def test_creates_probabilities(self):
activations_model = pytorch_custom()
brain_model = ModelCommitment(identifier=activations_model.identifier,
activations_model=activations_model,
layers=None,
behavioral_readout_layer='relu2')
fitting_stimuli = StimulusSet({
'image_id': ['rgb1', 'rgb2'],
'image_label': ['label1', 'label2']
})
fitting_stimuli.image_paths = {
'rgb1': os.path.join(os.path.dirname(__file__), 'rgb1.jpg'),
'rgb2': os.path.join(os.path.dirname(__file__), 'rgb2.jpg')
}
fitting_stimuli.name = 'test_probabilities_mapping.creates_probabilities'
brain_model.start_task(BrainModel.Task.probabilities, fitting_stimuli)
probabilities = brain_model.look_at(fitting_stimuli)
np.testing.assert_array_equal(probabilities.dims,
['presentation', 'choice'])
np.testing.assert_array_equal(probabilities.shape, [2, 2])
np.testing.assert_array_almost_equal(
probabilities.sel(image_id='rgb1', choice='label1').values,
probabilities.sel(image_id='rgb2', choice='label2').values)
assert probabilities.sel(image_id='rgb1', choice='label1') + \
probabilities.sel(image_id='rgb1', choice='label2') == approx(1)
def test_model(self):
target_model_pool = pytorch_custom(28)
stimuli_model_pool = pytorch_custom(224)
search_target_model_param = {}
search_stimuli_model_param = {}
search_target_model_param['target_model'] = target_model_pool
search_stimuli_model_param['stimuli_model'] = stimuli_model_pool
search_target_model_param['target_layer'] = 'relu1'
search_stimuli_model_param['stimuli_layer'] = 'relu1'
search_target_model_param['target_img_size'] = 28
search_stimuli_model_param['search_image_size'] = 224
model = ModelCommitment(identifier=stimuli_model_pool.identifier, activations_model=None, layers=['relu1'], search_target_model_param=search_target_model_param, search_stimuli_model_param=search_stimuli_model_param)
assemblies = brainscore.get_assembly('klab.Zhang2018search_obj_array')
stimuli = assemblies.stimulus_set
fix = [[640, 512],
[365, 988],
[90, 512],
[365, 36],
[915, 36],
[1190, 512],
[915, 988]]
max_fix = 6
data_len = 300
model.start_task(BrainModel.Task.visual_search_obj_arr, fix=fix, max_fix=max_fix, data_len=data_len)
cumm_perf, saccades = model.look_at(stimuli)
assert saccades.shape == (300, 8, 2)
assert cumm_perf.shape == (7, 2)
def load(identifier=basemodel_identifier, activations_model=activations_model, layers=layers):
assert hasattr(activations_model, 'reload')
activations_model.reload()
brain_model = ModelCommitment(identifier=identifier, activations_model=activations_model,
layers=layers)
for region in regions:
brain_model.commit_region(region)
return brain_model
def test_commit_record(self, model_ctr, layers, expected_layer, region):
activations_model = model_ctr()
brain_model = ModelCommitment(identifier=activations_model.identifier, activations_model=activations_model,
layers=layers)
brain_model.start_recording(region, [(70, 170)])
predictions = brain_model.look_at([Path(__file__).parent / 'rgb1.jpg'])
assert set(predictions['region'].values) == {region}
assert set(predictions['layer'].values) == {expected_layer}
def test_creates_synset(self, model_ctr):
np.random.seed(0)
activations_model = model_ctr()
brain_model = ModelCommitment(identifier=activations_model.identifier, activations_model=activations_model,
layers=None, behavioral_readout_layer='dummy') # not needed
stimuli = StimulusSet({'image_id': ['abc123']})
stimuli.image_paths = {'abc123': os.path.join(os.path.dirname(__file__), 'rgb1.jpg')}
stimuli.name = 'test_logits_behavior.creates_synset'
brain_model.start_task(BrainModel.Task.label, 'imagenet')
synsets = brain_model.look_at(stimuli)
assert len(synsets) == 1
assert synsets[0].startswith('n')
def check_processing(model, module):
os.environ['RESULTCACHING_DISABLE'] = '1'
model_instance = module.get_model(model)
layers = module.get_layers(model)
brain_model = ModelCommitment(identifier=model,
activations_model=model_instance,
layers=layers,
region_benchmarks={'IT': _MockBenchmark()})
brain_model.commit_region('IT')
benchmark = _MockBenchmark()
score = benchmark(brain_model, do_behavior=True)
assert score is not None
assert score.sel(aggregation='center')
def mobilenet_brainmodel(identifier, init_weights=True, function=None):
model = get_mobilenet(identifier, init_weights)
if identifier == 'mobilenet_v1_1.0_224':
lay = layers['mobilenet_v1']
else:
lay = layers['mobilenet_pytorch']
brain_model = ModelCommitment(identifier=identifier,
activations_model=model,
layers=lay)
return brain_model
def load(identifier=basemodel_identifier,
activations_model=activations_model,
layers=model_layer):
assert hasattr(activations_model, 'reload')
activations_model.reload()
brain_model = ModelCommitment(
identifier=identifier,
activations_model=activations_model,
layers=layers)
return brain_model
