code
stringlengths
17
6.64M
def get_blocks(n, num_threads): n_per_instance = (((n + (num_threads * CUDA_MAX_GRID_DIM)) - 1) // (num_threads * CUDA_MAX_GRID_DIM)) return (((n + (num_threads * n_per_instance)) - 1) // (num_threads * n_per_instance))
def compile_kernel(kernel, filename, functioname): program = Program(kernel, filename) ptx = program.compile() m = function.Module() m.load(bytes(ptx.encode())) f = m.get_function(functioname) return f
class WaitPrint(threading.Thread): def __init__(self, t, message): super().__init__() self.t = t self.message = message self.running = True def stop(self): self.running = False def run(self): for _ in range(int((self.t // 0.1))): time.sleep(0.1) if (not self.running): return print(self.message, end='')
def show_running(func): @wraps(func) def g(*args, **kargs): x = WaitPrint(2, '{}({})... '.format(func.__name__, ', '.join(([repr(x) for x in args] + ['{}={}'.format(key, repr(value)) for (key, value) in kargs.items()])))) x.start() t = time.perf_counter() r = func(*args, **kargs) if x.is_alive(): x.stop() else: print('done in {:.0f} seconds'.format((time.perf_counter() - t))) return r return g
def cached_dirpklgz(dirname): '\n Cache a function with a directory\n ' def decorator(func): '\n The actual decorator\n ' @lru_cache(maxsize=None) @wraps(func) def wrapper(*args): '\n The wrapper of the function\n ' try: os.makedirs(dirname) except FileExistsError: pass indexfile = os.path.join(dirname, 'index.pkl') try: with open(indexfile, 'rb') as file: index = pickle.load(file) except FileNotFoundError: index = {} try: filename = index[args] except KeyError: index[args] = filename = '{}.pkl.gz'.format(len(index)) with open(indexfile, 'wb') as file: pickle.dump(index, file) filepath = os.path.join(dirname, filename) try: with gzip.open(filepath, 'rb') as file: print('load {}... '.format(filename), end='') result = pickle.load(file) except FileNotFoundError: print('compute {}... '.format(filename), end='') sys.stdout.flush() result = func(*args) print('save {}... '.format(filename), end='') with gzip.open(filepath, 'wb') as file: pickle.dump(result, file) print('done') return result return wrapper return decorator
def test_so3_rfft(b_in, b_out, device): x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), dtype=torch.float, device=device) from s2cnn.soft.so3_fft import so3_rfft y1 = so3_rfft(x, b_out=b_out) from s2cnn import so3_rft, so3_soft_grid import lie_learn.spaces.S3 as S3 weights = torch.tensor(S3.quadrature_weights(b_in), dtype=torch.float, device=device) x2 = torch.einsum('bac,b->bac', (x, weights)) y2 = so3_rft(x2.view((- 1)), b_out, so3_soft_grid(b_in)) assert ((y1 - y2).abs().max().item() < (0.0001 * y1.abs().mean().item()))
def test_inverse(f, g, b_in, b_out, device, complex): if complex: x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), 2, dtype=torch.float, device=device) else: x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), dtype=torch.float, device=device) x = g(f(x, b_out=b_out), b_out=b_in) y = g(f(x, b_out=b_out), b_out=b_in) assert ((x - y).abs().max().item() < (0.0001 * y.abs().mean().item()))
def test_inverse2(f, g, b_in, b_out, device): x = torch.randn(((b_in * ((4 * (b_in ** 2)) - 1)) // 3), 2, dtype=torch.float, device=device) x = g(f(x, b_out=b_out), b_out=b_in) y = g(f(x, b_out=b_out), b_out=b_in) assert ((x - y).abs().max().item() < (0.0001 * y.abs().mean().item()))
def compare_cpu_gpu(f, x): z1 = f(x.cpu()) z2 = f(x.cuda()).cpu() q = ((z1 - z2).abs().max().item() / z1.std().item()) assert (q < 0.0001)
class BatchGenerator(): '\n Iterate over a video datasets, returning filenames of frames to laod.\n preprocessing.py can be used in combination with BatchGenerator to read and preprocess frames.\n\n (num_labels) number of labels in dataset.\n (filename) part of filename before _train.pkl or _test.pkl in annotations folder (e.g. Annotations/D1).\n (temporal_window) number of frames to be sampled in a single video segment.\n (flow_data_path) Path to folder containing flow frames.\n (rgb_data_path) Path to folder containing rgb frames.\n (random_sync) Half the batch with corresponding examples, half the batch with non-corresponding examples.\n (synchronised) flow and rgb frames are temporally synchronised for corresponding examples.\n\n nextBatch - returns rgb and flow filenames of frames to load in a training batch, labels, and correspondence labels\n nextBatchEval - returns rgb and flow filenames of frames to load in a training batch, labels, and correspondence labels\n ' def __init__(self, num_labels, filename, temporal_window=16, flow_data_path='flow_frames_parent/flow', rgb_data_path='rgb_frames_parent/frames', synchronised=False, random_sync=False): self.random_sync = random_sync self.synchronised = synchronised self.temporal_window = temporal_window self.flow_data_path = (flow_data_path + '/') self.rgb_data_path = (rgb_data_path + '/') self.filename = filename self.num_labels = num_labels (dataset_data, dataset_labels) = self._parse_inputs_df((filename + '_train.pkl')) (dataset_data_test, dataset_labels_test) = self._parse_inputs_df((filename + '_test.pkl')) dataset_data_total = np.arange(dataset_data.shape[0]) dataset_test_total = np.arange(dataset_data_test.shape[0]) dataset_data_train = (dataset_data, dataset_labels) dataset_data_test = (dataset_data_test, dataset_labels_test) self.dataset_data = {False: dataset_data_train, True: dataset_data_test} dataset_data_train_total = dataset_data_total dataset_data_test_total = dataset_test_total self.dataset_total = {False: dataset_data_train_total, True: dataset_data_test_total} def reset_dataset(self, test=False): ' Reset dataset iterator to include all data' self.dataset_total[test] = np.arange(self.dataset_data[test][0].shape[0]) def _parse_inputs_df(self, filename): ' Read annotation file ' df = pd.read_pickle(filename) data = [] for (_, line) in df.iterrows(): image = [((line['participant_id'] + '/') + line['video_id']), line['start_frame'], line['stop_frame']] labels = line['verb_class'] one_hot = np.zeros(self.num_labels) one_hot[labels] = 1.0 data.append((image[0], image[1], image[2], one_hot)) (segment, start, end, softmaxlabels) = list(zip(*data)) labels = list(softmaxlabels) train = list(zip(segment, start, end)) train = np.array(train) labels = np.array(labels) return (train, labels) def nextBatch(self, batch_size, test=False): ' Get next training samples\n loops through datasets, randomly sampling data without replacement to add to batch.' batch_size = int(batch_size) (dataset_data, dataset_labels) = self.dataset_data[test] dataset_total = self.dataset_total[test] file = self.filename if (len(dataset_total) > batch_size): sample_idx_t = np.random.choice(range(dataset_total.shape[0]), size=batch_size, replace=False) sample_idx = dataset_total[sample_idx_t] self.dataset_total[test] = np.delete(dataset_total, sample_idx_t, axis=0) else: sample_idx = dataset_total remaining = (batch_size - sample_idx.shape[0]) dataset_total_temp = np.arange(dataset_data.shape[0]) dataset_total_temp = np.delete(dataset_total_temp, sample_idx, axis=0) sample_idx_t = np.random.choice(range(dataset_total_temp.shape[0]), size=remaining, replace=False) sample_idx = np.concatenate((sample_idx, dataset_total_temp[sample_idx_t])) self.dataset_total[test] = np.delete(dataset_total_temp, sample_idx_t, axis=0) print('Done