# In[13]:
gtf.Model();
# In[14]:
gtf.Set_Hyperparams(lr=0.0001, val_interval=1, es_min_delta=0.0, es_patience=0)
# In[ ]:
gtf.Train(num_epochs=30, model_output_dir="trained/");
# In[ ]:
# In[2]:
# import sys
# sys.path.append("Monk_Object_Detection/4_efficientdet/lib/");
# from src.dataset import CocoDataset
# root_dir = "coco_dataset_3class";
gtf = Detector()
gtf.Train_Dataset(root_dir,
coco_dir,
img_dir,
set_dir,
batch_size=system["batch_size"],
image_size=system["image_size"],
use_gpu=system["use_gpu"])
if (system["val_data"] == "yes"):
gtf.Val_Dataset(val_root_dir, val_coco_dir, val_img_dir, val_set_dir)
tmp = system["devices"].split(",")
gpu_devices = []
for i in range(len(tmp)):
gpu_devices.append(int(tmp[i]))
gtf.Model(gpu_devices=gpu_devices)
gtf.Set_Hyperparams(lr=system["lr"],
val_interval=system["val_interval"],
es_min_delta=system["es_min_delta"],
es_patience=system["es_patience"])
gtf.Train(num_epochs=system["epochs"],
model_output_dir=system["output_model_dir"])
print("Completed")
system["val_interval"] = int(system["val_interval"])
system["lr"] = float(system["lr"])
gtf = Detector()
gtf.Train_Dataset(root_dir,
coco_dir,
img_dir,
set_dir,
batch_size=system["batch_size"],
use_gpu=system["use_gpu"])
if (system["val_data"] == "yes"):
gtf.Val_Dataset(val_root_dir, val_coco_dir, val_img_dir, val_set_dir)
tmp = system["devices"].split(",")
gpu_devices = []
for i in range(len(tmp)):
gpu_devices.append(int(tmp[i]))
gtf.Model(model_name=system["model"], gpu_devices=gpu_devices)
gtf.Set_Hyperparams(lr=system["lr"],
val_interval=system["val_interval"],
print_interval=system["print_interval"])
gtf.Train(num_epochs=system["epochs"],
output_model_name=system["output_model_name"] + ".pt")
print("Completed")
timg_dir = "images";
tset_dir = "Train";
vroot_dir = "Root_Dir";
vcoco_dir = "Coco_style";
vimg_dir = "images";
vset_dir = "Val";
model.Train_Dataset(troot_dir, tcoco_dir, timg_dir, tset_dir, batch_size=8, image_size=352, use_gpu=True)
model.Val_Dataset(vroot_dir, vcoco_dir, vimg_dir, vset_dir)
model.Model(model_name="resnet34"); # resnet 50 brought cuda memory error.
model.Set_Hyperparams(lr=0.0001, val_interval=1, print_interval=20)
model.Train(num_epochs=300,output_model_name="karen_model.pt");
from infer_detector import Infer
gtf = Infer();
gtf.Model(model_path="/content/karen_model.pt");
#predictions are quite bad at the moment.
class_list=[]
with open("/content/Root_Dir/Coco_style/annotations/classes.txt") as file:
for line in file:
class_list.append(line.rstrip("\n"))
class_list=class_list[:-1]
img_p="/content/Images_and_Labels/Images/0000002_00005_d_0000014_jpg.rf.555bf2106d899e56d45da0a48295f04c.jpg"
scores, labels, boxes = gtf.Predict(img_p, class_list, vis_threshold=0.4);
from IPython.display import Image
Image(filename='output.jpg')
gtf = Detector();
gtf.Train_Dataset(root_dir, coco_dir, img_dir, set_dir,
batch_size=system["batch_size"],
num_workers=4)
if(system["val_data"] == "yes"):
gtf.Val_Dataset(val_root_dir, val_coco_dir, val_img_dir, val_set_dir)
gtf.Model(model_name=system["model"]);
gtf.Hyper_Params(lr=system["lr"],
total_iterations=system["iterations"],
val_interval=system["val_interval"])
gtf.Setup();
gtf.Train(display_interval=system["print_interval"]);
print("Completed");
val_classes_file = system["val_yolo_classes_file"]
from train_detector import Detector
gtf = Detector()
gtf.set_train_dataset(root_dir + "/" + img_dir,
root_dir + "/" + anno_dir,
root_dir + "/" + classes_file,
batch_size=system["batch_size"],
img_size=system["img_size"],
cache_images=system["cache_images"])
if (system["val_data"] == "yes"):
gtf.set_val_dataset(val_root_dir + "/" + val_img_dir,
val_root_dir + "/" + val_anno_dir)
gtf.set_model(model_name=system["model"])
gtf.set_hyperparams(optimizer=system["optimizer"],
lr=system["lr"],
multi_scale=system["multi_scale"],
evolve=system["evolve"],
num_generations=system["num_generations"],
mixed_precision=system["mixed_precision"],
gpu_devices=system["devices"])
gtf.Train(num_epochs=system["epochs"])
print("Completed")