diff --git a/pysegcnn/core/logging.py b/pysegcnn/core/logging.py
new file mode 100644
index 0000000000000000000000000000000000000000..4f42dc0535a83d6ff2e56c72abc7be273e588dc2
--- /dev/null
+++ b/pysegcnn/core/logging.py
@@ -0,0 +1,47 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Fri Aug 14 10:07:12 2020
+
+@author: Daniel
+"""
+
+
+# the logging configuration dictionary
+def log_conf(logfile):
+
+ LOGGING_CONFIG = {
+ 'version': 1,
+ 'disable_existing_loggers': False,
+ 'formatters': {
+ 'brief': {
+ 'format': '%(name)s: %(message)s'
+ },
+ 'standard': {
+ 'format': '%(asctime)s [%(levelname)s] %(name)s: %(message)s'
+ },
+ },
+ 'handlers': {
+ 'console': {
+ 'class': 'logging.StreamHandler',
+ 'formatter': 'brief',
+ 'level': 'INFO',
+ 'stream': 'ext://sys.stderr',
+ },
+
+ 'file': {
+ 'class': 'logging.FileHandler',
+ 'formatter': 'standard',
+ 'level': 'INFO',
+ 'filename': logfile,
+ 'mode': 'w'
+ }
+ },
+ 'loggers': {
+ '': {
+ 'handlers': ['console', 'file'],
+ 'level': 'INFO',
+ },
+ }
+ }
+
+ return LOGGING_CONFIG
diff --git a/pysegcnn/core/models.py b/pysegcnn/core/models.py
index 6d840fc67393d86b0bef7086c27561feb96aee8a..2d5cc84097f4b0bea00d565c1460581823a67092 100644
--- a/pysegcnn/core/models.py
+++ b/pysegcnn/core/models.py
@@ -8,6 +8,7 @@ Created on Fri Jun 26 16:31:36 2020
# builtins
import os
import enum
+import logging
# externals
import numpy as np
@@ -19,6 +20,9 @@ import torch.optim as optim
from pysegcnn.core.layers import (Encoder, Decoder, Conv2dPool, Conv2dUnpool,
Conv2dUpsample, Conv2dSame)
+# module level logger
+LOGGER = logging.getLogger(__name__)
+
class Network(nn.Module):
@@ -43,7 +47,10 @@ class Network(nn.Module):
# initialize dictionary to store network parameters
model_state = {}
- # store input bands
+ # store model name
+ model_state['cls'] = self.__class__
+
+ # store the bands the model was trained with
model_state['bands'] = bands
# store construction parameters to instanciate the network
@@ -67,7 +74,7 @@ class Network(nn.Module):
# model state dictionary stores the values of all trainable parameters
state = os.path.join(outpath, state_file)
torch.save(model_state, state)
- print('Network parameters saved in {}'.format(state))
+ LOGGER.info('Network parameters saved in {}'.format(state))
return state
@@ -79,13 +86,13 @@ class Network(nn.Module):
model_state = torch.load(state)
# resume network parameters
- print('Loading network parameters from {} ...'.format(state))
+ LOGGER.info('Loading model parameters ...'.format(state))
self.load_state_dict(model_state['model_state_dict'])
self.epoch = model_state['epoch']
# resume optimizer parameters
if optimizer is not None:
- print('Loading optimizer parameters from {} ...'.format(state))
+ LOGGER.info('Loading optimizer parameters ...'.format(state))
optimizer.load_state_dict(model_state['optim_state_dict'])
return state
diff --git a/pysegcnn/core/predict.py b/pysegcnn/core/predict.py
index 01b07cfb80e0636029058e49d356512740ea6f8d..ae539052e8177b6ca63f8c43d3ca4dcf7c33e812 100644
--- a/pysegcnn/core/predict.py
+++ b/pysegcnn/core/predict.py
@@ -1,6 +1,7 @@
# builtins
import os
import pathlib
+import logging
# externals
import numpy as np
@@ -14,6 +15,9 @@ from pysegcnn.core.utils import reconstruct_scene, accuracy_function
from pysegcnn.core.graphics import plot_sample
from pysegcnn.core.split import RandomSubset, SceneSubset
+# module level logger
+LOGGER = logging.getLogger(__name__)
+
def get_scene_tiles(ds, scene_id):
@@ -33,7 +37,6 @@ def predict_samples(ds, model, optimizer, state_file, cm=False,
# check whether the dataset is a valid subset, i.e.
# an instance of pysegcnn.core.split.SceneSubset or
# an instance of pysegcnn.core.split.RandomSubset
- _name = type(ds).__name__
if not isinstance(ds, RandomSubset) or not isinstance(ds, SceneSubset):
raise TypeError('ds should be an instance of {} or of {}.'
.format(repr(RandomSubset), repr(SceneSubset)))
@@ -50,7 +53,7 @@ def predict_samples(ds, model, optimizer, state_file, cm=False,
_ = model.load(state_file.name, optimizer, state_file.parent)
# set the model to evaluation mode
- print('Setting model to evaluation mode ...')