Epoch') sample = dataset_data[sample_idx] if self.synchronised: sychron = ([True] * len(sample)) else: sychron = ([False] * len(sample)) sample = [self.sample_segment(filen, synchronise=to_sync) for (filen, to_sync) in zip(sample, sychron)] (sample_rgb, sample_flow) = zip(*sample) if self.random_sync: half = int((len(sample) / 2)) fixed_sample_rgb = sample_rgb[:half] fixed_sample_flow = sample_flow[:half] variate_sample_rgb = sample_rgb[half:] variate_sample_flow = sample_flow[half:] variate_sample_flow = (variate_sample_flow[1:] + variate_sample_flow[:1]) sample_flow = (fixed_sample_flow + variate_sample_flow) sample_rgb = (fixed_sample_rgb + variate_sample_rgb) sychron = (([True] * len(fixed_sample_rgb)) + ([False] * len(variate_sample_rgb))) elif self.synchronised: sychron = ([True] * len(sample)) else: sychron = ([True] * len(sample)) sample_labels = dataset_labels[sample_idx] batch_labels = sample_labels batch_rgb = list(sample_rgb) batch_flow = list(sample_flow) combined = list(zip(batch_rgb, batch_flow, batch_labels, sychron)) shuffle(combined) (batch_rgb, batch_flow, batch_labels, sychron) = list(zip(*combined)) batch_rgb = np.array(batch_rgb) batch_flow = np.array(batch_flow) return (batch_rgb, batch_flow, batch_labels, sychron) def nextBatchEval(self, batch_size, test=True): ' Get next testing samples, return 5 equidistant frames along a action segment\n loops through datasets, randomly sampling data without replacement to add to batch. ' (dataset_data, dataset_labels) = self.dataset_data[test] dataset_total = self.dataset_total[test] dataset_data = dataset_data dataset_labels = dataset_labels dataset_total = dataset_total batch_rgb = [] batch_flow = [] batch_labels = np.empty(shape=[0, self.num_labels]) done = True if (len(dataset_total) != 0): done = False if (len(dataset_total) > batch_size): sample_idx = np.random.choice(range(dataset_total.shape[0]), size=batch_size, replace=False) else: sample_idx = range(dataset_total.shape[0]) done = True sample = dataset_data[dataset_total[sample_idx]] sample = [self.sample_segment_test(filen) for filen in sample] (sample_rgb, sample_flow) = zip(*sample) sample_labels = dataset_labels[dataset_total[sample_idx]] self.dataset_total[test] = np.delete(dataset_total, sample_idx, axis=0) batch_labels = np.concatenate((sample_labels, batch_labels)) batch_rgb = (list(sample_rgb) + batch_rgb) batch_flow = (list(sample_flow) + batch_flow) if done: self.dataset_total[test] = np.arange(dataset_data.shape[0]) batch_rgb = np.array(batch_rgb) batch_flow = np.array(batch_flow) return (done, batch_rgb, batch_flow, batch_labels) def sample_segment_test(self, s): ' Samples rgb and flow frame windows from a video segment s.\n Sampling 5 windows, equidistant along a video segment\n s = ["filename", start_frame, end_frame]' def _path_to_dataset(flow): if flow: left = self.flow_data_path else: left = self.rgb_data_path right = '/frame_' numframe = 10 return (left, right, numframe) def flow_filename(frameno, num_stack=1): (left, right, fill_frame) = _path_to_dataset(True) left_frame = (frameno - int(((num_stack - 1) / 2))) right_frame = (frameno + int((num_stack / 2))) filename = [] for no in range(left_frame, (right_frame + 1)): filename.append((((((left + str(s[0])) + '/u') + right) + str(no).zfill(fill_frame)) + '.jpg')) filename.append((((((left + str(s[0])) + '/v') + right) + str(no).zfill(fill_frame)) + '.jpg')) return filename def rgb_filename(frameno): (left, right, fill_frame) = _path_to_dataset(False) filename = ((((left + str(s[0])) + right) + str(frameno).zfill(fill_frame)) + '.jpg') return filename def c3d_sampling(): num_sample_frame = self.temporal_window half_sample_frame = int((self.temporal_window / 2)) segment_images = [] segment_flow = [] step = 2 segment_start = (int(s[1]) + (step * half_sample_frame)) segment_end = ((int(s[2]) + 1) - (step * half_sample_frame)) if (segment_start >= segment_end): segment_start = int(s[1]) segment_end = int(s[2]) if (segment_start <= ((half_sample_frame * step) + 1)): segment_start = ((half_sample_frame * step) + 2) for center_frame in np.linspace(segment_start, segment_end, 7, dtype=np.int32)[1:(- 1)]: seg_f = [] seg_i = [] for no in range((center_frame - (step * half_sample_frame)), (center_frame + (step * half_sample_frame)), step): seg_f.append(flow_filename(int((no / 2)))) seg_i.append(rgb_filename(no)) segment_flow.append(seg_f) segment_images.append(seg_i) return (segment_images, segment_flow) return c3d_sampling() def sample_segment(self, s, synchronise=False): ' Samples rgb and flow frame windows from a video segment s.\n Sampling temporal windows randomly in video segment.\n s = ["filename", start_frame, end_frame]' def _path_to_dataset(flow): if flow: left = self.flow_data_path else: left = self.rgb_data_path right = '/frame_' numframe = 10 return (left, right, numframe) def flow_filename(frameno, num_stack=1): (left, right, fill_frame) = _path_to_dataset(True) left_frame = (frameno - int(((num_stack - 1) / 2))) right_frame = (frameno + int((num_stack / 2))) filename = [] for no in range(left_frame, (right_frame + 1)): filename.append((((((left + str(s[0])) + '/u') + right) + str(no).zfill(fill_frame)) + '.jpg')) filename.append((((((left + str(s[0])) + '/v') + right) + str(no).zfill(fill_frame)) + '.jpg')) return filename def rgb_filename(frameno): (left, right, fill_frame) = _path_to_dataset(False) filename = ((((left + str(s[0])) + right) + str(frameno).zfill(fill_frame)) + '.jpg') return filename def c3d_sampling(): num_sample_frame = self.temporal_window half_sample_frame = int((self.temporal_window / 2)) segment_images = [] segment_flow = [] step = 2 segment_start = (int(s[1]) + (step * half_sample_frame)) segment_end = ((int(s[2]) + 1) - (step * half_sample_frame)) if (segment_start >= segment_end): segment_start = int(s[1]) segment_end = int(s[2]) if (segment_start <= ((half_sample_frame * step) + 1)): segment_start = ((half_sample_frame * step) + 2) if synchronise: center_frame_rgb = center_frame_flow = randint(segment_start, segment_end) else: center_frame_rgb = randint(segment_start, segment_end) center_frame_flow = randint(segment_start, segment_end) for no in range((center_frame_rgb - (step * half_sample_frame)), (center_frame_rgb + (step * half_sample_frame)), step): segment_images.append(rgb_filename(no)) for no in range((center_frame_flow - (step * half_sample_frame)), (center_frame_flow + (step * half_sample_frame)), step): segment_flow.append(flow_filename(int((no / 2)))) return (segment_images, segment_flow) return c3d_sampling()
def i3d_model(input_images, is_training, num_labels, dropout, flip_classifier_gradient=False, flip_weight=1.0, aux_classifier=False, feat_level='features'): rgb_model = i3d.InceptionI3d((num_labels + num_labels), spatial_squeeze=True, final_endpoint='Logits', aux_classifier=aux_classifier) (logits, endpoints) = rgb_model(input_images, is_training=is_training, dropout_keep_prob=dropout, flip_classifier_gradient=flip_classifier_gradient, flip_weight=flip_weight) if aux_classifier: aux_classifier_logits = endpoints['aux_classifier'] else: aux_classifier_logits = None features = endpoints[feat_level] return (logits, aux_classifier_logits, features)
