import os
import pandas as pd
import glob
import numpy as np
from datetime import datetime
from DeepForest.config import load_config
from DeepForest import preprocess
#save time for logging
dirname = datetime.now().strftime("%Y%m%d_%H%M%S")
#Load DeepForest_config and data file based on training or retraining mode
DeepForest_config = load_config("train")
data = preprocess.load_data(DeepForest_config["training_csvs"],
DeepForest_config["rgb_res"],
DeepForest_config["lidar_path"])
#Log site
site = DeepForest_config["evaluation_site"]
##Preprocess Filters##
if DeepForest_config['preprocess']['zero_area']:
data = preprocess.zero_area(data)
#pass an args object instead of using command line
args = [
"--epochs",
str(DeepForest_config["epochs"]), "--batch-size",
str(DeepForest_config['batch_size']), "--backbone",
str(DeepForest_config["backbone"]), "--score-threshold",
#parse logs
parser = argparse.ArgumentParser()
parser.add_argument("--logdir",
help="",
type=str,
default="20180523-181550/")
parser.add_argument(
"--box_file",
help="",
type=str,
default=
"data/bounding_boxes_NEON_D03_OSBS_DP1_398000_3280000_classified_point_cloud_laz.csv"
)
args = parser.parse_args()
#load model
model = load_model(args.logdir)
#load data
data = preprocess.load_data(args.box_file)
#predict
preds, labels = predict(model, data)
#confusion matrix
tn, fp, fn, tp = calculate_confusion(labels, preds)
print(
"True Negative Rate %.3f\nTrue Positive Rate %.3f\nFalse Negative Rate %.3f\nFalse Positive Rate %.3f"
% (tn, tp, fn, fp))
from DeepForest.config import config
from DeepForest import preprocess
import random
batch_size = config['batch_size']
#set experiment and log configs
experiment = Experiment(api_key="ypQZhYfs3nSyKzOfz13iuJpj2",
project_name='deepforest-retinanet')
##Set seed for reproducibility##
np.random.seed(2)
#Load data and combine into a large frame
#Training
data = preprocess.load_data(data_dir=config['training_csvs'])
#Evaluation
evaluation = preprocess.load_data(data_dir=config['evaluation_csvs'])
##Preprocess Filters##
if config['preprocess']['zero_area']:
data = preprocess.zero_area(data)
evaluation = preprocess.zero_area(evaluation)
#Write training and evaluation data to file for annotations
data.to_csv("data/training/detection.csv")
evaluation.to_csv("data/training/evaluation.csv")
#log data size and set number of steps
def sample(n=50):
"""
Grab n random images from across the site
"""
#Load config
DeepForest_config = config.load_config()
#Read in data
data = preprocess.load_data(data_dir=DeepForest_config['training_csvs'],
res=0.1,
lidar_path=DeepForest_config["lidar_path"])
#Create windows
windows = preprocess.create_windows(
data,
DeepForest_config,
base_dir=DeepForest_config["evaluation_tile_dir"])
selected_windows = windows[["tile",
"window"]].drop_duplicates().sample(n=n)
generator = onthefly_generator.OnTheFlyGenerator(
data=data,
windowdf=selected_windows,
DeepForest_config=DeepForest_config)
folder_dir = os.path.join("data", DeepForest_config["evaluation_site"],
"samples")
if not os.path.exists(folder_dir):
os.mkdir(folder_dir)
for i in range(generator.size()):
#Load image - done for side effects, allow to skip bad tiles.
try:
three_channel = generator.load_image(i)
except:
continue
#load lidar
generator.load_lidar_tile()
generator.clip_las()
if generator.clipped_las == None:
continue
#name RGB
tilename = os.path.splitext(generator.image_data[i]["tile"])[0]
tilename = tilename + "_" + str(
generator.image_data[i]["window"]) + ".tif"
filename = os.path.join(folder_dir, tilename)
#Write
cv2.imwrite(filename, generator.image)
#name .laz
tilename = os.path.splitext(generator.image_data[i]["tile"])[0]
tilename = tilename + "_" + str(
generator.image_data[i]["window"]) + ".laz"
filename = os.path.join(folder_dir, tilename)
#Write .laz
generator.clipped_las.write(filename)
from DeepForest import preprocess, evaluate
from models import inception
batch_size = config['data_generator_params']['batch_size']
#set experiment and log configs
experiment = Experiment(api_key="ypQZhYfs3nSyKzOfz13iuJpj2",
project_name='deepforest')
experiment.log_multiple_params(config['data_generator_params'])
experiment.log_multiple_params(config['training'])
##Set seed for reproducibility##
np.random.seed(2)
#Load data and comine into a large
data = preprocess.load_data(data_dir=config['bbox_data_dir'],
nsamples=config["subsample"])
##Preprocess Filters##
data = preprocess.zero_area(data)
#Partition data in training and testing dataframes
msk = np.random.rand(len(data)) < 0.8
train = data[msk]
test = data[~msk]
#log data size
experiment.log_parameter("training_samples", train.shape[0])
experiment.log_parameter("testing_samples", test.shape[0])
#Create dictionaries to keep track of labels and splits
partition = {"train": train.index.values, "test": test.index.values}