def hmax_brainmodel(identifier, init_weights=True, function=None):
model = get_hmax(identifier, 224)
if identifier.startswith('hmax_2'):
lay = layers['hmax_2']
elif identifier.startswith('hmax_3'):
lay = layers['hmax_3']
else:
lay = layers['hmax']
brain_model = ModelCommitment(identifier=identifier,
activations_model=model,
layers=lay)
return brain_model
def load(identifier=basemodel_identifier,
activations_model=activations_model,
layers=layers,
target_model=target_model,
stimuli_model=stimuli_model):
assert hasattr(activations_model, 'reload')
activations_model.reload()
search_target_model_param = {}
search_stimuli_model_param = {}
if (vs_model_param is not None) and (
identifier == 'vgg-16'
): #as vs_layer is implemented only for vgg-16 as of now
search_target_model_param['target_model'] = target_model
search_stimuli_model_param['stimuli_model'] = stimuli_model
search_target_model_param[
'target_layer'] = visual_search_layer[identifier][0]
search_stimuli_model_param[
'stimuli_layer'] = visual_search_layer[identifier][0]
search_target_model_param[
'target_img_size'] = target_img_size
search_stimuli_model_param[
'search_image_size'] = stimuli_img_size
else:
search_target_model_param['target_model'] = None
search_stimuli_model_param['stimuli_model'] = None
search_target_model_param['target_layer'] = None
search_stimuli_model_param['stimuli_layer'] = None
search_target_model_param['target_img_size'] = None
search_stimuli_model_param['search_image_size'] = None
brain_model = ModelCommitment(
identifier=identifier,
activations_model=activations_model,
layers=layers,
search_target_model_param=search_target_model_param,
search_stimuli_model_param=search_stimuli_model_param)
for region in regions:
brain_model.commit_region(region)
return brain_model
def test_creates_synset(self, model_ctr):
np.random.seed(0)
activations_model = model_ctr()
brain_model = ModelCommitment(
identifier=activations_model.identifier,
activations_model=activations_model,
layers=None,
behavioral_readout_layer='dummy') # not needed
stimuli = StimulusSet({
'image_id': ['1', '2'],
'filename': ['rgb1', 'rgb2']
})
stimuli.image_paths = {
'1': os.path.join(os.path.dirname(__file__), 'rgb1.jpg'),
'2': os.path.join(os.path.dirname(__file__), 'rgb2.jpg')
}
stimuli.identifier = 'test_logits_behavior.creates_synset'
brain_model.start_task(BrainModel.Task.label, 'imagenet')
behavior = brain_model.look_at(stimuli)
assert isinstance(behavior, BehavioralAssembly)
assert set(behavior['image_id'].values) == {'1', '2'}
assert len(behavior['synset']) == 2
assert behavior['synset'].values[0].startswith('n')
def get_vonenet_commitment(identifier,
activations_model,
layers,
visual_degrees=8,
stochastic=True):
if stochastic:
model_commitment = StochasticModelCommitment(
identifier=identifier,
activations_model=activations_model,
layers=layers,
visual_degrees=visual_degrees)
else:
model_commitment = ModelCommitment(identifier=identifier,
activations_model=activations_model,
layers=layers,
visual_degrees=visual_degrees)
model_commitment.layer_model.region_layer_map['V1'] = 'vone_block.output'
return model_commitment
def check_processing(model_identifier, module):
os.environ['RESULTCACHING_DISABLE'] = '1'
model_instance = module.get_model(model_identifier)
layers = module.get_layers(model_identifier)
benchmark = _MockBenchmark()
layer_selection = LayerSelection(model_identifier=model_identifier,
activations_model=model_instance,
layers=layers,
visual_degrees=8)
region_layer_map = RegionLayerMap(layer_selection=layer_selection,
region_benchmarks={'IT': benchmark})
brain_model = ModelCommitment(identifier=model_identifier,
activations_model=model_instance,
layers=layers,
region_layer_map=region_layer_map)
score = benchmark(brain_model, do_behavior=True)
assert score is not None
assert score.sel(aggregation='center')
def resnet_brainmodel(identifier, init_weights=True, function=None):
model = resnet50(identifier, init_weights)
brain_model = ModelCommitment(identifier=identifier,
activations_model=model,
layers=layers['resnet-50'])
return brain_model
def alexnet_brainmodel(identifier, init_weights=True, function=None):
model = alexnet(identifier, init_weights)
brain_model = ModelCommitment(identifier=identifier,
activations_model=model,
layers=layers['alexnet'])
return brain_model
def load(identifier=basemodel_identifier, activations_model=activations_model, layers=model_layer):
assert hasattr(activations_model, 'reload')
activations_model.reload()
from model_tools.brain_transformation import ModelCommitment
brain_model = ModelCommitment(identifier=identifier, activations_model=activations_model, layers=layers)
return brain_model
def test_standard_commitment(self):
brain_model = ModelCommitment(identifier=fullname(self),
activations_model=None,
layers=['dummy'])
assert brain_model.visual_degrees() == 8