def build_i3d(reuse_variables, input_images, is_training, num_labels, flow, temporal_window, dropout, flip_classifier_gradient, flip_weight=1.0, aux_classifier=False, feat_level='features'): with tf.variable_scope(tf.get_variable_scope(), reuse=reuse_variables): return i3d_model(input_images, is_training, num_labels, dropout, flip_classifier_gradient=flip_classifier_gradient, flip_weight=flip_weight, aux_classifier=aux_classifier, feat_level=feat_level)
def weight_variable(shape): initial = tf.truncated_normal(shape, stddev=0.1) return tf.Variable(initial)
def bias_variable(shape): initial = tf.constant(0.1, shape=shape) return tf.Variable(initial)
def domain_classifier(feat): shape = feat.get_shape().as_list() dim = np.prod(shape[1:]) feat = tf.reshape(feat, [(- 1), dim]) with tf.variable_scope('Domain_Classifier'): d_h_fc0 = tf.layers.dense(feat, 100, kernel_initializer=tf.initializers.truncated_normal(stddev=0.1), name='first') d_h_fc0 = tf.nn.relu(d_h_fc0) d_logits = tf.layers.dense(d_h_fc0, 2, kernel_initializer=tf.initializers.truncated_normal(stddev=0.1), name='second') return d_logits
def predict_synch(feat): shape = feat.get_shape().as_list() dim = np.prod(shape[1:]) feat = tf.reshape(feat, [(- 1), dim]) d_h_fc0 = tf.layers.dense(feat, 100, kernel_initializer=tf.initializers.truncated_normal(stddev=0.1), name='first') d_h_fc0 = tf.nn.relu(d_h_fc0) d_logits = tf.layers.dense(d_h_fc0, 2, kernel_initializer=tf.initializers.truncated_normal(stddev=0.1), name='second') return (d_logits, d_h_fc0)
class FlipGradientBuilder(object): def __init__(self): self.num_calls = 0 def __call__(self, x, l=1.0): grad_name = ('FlipGradient%d' % self.num_calls) @ops.RegisterGradient(grad_name) def _flip_gradients(op, grad): return [(tf.negative(grad) * l)] g = tf.get_default_graph() with g.gradient_override_map({'Identity': grad_name}): y = tf.identity(x) self.num_calls += 1 return y
class Unit3D(snt.AbstractModule): 'Basic unit containing Conv3D + BatchNorm + non-linearity.' def __init__(self, output_channels, kernel_shape=(1, 1, 1), stride=(1, 1, 1), activation_fn=tf.nn.relu, use_batch_norm=True, use_bias=False, name='unit_3d'): 'Initializes Unit3D module.' super(Unit3D, self).__init__(name=name) self._output_channels = output_channels self._kernel_shape = kernel_shape self._stride = stride self._use_batch_norm = use_batch_norm self._activation_fn = activation_fn self._use_bias = use_bias def _build(self, inputs, is_training): 'Connects the module to inputs.\n\n Args:\n inputs: Inputs to the Unit3D component.\n is_training: whether to use training mode for snt.BatchNorm (boolean).\n\n Returns:\n Outputs from the module.\n ' net = snt.Conv3D(output_channels=self._output_channels, kernel_shape=self._kernel_shape, stride=self._stride, padding=snt.SAME, use_bias=self._use_bias)(inputs) if self._use_batch_norm: bn = snt.BatchNorm() net = bn(net, is_training=is_training, test_local_stats=False) if (self._activation_fn is not None): net = self._activation_fn(net) return net
class InceptionI3d(snt.AbstractModule): 'Inception-v1 I3D architecture.\n\n The model is introduced in:\n\n Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset\n Joao Carreira, Andrew Zisserman\n https://arxiv.org/pdf/1705.07750v1.pdf.\n\n See also the Inception architecture, introduced in:\n\n Going deeper with convolutions\n Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed,\n Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich.\n http://arxiv.org/pdf/1409.4842v1.pdf.\n ' VALID_ENDPOINTS = ('Conv3d_1a_7x7', 'MaxPool3d_2a_3x3', 'Conv3d_2b_1x1', 'Conv3d_2c_3x3', 'MaxPool3d_3a_3x3', 'Mixed_3b', 'Mixed_3c', 'MaxPool3d_4a_3x3', 'Mixed_4b', 'Mixed_4c', 'Mixed_4d', 'Mixed_4e', 'Mixed_4f', 'MaxPool3d_5a_2x2', 'Mixed_5b', 'Mixed_5c', 'Logits', 'Predictions') def __init__(self, num_classes=400, spatial_squeeze=True, final_endpoint='Logits', name='inception_i3d', aux_classifier=False): "Initializes I3D model instance.\n\n Args:\n num_classes: The number of outputs in the logit layer (default 400, which\n matches the Kinetics dataset).\n spatial_squeeze: Whether to squeeze the spatial dimensions for the logits\n before returning (default True).\n final_endpoint: The model contains many possible endpoints.\n `final_endpoint` specifies the last endpoint for the model to be built\n up to. In addition to the output at `final_endpoint`, all the outputs\n at endpoints up to `final_endpoint` will also be returned, in a\n dictionary. `final_endpoint` must be one of\n InceptionI3d.VALID_ENDPOINTS (default 'Logits').\n name: A string (optional). The name of this module.\n\n Raises:\n ValueError: if `final_endpoint` is not recognized.\n " if (final_endpoint not in self.VALID_ENDPOINTS): raise ValueError(('Unknown final endpoint %s' % final_endpoint)) super(InceptionI3d, self).__init__(name=name) self._num_classes = num_classes self._spatial_squeeze = spatial_squeeze self._final_endpoint = final_endpoint self._aux_classifier = aux_classifier def _build(self, inputs, is_training, dropout_keep_prob=1.0, flip_classifier_gradient=False, flip_weight=1.0): 'Connects the model to inputs.\n\n Args:\n inputs: Inputs to the model, which should have dimensions\n `batch_size` x `num_frames` x 224 x 224 x `num_channels`.\n is_training: whether to use training mode for snt.BatchNorm (boolean).\n dropout_keep_prob: Probability for the tf.nn.dropout layer (float in\n [0, 1)).\n\n Returns:\n A tuple consisting of:\n 1. Network output at location `self._final_endpoint`.\n 2. Dictionary containing all endpoints up to `self._final_endpoint`,\n indexed by endpoint name.\n\n Raises:\n ValueError: if `self._final_endpoint` is not recognized.\n ' if (self._final_endpoint not in self.VALID_ENDPOINTS): raise ValueError(('Unknown final endpoint %s' % self._final_endpoint)) net = inputs end_points = {} end_point = 'Conv3d_1a_7x7' net = Unit3D(output_channels=64, kernel_shape=[7, 7, 7], stride=[2, 2, 2], name=end_point)(net, is_training=is_training) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'MaxPool3d_2a_3x3' net = tf.nn.max_pool3d(net, ksize=[1, 1, 3, 3, 1], strides=[1, 1, 2, 2, 1], padding=snt.SAME, name=end_point) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Conv3d_2b_1x1' net = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name=end_point)(net, is_training=is_training) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Conv3d_2c_3x3' net = Unit3D(output_channels=192, kernel_shape=[3, 3, 3], name=end_point)(net, is_training=is_training) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'MaxPool3d_3a_3x3' net = tf.nn.max_pool3d(net, ksize=[1, 1, 3, 3, 1], strides=[1, 1, 2, 2, 1], padding=snt.SAME, name=end_point) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_3b' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=96, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=128, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=16, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=32, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=32, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_3c' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=192, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=32, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=96, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'MaxPool3d_4a_3x3' net = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 2, 2, 2, 