+ LOGGER.info('Setting model to evaluation mode ...')
model.eval()
model.to(device)
@@ -66,7 +69,7 @@ def predict_samples(ds, model, optimizer, state_file, cm=False,
# iterate over the samples and plot inputs, ground truth and
# model predictions
output = {}
- print('Predicting samples of the {} dataset ...'.format(ds.name))
+ LOGGER.info('Predicting samples of the {} dataset ...'.format(ds.name))
for batch, (inputs, labels) in enumerate(dataloader):
# send inputs and labels to device
@@ -80,9 +83,8 @@ def predict_samples(ds, model, optimizer, state_file, cm=False,
# store output for current batch
output[batch] = {'input': inputs, 'labels': labels, 'prediction': prd}
- print('Sample: {:d}/{:d}, Accuracy: {:.2f}'
- .format(batch + 1, len(dataloader),
- accuracy_function(prd, labels)))
+ LOGGER.info('Sample: {:d}/{:d}, Accuracy: {:.2f}'.format(
+ batch + 1, len(dataloader), accuracy_function(prd, labels)))
# update confusion matrix
if cm:
@@ -126,7 +128,7 @@ def predict_scenes(ds, model, optimizer, state_file,
_ = model.load(state_file.name, optimizer, state_file.parent)
# set the model to evaluation mode
- print('Setting model to evaluation mode ...')
+ LOGGER.info('Setting model to evaluation mode ...')
model.eval()
model.to(device)
@@ -147,7 +149,7 @@ def predict_scenes(ds, model, optimizer, state_file,
scene_size = (ds.dataset.height, ds.dataset.width)
# iterate over the scenes
- print('Predicting scenes of the {} dataset ...'.format(ds.name))
+ LOGGER.info('Predicting scenes of the {} dataset ...'.format(ds.name))
scenes = {}
for i, sid in enumerate(scene_ids):
@@ -187,7 +189,7 @@ def predict_scenes(ds, model, optimizer, state_file,
prdtcn = reconstruct_scene(prd, scene_size, nbands=1)
# print progress
- print('Scene {:d}/{:d}, Id: {}, Accuracy: {:.2f}'.format(
+ LOGGER.info('Scene {:d}/{:d}, Id: {}, Accuracy: {:.2f}'.format(
i + 1, len(scene_ids), sid, accuracy_function(prdtcn, labels)))
# save outputs to dictionary
diff --git a/pysegcnn/core/utils.py b/pysegcnn/core/utils.py
index 617aff19a1dd9854991296076a8308e5b4718a5d..c2c1d5d0042dfd1c2ad19398e2ad96a406536d6a 100644
--- a/pysegcnn/core/utils.py
+++ b/pysegcnn/core/utils.py
@@ -7,6 +7,7 @@ Created on Tue Jul 14 15:02:23 2020
# builtins
import os
import re
+import logging
import datetime
# externals
@@ -17,6 +18,9 @@ import numpy as np
# the following functions are utility functions for common image
# manipulation operations
+# module level logger
+LOGGER = logging.getLogger(__name__)
+
# this function reads an image to a numpy array
def img2np(path, tile_size=None, tile=None, pad=False, cval=0, verbose=False):
@@ -36,7 +40,7 @@ def img2np(path, tile_size=None, tile=None, pad=False, cval=0, verbose=False):
width = img.RasterXSize
elif path is None:
- print('Path is of NoneType, returning.')
+ LOGGER.warning('Path is of NoneType, returning.')
return
# accept numpy arrays as input
@@ -81,13 +85,12 @@ def img2np(path, tile_size=None, tile=None, pad=False, cval=0, verbose=False):
y_size = height + padding[0] + padding[2]
x_size = width + padding[1] + padding[3]
- if verbose:
- print('Image size: {}'.format((height, width)))
- print('Dividing image into {} tiles of size {} ...'
- .format(ntiles, (tile_size, tile_size)))
- print('Padding image: bottom = {}, left = {}, top = {}, right = {}'
- ' ...'.format(*list(padding)))
- print('Padded image size: {}'.format((y_size, x_size)))
+ # print progress
+ LOGGER.debug('Image size: {}'.format((height, width)))
+ LOGGER.debug('Dividing image into {} tiles of size {} ...'
+ .format(ntiles, (tile_size, tile_size)))
+ LOGGER.debug('Padding image (b, l, t, r): {}'.format(tuple(padding)))
+ LOGGER.debug('Padded image size: {}'.format((y_size, x_size)))
# get the indices of the top left corner for each tile
topleft = tile_topleft_corner((y_size, x_size), tile_size)
@@ -102,21 +105,18 @@ def img2np(path, tile_size=None, tile=None, pad=False, cval=0, verbose=False):
# iterate over the topleft corners of the tiles
for k, corner in topleft.items():
- if verbose:
- print('Creating tile {} with top-left corner {} ...'
- .format(k, corner))
-
# in case a single tile is required, skip the rest of the tiles
if tile is not None:
if k != tile:
- if verbose:
- print('Skipping tile {} ...'.format(k))
continue
+
# set the key to 0 for correct array indexing when reading
# a single tile from the image
- if verbose:
- print('Processing tile {} ...'.format(k))
+ LOGGER.debug('Processing tile {} ...'.format(k))
k = 0
+ else:
+ LOGGER.debug('Creating tile {} with top-left corner {} ...'
+ .format(k, corner))
# calculate shift between padded and original image
row = corner[0] - padding[2] if corner[0] > 0 else corner[0]