from datascience.ml.neural.models import load_create_nn, InceptionEnv
from datascience.data.loader import occurrence_loader
from datascience.data.datasets import EnvironmentalDataset
from datascience.ml.neural.supervised import fit
from sklearn.model_selection import train_test_split
from projects.ecography.configs.inception import model_params, training_params
# loading/creating model
model = load_create_nn(model_class=InceptionEnv, model_params=model_params)
# loading dataset
train, val, test = occurrence_loader(EnvironmentalDataset,
source='gbif_taxref',
splitter=train_test_split)
# training model
validation_params = {
'metrics':
(ValidationAccuracyMultipleBySpecies([1, 10,
30]), ValidationMRRBySpecies(),
ValidationAccuracyRangeBySpecies(max_top_k=100, final_validation=True),
ValidationAccuracyForAllSpecies(train=train, final_validation=True))
}
fit(model,
train=train,
val=val,
test=test,
training_params=training_params,
validation_params=validation_params)
training_params = {
'iterations': [120],
'log_modulo': -1,
'val_modulo': 1,
'loss': HebbLoss()
}
optim_params = {
'momentum': 0.0,
'lr': 0.1,
}
validation_params = {'metrics': (ValidationAccuracy(1), )}
fit(model,
train=train,
test=test,
training_params=training_params,
optim_params=optim_params,
validation_params=validation_params)
# plot results
ax = plot_db_partitions(train.dataset, train.labels,
model) # plot_db_partitions_gradients
plot_separator(train.separator, ax=ax)
remove_axis()
save_fig()
training_params = {
'iterations': [100, 130, 150, 160],
'batch_size': 256,
}
optim_params = {'lr': 0.001}
validation_params = {'metrics': (ValidationAccuracy(1), )}
model_selection_params = {'cross_validation': True, 'min_epochs': 50}
stats = fit(model,
train=train,
val=val,
test=test,
training_params=training_params,
validation_params=validation_params,
optim_params=optim_params,
model_selection_params=model_selection_params)
score = stats.final_metric().metric_score()
score = score if test[0][1] == 1. else 1. - score
# write the score in a csv
with open(export_result, 'a') as f:
f.write('%s;%s;%.5f;%ld\n' %
(painter_val, painter_test, score, test[0][1]))
del stats
del model
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
'test': transforms.Compose([
transforms.Resize(input_size),
transforms.CenterCrop(input_size),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
}
train, test = cifar10(transform)
training_params = {
'iterations': [50, 75, 100],
'batch_size': 256,
}
optim_params = {
'lr': 0.001
}
validation_params = {
'metrics': (ValidationAccuracy(1),)
}
fit(
model, train=train, test=test, training_params=training_params, validation_params=validation_params,
optim_params=optim_params, cross_validation=True
)