1], padding=snt.SAME, name=end_point) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_4b' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=192, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=96, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=208, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=16, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=48, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_4c' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=160, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=112, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=224, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=24, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=64, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_4d' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=256, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=24, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=64, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_4e' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=112, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=144, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=288, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=32, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=64, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=64, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_4f' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=256, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=160, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=320, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=32, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=128, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'MaxPool3d_5a_2x2' net = tf.nn.max_pool3d(net, ksize=[1, 2, 2, 2, 1], strides=[1, 2, 2, 2, 1], padding=snt.SAME, name=end_point) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_5b' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=256, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=160, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=320, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=32, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=128, kernel_shape=[3, 3, 3], name='Conv3d_0a_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Mixed_5c' with tf.variable_scope(end_point): with tf.variable_scope('Branch_0'): branch_0 = Unit3D(output_channels=384, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) with tf.variable_scope('Branch_1'): branch_1 = Unit3D(output_channels=192, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_1 = Unit3D(output_channels=384, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_1, is_training=is_training) with tf.variable_scope('Branch_2'): branch_2 = Unit3D(output_channels=48, kernel_shape=[1, 1, 1], name='Conv3d_0a_1x1')(net, is_training=is_training) branch_2 = Unit3D(output_channels=128, kernel_shape=[3, 3, 3], name='Conv3d_0b_3x3')(branch_2, is_training=is_training) with tf.variable_scope('Branch_3'): branch_3 = tf.nn.max_pool3d(net, ksize=[1, 3, 3, 3, 1], strides=[1, 1, 1, 1, 1], padding=snt.SAME, name='MaxPool3d_0a_3x3') branch_3 = Unit3D(output_channels=128, kernel_shape=[1, 1, 1], name='Conv3d_0b_1x1')(branch_3, is_training=is_training) net = tf.concat([branch_0, branch_1, branch_2, branch_3], 4) end_points[end_point] = net if (self._final_endpoint == end_point): return (net, end_points) end_point = 'Logits' with tf.variable_scope(end_point): net = tf.nn.avg_pool3d(net, ksize=[1, 2, 7, 7, 1], strides=[1, 1, 1, 1, 1], padding=snt.VALID) end_points['features'] = net net = tf.nn.dropout(net, dropout_keep_prob) if flip_classifier_gradient: net = flip_gradient(net, flip_weight) if self._aux_classifier: with tf.variable_scope('AuxLogits'): logits_aux = Unit3D(output_channels=self._num_classes, kernel_shape=[1, 1, 1], activation_fn=None, use_batch_norm=False, use_bias=True, name='Conv3d_0c_1x1')(net, is_training=is_training) if self._spatial_squeeze: logits_aux = tf.squeeze(logits_aux, [2, 3], name='SpatialSqueezeAux') end_points['aux_classifier'] = tf.reduce_mean(logits_aux, axis=1) logits = Unit3D(output_channels=self._num_classes, kernel_shape=[1, 1, 1], activation_fn=None, use_batch_norm=False, use_bias=True, name='Conv3d_0c_1x1')(net, is_training=is_training) if self._spatial_squeeze: logits = tf.squeeze(logits, [2, 3], name='SpatialSqueeze') averaged_logits = tf.reduce_mean(logits, axis=1) end_points[end_point] = averaged_logits if (self._final_endpoint == end_point): return (averaged_logits, end_points) end_point = 'Predictions' predictions = tf.nn.softmax(averaged_logits) end_points[end_point] = predictions return (predictions, end_points)
def _mix_rbf_kernel(X, Y, gammas, wts=None): if (wts is None): wts = ([1] * len(gammas)) XX = tf.matmul(X, X, transpose_b=True) XY = tf.matmul(X, Y, transpose_b=True) YY = tf.matmul(Y, Y, transpose_b=True) X_sqnorms = tf.diag_part(XX) Y_sqnorms = tf.diag_part(YY) r = (lambda x: tf.expand_dims(x, 0)) c = (lambda x: tf.expand_dims(x, 1)) (K_XX, K_XY, K_YY) = (0, 0, 0) for (gamma_t, wt) in zip(gammas, wts): gamma = (1 / gamma_t) K_XX += (wt * tf.exp(((- gamma) * ((((- 2) * XX) + c(X_sqnorms)) + r(X_sqnorms))))) K_XY += (wt * tf.exp(((- gamma) * ((((- 2) * XY) + c(X_sqnorms)) + r(Y_sqnorms))))) K_YY += (wt * tf.exp(((- gamma) * ((((- 2) * YY) + c(Y_sqnorms)) + r(Y_sqnorms))))) return (K_XX, K_XY, K_YY, tf.reduce_sum(wts))
def rbf_mmd2(X, Y, gammas=1, biased=True): return mix_rbf_mmd2(X, Y, gammas=[gammas], biased=biased)
def mix_rbf_mmd2(X, Y, gammas=(1,), wts=None, biased=True): (K_XX, K_XY, K_YY, d) = _mix_rbf_kernel(X, Y, gammas, wts) return _mmd2(K_XX, K_XY, K_YY, const_diagonal=d, biased=biased)
def _mmd2(K_XX, K_XY, K_YY, const_diagonal=False, biased=False): m = tf.cast(tf.shape(K_XX)[0], tf.float32) n = tf.cast(tf.shape(K_YY)[0], tf.float32) if biased: mmd2 = (((tf.reduce_sum(K_XX) / (m * m)) + (tf.reduce_sum(K_YY) / (n * n))) - ((2 * tf.reduce_sum(K_XY)) / (m * n))) else: if (const_diagonal is not False): trace_X = (m * const_diagonal) trace_Y = (n * const_diagonal) else: trace_X = tf.trace(K_XX) trace_Y = tf.trace(K_YY) mmd2 = ((((tf.reduce_sum(K_XX) - trace_X) / (m * (m - 1))) + ((tf.reduce_sum(K_YY) - trace_Y) / (n * (n - 1)))) - ((2 * tf.reduce_sum(K_XY)) / (m * n))) return (mmd2, n)
def main(): seen = tf.placeholder(tf.float32, shape=[None, 1024]) unseen = tf.placeholder(tf.float32, shape=[None, 1024]) (mmd, n) = rbf_mmd2(seen, unseen) (mmd, n) = mix_rbf_mmd2(seen, unseen, gammas=[10.0, 1.0, 0.1, 0.01, 0.001]) source_numpy = np.load(sys.argv[1]) target_numpy = np.load(sys.argv[2]) source_numpy_labels = np.load(sys.argv[3]) target_numpy_labels = np.load(sys.argv[4]) with tf.Session() as sess: print('Total', sess.run(mmd, feed_dict={seen: source_numpy, unseen: target_numpy})) for i in np.unique(source_numpy_labels): print(i, sess.run(mmd, feed_dict={seen: source_numpy[(source_numpy_labels == i)], unseen: target_numpy[(target_numpy_labels == i)]}))
def _get_variables_to_restore_load(to_ignore, flow): to_ignore.append('domain_accumulators') to_ignore.append('accum_accumulators') to_ignore.append('accumulators') scope_to_ignore = to_ignore if flow: scope_to_ignore.append('RGB') scope_to_ignore.append('Joint') else: scope_to_ignore.append('Flow') scope_to_ignore.append('Joint') variables = slim.get_variables_to_restore(exclude=scope_to_ignore) keyword_filter = to_ignore return [x for x in variables if (not any(((word in x.name) for word in keyword_filter)))]
