = relay.frontend.from_mxnet(sym, {"data": dshape}, arg_params=arg_params, aux_params=aux_params)
with relay.build_config(opt_level=3):
graph, lib, params = relay.build(net, target, params=params)
elif args.frontend == "nnvm":
net, params = from_mxnet(sym, arg_params, aux_params)
download(model_url, model_file_path)
download(image_url, test_image_path)
download(inference_symbol_url, inference_symbol_path)
zip_ref = zipfile.ZipFile(model_file_path, 'r')
zip_ref.extractall(dir)
zip_ref.close()
zip_ref = zipfile.ZipFile(inference_symbol_path)
zip_ref.extractall(dir)
zip_ref.close()
######################################################################
# Convert and compile model with NNVM for CPU.
sym = mx.sym.load("%s/%s/ssd_resnet50_inference.json" % (dir, inference_symbol_folder))
_, arg_params, aux_params = load_checkpoint("%s/%s" % (dir, model_name), 0)
net, params = from_mxnet(sym, arg_params, aux_params)
with compiler.build_config(opt_level=3):
graph, lib, params = compiler.build(net, target, {"data": dshape}, params=params)
######################################################################
# Create TVM runtime and do inference
# Preprocess image
image = cv2.imread(test_image_path)
img_data = cv2.resize(image, (dshape[2], dshape[3]))
img_data = img_data[:, :, (2, 1, 0)].astype(np.float32)
img_data -= np.array([123, 117, 104])
img_data = np.transpose(np.array(img_data), (2, 0, 1))
img_data = np.expand_dims(img_data, axis=0)
# Build TVM runtime
inf_json = "deploy_ssd_inceptionv3_512/deploy_ssd_inceptionv3_512-symbol.json"
#inf_json = "deploy_ssd_vgg16_reduced_512/deploy_ssd_vgg16_reduced_512-symbol.json"
#inf_json = "deploy_ssd_vgg16_reduced_300/deploy_ssd_vgg16_reduced_300-symbol.json"
#inf_json = "deploy_ssd_mobilenet_512/deploy_ssd_mobilenet_512-symbol.json"
#inf_json = "deploy_ssd_mobilenet_608/deploy_ssd_mobilenet_608-symbol.json"
print("mx.sym.load: " + inf_json)
sym = mx.sym.load(inf_json)
#checkp = "deploy_ssd_resnet50_512/deploy_ssd_resnet50_512"
checkp = "deploy_ssd_inceptionv3_512/deploy_ssd_inceptionv3_512"
#checkp = "deploy_ssd_vgg16_reduced_512/deploy_ssd_vgg16_reduced_512"
#checkp = "deploy_ssd_vgg16_reduced_300/deploy_ssd_vgg16_reduced_300"
#checkp = "deploy_ssd_mobilenet_512/deploy_ssd_mobilenet_512"
#checkp = "deploy_ssd_mobilenet_608/deploy_ssd_mobilenet_608"
print("load_checkpoint: " + checkp)
_, arg_params, aux_params = load_checkpoint(checkp, 0)
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("-f",
"--frontend",
help="Frontend for compilation, nnvm or relay",
type=str,
default="nnvm")
args = parser.parse_args()
if args.frontend == "relay":
net, params = relay.frontend.from_mxnet(sym, {"data": dshape}, arg_params=arg_params, \
aux_params=aux_params)
with relay.build_config(opt_level=3):
graph, lib, params = relay.build(net,
target,
def test_ssd():
model_name = "ssd_resnet50_512"
model_file = "%s.zip" % model_name
test_image = "dog.jpg"
dshape = (1, 3, 512, 512)
######################################################################
# Download MXNet SSD pre-trained model and demo image
# ---------------------------------------------------
# Pre-trained model available at
# https://github.com/apache/incubator-\mxnet/tree/master/example/ssd
model_url = "https://github.com/zhreshold/mxnet-ssd/releases/download/v0.6/" \
"resnet50_ssd_512_voc0712_trainval.zip"
image_url = "https://cloud.githubusercontent.com/assets/3307514/20012567/" \
"cbb60336-a27d-11e6-93ff-cbc3f09f5c9e.jpg"
inference_symbol_folder = "c1904e900848df4548ce5dfb18c719c7-a28c4856c827fe766aa3da0e35bad41d44f0fb26"
inference_symbol_url = "https://gist.github.com/kevinthesun/c1904e900848df4548ce5dfb18c719c7/" \
"archive/a28c4856c827fe766aa3da0e35bad41d44f0fb26.zip"
dir = "ssd_model"
if not os.path.exists(dir):
os.makedirs(dir)
model_file_path = "%s/%s" % (dir, model_file)
test_image_path = "%s/%s" % (dir, test_image)
inference_symbol_path = "%s/inference_model.zip" % dir
download(model_url, model_file_path)
download(image_url, test_image_path)
download(inference_symbol_url, inference_symbol_path)
zip_ref = zipfile.ZipFile(model_file_path, 'r')
zip_ref.extractall(dir)
zip_ref.close()
zip_ref = zipfile.ZipFile(inference_symbol_path)
zip_ref.extractall(dir)
zip_ref.close()
######################################################################
# Convert and compile model with NNVM for CPU.
sym = mx.sym.load("%s/%s/ssd_resnet50_inference.json" %
(dir, inference_symbol_folder))
_, arg_params, aux_params = load_checkpoint(
"%s/%s" % (dir, model_name), 0)
net, params = from_mxnet(sym, arg_params, aux_params)
shape_dict = {"data": dshape}
with nnvm.compiler.build_config(opt_level=3):
image, tvm_output = heterogeneous_ssd(net, ['nms'],
shape_dict,
params, test_image_path)
#####################################################################
# Display result
class_names = ["aeroplane", "bicycle", "bird", "boat", "bottle", "bus",
"car", "cat", "chair",
"cow", "diningtable", "dog", "horse", "motorbike",
"person", "pottedplant",
"sheep", "sofa", "train", "tvmonitor"]
def display(img, out, thresh=0.5):
import random
import matplotlib as mpl
import matplotlib.pyplot as plt
mpl.rcParams['figure.figsize'] = (10, 10)
pens = dict()
plt.clf()
plt.imshow(img)
for det in out:
cid = int(det[0])
if cid < 0:
continue
score = det[1]
if score < thresh:
continue
if cid not in pens:
pens[cid] = (random.random(),
random.random(), random.random())
scales = [img.shape[1], img.shape[0]] * 2
xmin, ymin, xmax, ymax = [
int(p * s) for p, s in zip(det[2:6].tolist(), scales)]
rect = plt.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin,
fill=False,
edgecolor=pens[cid], linewidth=3)
plt.gca().add_patch(rect)
text = class_names[cid]
plt.gca().text(xmin, ymin - 2,
'{:s} {:.3f}'.format(text, score),
bbox=dict(facecolor=pens[cid], alpha=0.5),
fontsize=12, color='white')
plt.show()
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
display(image, tvm_output.asnumpy()[0], thresh=0.45)
