From 7cda65423e7a59fe19530736265c6f12ec235892 Mon Sep 17 00:00:00 2001
From: "Daniel.Frisinghelli" <daniel.frisinghelli@eurac.edu>
Date: Thu, 30 Jul 2020 10:40:17 +0200
Subject: [PATCH] Removed option to only predict some samples; not required
---
pysegcnn/core/predict.py | 57 ++++++++++++++--------------------------
pysegcnn/main/config.py | 6 -----
pysegcnn/main/eval.py | 3 ---
3 files changed, 19 insertions(+), 47 deletions(-)
diff --git a/pysegcnn/core/predict.py b/pysegcnn/core/predict.py
index 874d978..fe9a33e 100644
--- a/pysegcnn/core/predict.py
+++ b/pysegcnn/core/predict.py
@@ -25,9 +25,12 @@ def get_scene_tiles(ds, scene_id):
return indices
-def predict_samples(ds, model, optimizer, state_path, state_file, nsamples,
- seed, batch_size=None, cm=False, plot_samples=False,
- **kwargs):
+def predict_samples(ds, model, optimizer, state_path, state_file, cm=False,
+ plot=False, **kwargs):
+
+ # check whether the dataset is a subset
+ if not isinstance(ds, Subset):
+ raise TypeError('ds should be of type {}'.format(Subset))
# the device to compute on, use gpu if available
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
@@ -49,26 +52,13 @@ def predict_samples(ds, model, optimizer, state_path, state_file, nsamples,
# initialize confusion matrix
cmm = np.zeros(shape=(model.nclasses, model.nclasses))
- # set random seed for reproducibility
- np.random.seed(seed)
-
- # draw a number of samples from the dataset
- samples = np.arange(0, len(ds))
- if nsamples > 0:
- batch_size = nsamples
- samples = np.random.choice(samples, size=min(nsamples, len(ds)))
-
- # create a subset of the dataset
- smpl_subset = Subset(ds, samples.tolist())
- if batch_size is None:
- raise ValueError('If you specify "nsamples"=-1, you have to provide '
- 'a batch size, e.g. trainer.batch_size.')
- smpl_loader = DataLoader(smpl_subset, batch_size=batch_size, shuffle=False)
+ # create the dataloader
+ dataloader = DataLoader(ds, batch_size=1, shuffle=False, drop_last=False)
# iterate over the samples and plot inputs, ground truth and
# model predictions
output = {}
- for batch, (inputs, labels) in enumerate(smpl_loader):
+ for batch, (inputs, labels) in enumerate(dataloader):
# send inputs and labels to device
inputs = inputs.to(device)
@@ -87,22 +77,14 @@ def predict_samples(ds, model, optimizer, state_path, state_file, nsamples,
cmm[ytrue.long(), ypred.long()] += 1
# save plot of current batch to disk
- if plot_samples:
-
- # check whether the dataset is a subset
- if isinstance(ds, Subset):
- use_bands = ds.dataset.use_bands
- ds_labels = ds.dataset.labels
- else:
- use_bands = ds.use_bands
- ds_labels = ds.labels
+ if plot:
# plot inputs, ground truth and model predictions
- sname = fname + '_sample_{}.pt'.format(batch)
+ sname = fname + '_{}_{}.pt'.format(ds.name, batch)
fig, ax = plot_sample(inputs.numpy().clip(0, 1),
labels,
- use_bands,
- ds_labels,
+ ds.dataset.use_bands,
+ ds.dataset.labels,
y_pred=prd,
state=sname,
**kwargs)
@@ -113,6 +95,10 @@ def predict_samples(ds, model, optimizer, state_path, state_file, nsamples,
def predict_scenes(ds, model, optimizer, state_path, state_file,
scene_id=None, cm=False, plot_scenes=False, **kwargs):
+ # check if the dataset is an instance of torch.data.dataset.Subset
+ if not isinstance(ds, Subset):
+ raise TypeError('ds should be of type {}'.format(Subset))
+
# the device to compute on, use gpu if available
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
@@ -136,12 +122,6 @@ def predict_scenes(ds, model, optimizer, state_path, state_file,
# check whether a scene id is provided
if scene_id is None:
- # check if the dataset is an instance of torch.data.dataset.Subset
- if not isinstance(ds, Subset):
- raise TypeError('ds should be of type {}'.format(Subset))
-
- print('Predicting scenes of the subset ...')
-
# get the names of the scenes
try:
scene_ids = ds.ids
@@ -158,6 +138,7 @@ def predict_scenes(ds, model, optimizer, state_path, state_file,
scene_size = (ds.dataset.height, ds.dataset.width)
# iterate over the scenes
+ print('Predicting scenes of the subset ...')
scene = {}
for sid in scene_ids:
@@ -173,7 +154,7 @@ def predict_scenes(ds, model, optimizer, state_path, state_file,
# create the dataloader
scene_dl = DataLoader(scene_ds, batch_size=len(scene_ds),
- shuffle=False)
+ shuffle=False, drop_last=False)
# predict the current scene
for i, (inp, lab) in enumerate(scene_dl):
diff --git a/pysegcnn/main/config.py b/pysegcnn/main/config.py
index d453fa8..61bdaaa 100644
--- a/pysegcnn/main/config.py
+++ b/pysegcnn/main/config.py
@@ -261,12 +261,6 @@ config = {
# split_mode="date"
'predict_scene': True,
- # number of samples to validate model performance on
- # if nsamples': -1, the model is evaluated on all samples of the validation
- # set or test set
- # only takes effect if predict_scene=False
- 'nsamples': -1,
-
# whether to save plots of (input, ground truth, prediction) of the
# samples from the validation/test dataset to disk
# output path is: pysegcnn/main/_samples/
diff --git a/pysegcnn/main/eval.py b/pysegcnn/main/eval.py
index f082c35..85a5351 100644
--- a/pysegcnn/main/eval.py
+++ b/pysegcnn/main/eval.py
@@ -49,9 +49,6 @@ if __name__ == '__main__':
trainer.optimizer,
trainer.state_path,
trainer.state_file,
- trainer.nsamples,
- trainer.seed,
- trainer.batch_size,
trainer.cm,
trainer.plot_samples,
bands=trainer.plot_bands,
--
GitLab