def read_joint(mode=''): if (mode == 'restore'): to_ignore = ['Adam', 'adam', 'Momentum', 'beta1_power', 'beta2_power', 'global_step', 'Domain_Classifier'] elif (mode == 'continue'): to_ignore = [] else: raise Exception('Unknown mode for read_joint') to_ignore.append('domain_accumulators') to_ignore.append('accum_accumulators') to_ignore.append('accumulators') to_ignore.append('Flow') to_ignore.append('RGB') variables = slim.get_variables_to_restore(exclude=to_ignore) variables_restore = [x for x in variables if (not any(((word in x.name) for word in to_ignore)))] rgb_variable_map = {} for variable in variables_restore: name = variable.name.replace(':0', '') rgb_variable_map[name] = variable variable_loader = tf.train.Saver(var_list=rgb_variable_map, reshape=True, max_to_keep=20) return variable_loader
def read_i3d_checkpoint(mode='', flow=False, aux_logits=False): if (mode == 'pretrain'): to_ignore = ['inception_i3d/Logits', 'Adam', 'adam', 'Momentum', 'beta1_power', 'beta2_power', 'global_step', 'Domain_Classifier', 'arrow_test'] elif (mode == 'restore'): to_ignore = ['Adam', 'adam', 'Momentum', 'beta1_power', 'beta2_power', 'global_step', 'Domain_Classifier'] elif (mode == 'continue'): to_ignore = [] else: raise Exception('Unkown mode for read_i3d_checkpoint') variables_restore = _get_variables_to_restore_load(to_ignore, flow) rgb_variable_map = {} for variable in variables_restore: name = variable.name.replace(':0', '') rgb_variable_map[name] = variable variable_loader = tf.train.Saver(var_list=rgb_variable_map, reshape=True, max_to_keep=20) return variable_loader
def restore_base(sess, saver, checkpoint_path, model_to_restore, restore_mode='model'): if (restore_mode == 'continue'): ckpt = tf.train.get_checkpoint_state(checkpoint_path) if (ckpt and ckpt.model_checkpoint_path): saver['continue'].restore(sess, ckpt.model_checkpoint_path) else: raise Exception('Cannot find a model to continue training') start_step = ckpt.model_checkpoint_path.split('/')[(- 1)].split('-')[(- 1)] elif (restore_mode == 'model'): saver['model'].restore(sess, model_to_restore) start_step = 0 elif ((restore_mode == 'pretrain') or (restore_mode == 'pretrain_from_synch')): saver['pretrain'].restore(sess, model_to_restore) start_step = 0 else: raise Exception('A valid restore Mode must be set --restore_mode==[continue,model,pretrain]') return start_step
def restore_joint(sess, saver, checkpoint_path, model_to_restore, restore_mode='model'): if (restore_mode == 'continue'): ckpt = tf.train.get_checkpoint_state(checkpoint_path) if (ckpt and ckpt.model_checkpoint_path): saver['continue'].restore(sess, ckpt.model_checkpoint_path) else: raise Exception('Cannot find a model to continue training') start_step = ckpt.model_checkpoint_path.split('/')[(- 1)].split('-')[(- 1)] elif ((restore_mode == 'model') or (restore_mode == 'pretrain_from_synch')): saver['model'].restore(sess, model_to_restore) start_step = 0 elif (restore_mode == 'pretrain'): start_step = 0 else: raise Exception('A valid restore Mode must be set --restore_mode==[continue,model,pretrain]') return start_step
def init_savers_base(flow=False): pretrain_loader = read_i3d_checkpoint(mode='pretrain', flow=flow) model_loader = read_i3d_checkpoint(mode='restore', flow=flow) savesave = read_i3d_checkpoint(mode='continue', flow=flow) return (pretrain_loader, model_loader, savesave)
class TrainTestScript(): ' Creates a framework to train/test an MM-SADA model\n (FLAGS) TensorFlow flags\n (results_dir) Directory of tensorboard files and other testing logs\n (train_dir) Director of saved model\n\n Methods:\n train - train MM-SADA\n test - evaluate an MM-SADA saved model\n ' def __init__(self, FLAGS, results_dir, train_dir): self.FLAGS = FLAGS self.train_dir = train_dir self.datasets = FLAGS.datasets self.unseen_dataset = FLAGS.unseen_dataset self.num_gpus = FLAGS.num_gpus self.num_labels = FLAGS.num_labels self.target_data = (not (not FLAGS.domain_mode)) if self.target_data: if ((FLAGS.domain_mode == 'None') or (FLAGS.domain_mode == 'Pretrain')): self.target_data = False print('No adaptation') if FLAGS.domain_mode: self.domain_mode = FLAGS.domain_mode else: self.domain_mode = 'None' self.lr = FLAGS.lr if (not FLAGS.modality): raise Exception('Need to Specify modality') if ((FLAGS.modality != 'rgb') and (FLAGS.modality != 'flow') and (FLAGS.modality != 'joint')): raise Exception('Invalid Modality') self.results_dir = ((results_dir + '_') + FLAGS.modality) self.modality = FLAGS.modality self.model = Model(num_gpus=self.num_gpus, num_labels=self.num_labels, modality=self.modality, temporal_window=self.FLAGS.temporal_window, batch_norm_update=self.FLAGS.batch_norm_update, domain_mode=self.domain_mode, steps_per_update=FLAGS.steps_before_update, aux_classifier=self.FLAGS.aux_classifier, synchronised=self.FLAGS.synchronised, predict_synch=self.FLAGS.pred_synch, selfsupervised_lambda=self.FLAGS.self_lambda) def training_batch_gen(self): batch_gen = BatchGenerator(self.num_labels, self.datasets, temporal_window=self.FLAGS.temporal_window, rgb_data_path=self.FLAGS.rgb_data_path, flow_data_path=self.FLAGS.flow_data_path, synchronised=self.FLAGS.synchronised, random_sync=self.FLAGS.pred_synch) batch_gen_unseen = BatchGenerator(self.num_labels, self.unseen_dataset, temporal_window=self.FLAGS.temporal_window, rgb_data_path=self.FLAGS.rgb_data_path, flow_data_path=self.FLAGS.flow_data_path, synchronised=self.FLAGS.synchronised, random_sync=self.FLAGS.pred_synch) return (batch_gen, batch_gen_unseen) def testing_batch_gen(self): batch_gen = BatchGenerator(self.num_labels, self.datasets, temporal_window=self.FLAGS.temporal_window, rgb_data_path=self.FLAGS.rgb_data_path, flow_data_path=self.FLAGS.flow_data_path, synchronised=self.FLAGS.synchronised, random_sync=self.FLAGS.pred_synch) batch_gen_unseen = BatchGenerator(self.num_labels, self.unseen_dataset, temporal_window=self.FLAGS.temporal_window, rgb_data_path=self.FLAGS.rgb_data_path, flow_data_path=self.FLAGS.flow_data_path, synchronised=self.FLAGS.synchronised, random_sync=self.FLAGS.pred_synch) return (batch_gen, batch_gen_unseen) def train(self): ' Train MM-SADA model' g1 = tf.Graph() with g1.as_default(), tf.device('/cpu:0'): self.model.init_savers() train_writer = tf.summary.FileWriter((self.results_dir + '/train')) seen_writer = tf.summary.FileWriter((self.results_dir + '/seen')) unseen_writer = tf.summary.FileWriter((self.results_dir + '/unseen')) (batch_gen, batch_gen_unseen) = self.training_batch_gen() with tf.Session(graph=g1, config=tf.ConfigProto(allow_soft_placement=True)) as sess: print('init variables') sess.run(tf.global_variables_initializer()) start_step = self.model.restore_model_train(sess, self.train_dir, self.FLAGS.restore_model_flow, self.FLAGS.restore_model_rgb, self.FLAGS.restore_model_joint, self.FLAGS.restore_mode) for step in range(int(start_step), (self.FLAGS.max_steps + 1)): p = (float(step) / self.FLAGS.max_steps) lin = (((2 / (1.0 + np.exp(((- 10.0) * p)))) - 1) * self.FLAGS.lambda_in) start_time = time.time() (training_loss, training_accuracy, summary) = train_step(sess, self.model, self.FLAGS, batch_gen, batch_gen_unseen, lin, self.target_data) for s in summary: train_writer.add_summary(s, step) duration = (time.time() - start_time) if ((step % 50) == 0): num_examples_per_step = self.FLAGS.batch_size examples_per_sec = (num_examples_per_step / duration) sec_per_batch = duration format_str = '(Train) %s: step %d, loss %.3f, acc %.3f (%.1f examples/sec; %.3f sec/batch)' print((format_str % (datetime.now(), step, training_loss, training_accuracy, examples_per_sec, sec_per_batch))) (valaccuracy, domainaccuracy, average_class) = evaluate(sess, self.model, self.FLAGS, batch_gen, lin) domainaccuracy = (1.0 - domainaccuracy) val_summary = tf.Summary() val_summary.value.add(tag='acc/Accuracy', simple_value=valaccuracy) domain_summary = tf.Summary() domain_summary.value.add(tag='acc/Domain', simple_value=domainaccuracy) seen_writer.add_summary(val_summary, step) seen_writer.add_summary(domain_summary, step) format_str = '(Val) %s: domain:%s step:%d accuracy:%f avg_class %f domain_accuracy %f' print((format_str % (datetime.now(), 'Source', step, valaccuracy, average_class, domainaccuracy))) if self.FLAGS.pred_synch: synch_accuracy = evaluate_self_supervised(sess, self.model, self.FLAGS, batch_gen, lin, mode='synch') val_summary = tf.Summary() val_summary.value.add(tag='acc/Synch_Accuracy', simple_value=synch_accuracy) seen_writer.add_summary(val_summary, step) format_str = '(Val) %s: domain:%s step:%d synch_accuracy:%f' print((format_str % (datetime.now(), 'Source', step, synch_accuracy))) (valaccuracy, domainaccuracy, average_class) = evaluate(sess, self.model, self.FLAGS, batch_gen_unseen, lin) val_summary = tf.Summary() val_summary.value.add(tag='acc/Accuracy', simple_value=valaccuracy) domain_summary = tf.Summary() domain_summary.value.add(tag='acc/Domain', simple_value=domainaccuracy) unseen_writer.add_summary(val_summary, step) unseen_writer.add_summary(domain_summary, step) format_str = '(Val) %s: domain:%s step:%d accuracy:%f avg_class %f domain_accuracy %f' print((format_str % (datetime.now(), 'Target', step, valaccuracy, average_class, domainaccuracy))) if self.FLAGS.pred_synch: synch_accuracy = evaluate_self_supervised(sess, self.model, self.FLAGS, batch_gen_unseen, lin, mode='synch') val_summary = tf.Summary() val_summary.value.add(tag='acc/Synch_Accuracy', simple_value=synch_accuracy) unseen_writer.add_summary(val_summary, step) format_str = '(Val) %s: domain:%s step:%d synch_accuracy:%f' print((format_str % (datetime.now(), 'Target', step, synch_accuracy))) if (((step % 50) == 0) or (step == self.FLAGS.max_steps)): self.model.save_model(sess, self.train_dir, step) def test(self): ' Evaluate MM-SADA model' def _save_results(FLAGS, feature_list, label_list, predict_list, img_path_list, ident, test=True): ' Save statistics and extracted features to feature_path folder' if test: stringtest = 'test' else: stringtest = 'train' source_domain = os.path.basename(FLAGS.datasets) np.save(((((((((FLAGS.feature_path + '/') + stringtest) + '_feat_') + source_domain) + '_') + str(FLAGS.modelnum)) + '_') + str(ident)), feature_list) np.save(((((((((FLAGS.feature_path + '/') + stringtest) + '_label') + source_domain) + '_') + str(FLAGS.modelnum)) + '_') + str(ident)), label_list) np.save(((((((((FLAGS.feature_path + '/') + stringtest) + '_pred') + source_domain) + '_') + str(FLAGS.modelnum)) + '_') + str(ident)), predict_list) np.save(((((((((FLAGS.feature_path + '/') + stringtest) + '_filenames') + source_domain) + '_') + str(FLAGS.modelnum)) + '_') + str(ident)), img_path_list) with tf.Graph().as_default(), tf.device('/cpu:0'): (batch_gen, batch_gen_unseen) = self.testing_batch_gen() self.model.init_savers() with tf.Session(config=tf.ConfigProto(allow_soft_placement=True)) as sess: self.model.restore_model_test(sess, self.train_dir, self.FLAGS.modelnum) lin = 0.0 step = 0 seen_filenames = '' seen_accuracy = '' (valaccuracy, domainaccuracy, valperclass, valfeat, valfile, vallabel, valpredict) = evaluate(sess, self.model, self.FLAGS, batch_gen, lin, test=(not self.FLAGS.eval_train), out_features=self.FLAGS.features, extra_info=True) if self.FLAGS.features: _save_results(self.FLAGS, valfeat, vallabel, valpredict, valfile, 'Source', test=(not self.FLAGS.eval_train)) seen_accuracy = ((seen_accuracy + str(valaccuracy)) + ',') seen_filenames = ((seen_filenames + 'Source') + ',') format_str = '(Val) %s: domain:%s step:%d accuracy:%f domain_accuracy %f' print((format_str % (datetime.now(), 'Source', step, valaccuracy, domainaccuracy))) (valaccuracy, domainaccuracy, valperclass, valfeat, valfile, vallabel, valpredict) = evaluate(sess, self.model, self.FLAGS, batch_gen_unseen, lin, test=(not self.FLAGS.eval_train), out_features=self.FLAGS.features, extra_info=True) domainaccuracy = (1.0 - domainaccuracy) if self.FLAGS.features: _save_results(self.FLAGS, valfeat, vallabel, valpredict, valfile, 'Target', test=(not self.FLAGS.eval_train)) format_str = '(Val) %s: domain:%s step:%d accuracy:%f domain_accuracy %f' print((format_str % (datetime.now(), 'Target', step, valaccuracy, domainaccuracy))) results_log_file = '/logs/results.list' if (not os.path.exists((self.results_dir + '/logs'))): os.makedirs((self.results_dir + '/logs')) if (not os.path.isfile((self.results_dir + results_log_file))): f = open((self.results_dir + results_log_file), 'w') f.write(((seen_filenames + 'target,step,target_directory') + '\n')) f.close() f = open((self.results_dir + results_log_file), 'a') f.write(((((((seen_accuracy + str(valaccuracy)) + ',') + str(self.FLAGS.modelnum)) + ',') + self.FLAGS.unseen_dataset) + '\n')) f.close()
def parse_args(FLAGS): error = False if (FLAGS.train is None): print('Specify whether to train (True) or test (False) --train') error = True if (FLAGS.results_path is None): print('Specify path to save logs and models --results_path') error = True if (FLAGS.datasets is None): print('Specify the Source domain dataset --datasets') error = True if (FLAGS.unseen_dataset is None): print('Specify the Target domain dataset --unseen_dataset') error = True if (FLAGS.rgb_data_path is None): print('Specify the path to rgb frames --rgb_data_path') error = True if (FLAGS.flow_data_path is None): print('Specify the path to flow frames --flow_data_path') error = True if error: return True if FLAGS.train: if (FLAGS.restore_mode is None): print("Specify the restore mode --restore_mode ('pretrain', 'model', 'continue')") error = True if ((FLAGS.restore_model_flow is None) and FLAGS.flow): print('Specify pretrained model to use --pretrained_model') error = True if ((FLAGS.restore_model_rgb is None) and (not FLAGS.flow)): print('Specify pretrained model to use --pretrained_model') error = True else: if (FLAGS.modelnum is None): print('Specify model number to restore for testing --modelnum') error = True if FLAGS.features: if (FLAGS.feature_path is None): print('Specify path to store features --feature_path') error = True if error: return True return False
def input_parser(): flags = tf.app.flags flags.DEFINE_boolean('train', None, 'Weither to train or evaluate (False)') flags.DEFINE_string('results_path', None, 'Where to store the log files and saved models') flags.DEFINE_float('lr', 0.001, 'Initial Learning Rate') flags.DEFINE_float('batch_norm_update', 0.9, 'Update rate of batch norm statistics') flags.DEFINE_integer('num_gpus', 8, 'number of gpus to run') flags.DEFINE_integer('max_steps', 6000, 'Number of batches to run.') flags.DEFINE_integer('steps_before_update', 1, 'number of steps to run before updating weights') flags.DEFINE_string('domain_mode', None, 'background only for dataset2') flags.DEFINE_float('lambda_in', 1.0, 'grl hyperparameter') flags.DEFINE_float('self_lambda', 5.0, 'weigthing of self supervised loss') flags.DEFINE_string('datasets', None, 'Comma seperated list of datasets') flags.DEFINE_string('unseen_dataset', None, 'Specify file path to unseen dataset folder') flags.DEFINE_integer('num_labels', 8, 'Total number of combined labels') flags.DEFINE_integer('batch_size', 128, 'Size of a batch') flags.DEFINE_boolean('synchronised', None, 'Weither to synchronise flow and rgb') flags.DEFINE_string('modality', 'joint', 'rgb, flow or joint (default: joint)') flags.DEFINE_integer('temporal_window', 16, 'i3d temporal window') flags.DEFINE_boolean('aux_classifier', None, '2 classifiers') flags.DEFINE_boolean('pred_synch', None, 'Predict if modalities are synchronised') flags.DEFINE_boolean('features', None, 'Weither to produce features of evalutate') flags.DEFINE_string('feature_path', None, 'path to store features') flags.DEFINE_boolean('eval_train', None, 'Weither to evaludate training example rather than test') flags.DEFINE_integer('modelnum', None, 'model number to restore for testing') flags.DEFINE_string('restore_model_rgb', None, 'Load these weights excluding Logits') flags.DEFINE_string('restore_model_flow', None, 'Load these weights excluding Logits') flags.DEFINE_string('restore_model_joint', None, 'Load these weights excluding Logits') flags.DEFINE_string('rgb_data_path', None, 'path to rgb data') flags.DEFINE_string('flow_data_path', None, 'path to flow data') flags.DEFINE_string('restore_mode', None, 'pretrain (for base netwrok without logits), model (restore base model with classification logits) or continue (restore everything)') FLAGS = flags.FLAGS source_domain = os.path.basename(FLAGS.datasets) target_domain = os.path.basename(FLAGS.unseen_dataset) train_dir = ((((((((FLAGS.results_path + '/saved_model_') + source_domain) + '_') + target_domain) + '_') + str(FLAGS.lr)) + '_') + str(FLAGS.batch_norm_update)) if (not os.path.exists(train_dir)): os.makedirs(train_dir) results_dir = ((((((((FLAGS.results_path + '/results_') + source_domain) + '_') + target_domain) + '_') + str(FLAGS.lr)) + '_') + str(FLAGS.batch_norm_update)) return (FLAGS, train_dir, results_dir)
def main(): (flags, train_dir, results_dir) = input_parser() if parse_args(flags): return train_test = TrainTestScript(flags, results_dir, train_dir) if flags.train: train_test.train() else: train_test.test()
def eval_batch_mlp(mlp, data, batch_idxs, criterion, device_id=0): ' evaluate a batch for the baseline mlp ' atom_types = to_one_hot(data['features']['atom_types'][(batch_idxs, ...)], NUM_ATOM_TYPES) targets = data['targets'][(batch_idxs, ...)] atom_types = Variable(atom_types) targets = Variable(targets) if torch.cuda.is_available(): atom_types = atom_types.cuda(device_id) targets = targets.cuda(device_id) outputs = mlp(atom_types) loss = criterion(outputs, targets) return loss
def eval_batch_s2cnn(mlp, s2cnn, data, batch_idxs, criterion, device_id=0): ' evaluate a batch for the s2cnn ' geometry = data['features']['geometry'][(batch_idxs, ...)] atom_types = data['features']['atom_types'][(batch_idxs, ...)] atom_types_one_hot = to_one_hot(atom_types, NUM_ATOM_TYPES) targets = data['targets'][(batch_idxs, ...)] geometry = Variable(geometry) atom_types = Variable(atom_types) atom_types_one_hot = Variable(atom_types_one_hot) targets = Variable(targets) if torch.cuda.is_available(): atom_types_one_hot = atom_types_one_hot.cuda(device_id) geometry = geometry.cuda(device_id) atom_types = atom_types.cuda(device_id) targets = targets.cuda(device_id) outputs = mlp(atom_types_one_hot) outputs += s2cnn(geometry, atom_types) loss = criterion(outputs, targets) return loss
def train_baseline(mlp, data, train_batches, test_batches, num_epochs, learning_rate_mlp, device_id=0): ' train the baseline model ' optim = OPTIMIZER(mlp.parameters(), lr=learning_rate_mlp) criterion = nn.MSELoss() if torch.cuda.is_available(): criterion = criterion.cuda(device_id) for epoch in range(num_epochs): train_losses = [] print('training') for (iteration, batch_idxs) in enumerate(train_batches): mlp.train() optim.zero_grad() loss = eval_batch_mlp(mlp, data, batch_idxs, criterion, device_id) loss.backward() optim.step() train_losses.append(loss.item()) print('\riteration {}/{}'.format((iteration + 1), train_batches.num_iterations()), end='') print() test_losses = [] print('evaluating') for (iteration, batch_idxs) in enumerate(test_batches): mlp.eval() loss = eval_batch_mlp(mlp, data, batch_idxs, criterion) test_losses.append(loss.item()) print('\riteration {}/{}'.format((iteration + 1), test_batches.num_iterations()), end='') print() train_loss = np.sqrt(np.mean(train_losses)) test_loss = np.sqrt(np.mean(test_losses)) print('epoch {}/{} - avg train loss: {}, test loss: {}'.format((epoch + 1), num_epochs, train_loss, test_loss)) return (train_loss, test_loss)
def train_s2cnn(mlp, s2cnn, data, train_batches, test_batches, num_epochs, init_learning_rate_s2cnn, learning_rate_decay_epochs, device_id=0): ' train the s2cnn keeping the baseline frozen ' optim = OPTIMIZER(s2cnn.parameters(), lr=init_learning_rate_s2cnn) criterion = nn.MSELoss() if torch.cuda.is_available(): criterion = criterion.cuda(device_id) for epoch in range(num_epochs): optim = exp_lr_scheduler(optim, epoch, init_lr=init_learning_rate_s2cnn, lr_decay_epoch=learning_rate_decay_epochs) train_losses = [] print('training') for (iteration, batch_idxs) in enumerate(train_batches): s2cnn.train() mlp.eval() optim.zero_grad() loss = eval_batch_s2cnn(mlp, s2cnn, data, batch_idxs, criterion) loss.backward() optim.step() train_losses.append(loss.item()) print('\riteration {}/{} - batch loss: {}'.format((iteration + 1), train_batches.num_iterations(), np.sqrt(train_losses[(- 1)])), end='') print() test_losses = [] print('evaluating') for (iteration, batch_idxs) in enumerate(test_batches): s2cnn.eval() mlp.eval() loss = eval_batch_s2cnn(mlp, s2cnn, data, batch_idxs, criterion) test_losses.append(loss.item()) print('\riteration {}/{} - batch loss: {}'.format((iteration + 1), test_batches.num_iterations(), np.sqrt(test_losses[(- 1)])), end='') print() train_loss = np.sqrt(np.mean(train_losses)) test_loss = np.sqrt(np.mean(test_losses)) print('epoch {}/{} - avg train loss: {}, test loss: {}'.format((epoch + 1), num_epochs, train_loss, test_loss)) return (train_loss, test_loss)
def main(): parser = argparse.ArgumentParser() parser.add_argument('--data_path', type=str, default='data.joblib') parser.add_argument('--test_strat', type=int, default=0) parser.add_argument('--device_id', type=int, default=0) parser.add_argument('--num_epochs_s2cnn', type=int, default=30) parser.add_argument('--num_epochs_mlp', type=int, default=30) parser.add_argument('--batch_size_s2cnn', type=int, default=32) parser.add_argument('--batch_size_mlp', type=int, default=32) parser.add_argument('--init_learning_rate_s2cnn', type=int, default=0.001) parser.add_argument('--learning_rate_mlp', type=int, default=0.001) parser.add_argument('--learning_rate_decay_epochs', type=int, default=10) args = parser.parse_args() torch.cuda.set_device(args.device_id) print('evaluating on {}'.format(args.test_strat)) print('loading data...', end='') (data, train_idxs, test_idxs) = load_data(args.data_path, args.test_strat, cuda=args.device_id) print('done!') mlp = BaselineRegressor() s2cnn = S2CNNRegressor() if torch.cuda.is_available(): for model in [mlp, s2cnn]: model.cuda(args.device_id) print('training baseline model') print('mlp #params: {}'.format(count_params(mlp))) train_baseline(mlp, data, IndexBatcher(train_idxs, args.batch_size_mlp, cuda=args.device_id), IndexBatcher(test_idxs, args.batch_size_mlp, cuda=args.device_id), args.num_epochs_mlp, args.learning_rate_mlp, args.device_id) print('training residual s2cnn model') print('s2cnn #params: {}'.format(count_params(s2cnn))) train_s2cnn(mlp, s2cnn, data, IndexBatcher(train_idxs, args.batch_size_s2cnn, cuda=args.device_id), IndexBatcher(test_idxs, args.batch_size_s2cnn, cuda=args.device_id), args.num_epochs_s2cnn, args.init_learning_rate_s2cnn, args.learning_rate_decay_epochs, args.device_id)
class S2Block(nn.Module): ' simple s2 convolution block ' def __init__(self, b_in, b_out, f_in, f_out): ' b_in/b_out: bandwidth of input/output signals\n f_in/f_out: filters in input/output signals ' super(S2Block, self).__init__() self.grid_s2 = s2_near_identity_grid(n_alpha=(2 * b_in), n_beta=2) self.cnn = S2Convolution(nfeature_in=f_in, nfeature_out=f_out, b_in=b_in, b_out=b_out, grid=self.grid_s2) self.bn = nn.BatchNorm3d(f_out, affine=AFFINE) def forward(self, x): x = self.cnn(x) x = self.bn(x) x = nonlinearity(x) return x
class So3Block(nn.Module): ' simple so3 convolution block ' def __init__(self, b_in, b_out, f_in, f_out): ' b_in/b_out: bandwidth of input/output signals\n f_in/f_out: filters in input/output signals ' super(So3Block, self).__init__() self.grid_so3 = so3_near_identity_grid(n_alpha=(2 * b_in), n_beta=2, n_gamma=2) self.cnn = SO3Convolution(nfeature_in=f_in, nfeature_out=f_out, b_in=b_in, b_out=b_out, grid=self.grid_so3) self.bn = nn.BatchNorm3d(f_out, affine=AFFINE) def forward(self, x): x = self.cnn(x) x = self.bn(x) x = nonlinearity(x) return x
class DeepSet(nn.Module): ' deep set block ' def __init__(self, f, h1, h_latent, h2, n_objs): ' f: input filters\n h1, h2: hidden units for encoder/decoder mlps\n h_latent: dimensions\n n_objs: of objects to aggregate in latent space ' super(DeepSet, self).__init__() self.f = f self.h1 = h1 self.h3 = h2 self.n_objs = n_objs self.emb_h = nn.Linear(f, h1) self.emb_rep = nn.Linear(h1, h_latent) self.proj_h = nn.Linear(h_latent, h2) self.proj = nn.Linear(h2, 1) self.bn1 = nn.BatchNorm1d(h1, affine=AFFINE) self.bn2 = nn.BatchNorm1d(h_latent, affine=AFFINE) self.bn3 = nn.BatchNorm1d(h2, affine=AFFINE) def forward(self, x, mask): x = self.emb_h(x) x = self.bn1(x) x = nonlinearity(x) x = self.emb_rep(x) x = self.bn2(x) x = nonlinearity(x) (n, h_latent) = x.size() x = x.view((n // self.n_objs), self.n_objs, h_latent) x = torch.sum((x * mask), dim=1) x = self.proj_h(x) x = self.bn3(x) x = nonlinearity(x) x = self.proj(x) return x
class S2CNNRegressor(nn.Module): ' approximate energy using spherical representations ' def __init__(self): super(S2CNNRegressor, self).__init__() n_objs = 23 self.blocks = [S2Block(b_in=10, f_in=5, b_out=8, f_out=8), So3Block(b_in=8, b_out=6, f_in=8, f_out=16), So3Block(b_in=6, b_out=4, f_in=16, f_out=32), So3Block(b_in=4, b_out=2, f_in=32, f_out=64)] for (i, block) in enumerate(self.blocks): setattr(self, 'block{0}'.format(i), block) self.ds = DeepSet(64, 256, 64, 512, n_objs) def forward(self, x, atom_types): (n_batch, n_atoms, n_features, bandwidth, _) = x.size() mask = (atom_types > 0).view(n_batch, n_atoms, 1).float() x = x.view((n_batch * n_atoms), n_features, bandwidth, bandwidth) for block in self.blocks: x = block(x) x = so3_integrate(x) y = self.ds(x, mask) return y
class IndexBatcher(): def __init__(self, indices, n_batch, cuda=None): self.indices = indices.astype(np.int64) self.n_batch = n_batch self.pos = 0 self.cuda = cuda self.internal_indices = np.arange(len(indices)).astype(np.int64) np.random.shuffle(self.internal_indices) def __iter__(self): return self def reset(self): self.pos = 0 np.random.shuffle(self.internal_indices) def __next__(self): start = self.pos end = np.minimum((self.pos + self.n_batch), len(self.indices)) self.pos += self.n_batch if (self.pos >= len(self.indices)): self.reset() raise StopIteration tensor = torch.LongTensor(self.indices[self.internal_indices[start:end]]) if (self.cuda is not None): tensor.cuda(self.cuda) return tensor def num_iterations(self): return (len(self.indices) // self.n_batch) next = __next__
def to_one_hot(x, n): x_ = torch.unsqueeze(x, 2) dims = (*x.size(), n) one_hot = torch.FloatTensor(*dims).zero_() one_hot.scatter_(2, x_, 1) return one_hot
def load_data(path, test_strat_id=None, cuda=None): '\n Loads the data\n\n path: path to the molecule .gz\n batch_size: size of a mini batch\n test_strat_id: id of strat being used as test set\n ' data = joblib.load(path) type_remap = (- np.ones((int(data['features']['atom_types'].max()) + 1))) unique_types = np.unique(data['features']['atom_types']).astype(int) type_remap[unique_types] = np.arange(len(unique_types)) data['features']['atom_types'] = type_remap[data['features']['atom_types'].astype(int)] data['features']['geometry'] = torch.FloatTensor(data['features']['geometry'].astype(np.float32)) data['features']['atom_types'] = torch.LongTensor(data['features']['atom_types'].astype(np.int64)) data['targets'] = torch.from_numpy(data['targets']) if (cuda is not None): data['features']['geometry'].cuda(cuda) data['features']['atom_types'].cuda(cuda) data['targets'].cuda(cuda) train = np.ndarray(0) test = np.ndarray(0) if (not test_strat_id): test_strat_id = np.random.randint(len(data['strats'])) for i in range(len(data['strats'])): if (i != test_strat_id): train = np.concatenate((train, data['strats'][i])) else: test = np.concatenate((test, data['strats'][i])) return (data, train, test)
def exp_lr_scheduler(optimizer, epoch, init_lr=0.005, lr_decay_epoch=40): 'Decay learning rate by a factor of 0.1 every lr_decay_epoch epochs.' lr = (init_lr * (0.1 ** (epoch // lr_decay_epoch))) if ((epoch % lr_decay_epoch) == 0): print('LR is set to {}'.format(lr)) for param_group in optimizer.param_groups: param_group['lr'] = lr return optimizer
def count_params(model): return sum([np.prod(p.size()) for p in model.parameters() if p.requires_grad])
class Model(nn.Module): def __init__(self, nclasses): super().__init__() self.features = [6, 100, 100, nclasses] self.bandwidths = [64, 16, 10] assert (len(self.bandwidths) == (len(self.features) - 1)) sequence = [] grid = s2_equatorial_grid(max_beta=0, n_alpha=(2 * self.bandwidths[0]), n_beta=1) sequence.append(S2Convolution(self.features[0], self.features[1], self.bandwidths[0], self.bandwidths[1], grid)) for l in range(1, (len(self.features) - 2)): nfeature_in = self.features[l] nfeature_out = self.features[(l + 1)] b_in = self.bandwidths[l] b_out = self.bandwidths[(l + 1)] sequence.append(nn.BatchNorm3d(nfeature_in, affine=True)) sequence.append(nn.ReLU()) grid = so3_equatorial_grid(max_beta=0, max_gamma=0, n_alpha=(2 * b_in), n_beta=1, n_gamma=1) sequence.append(SO3Convolution(nfeature_in, nfeature_out, b_in, b_out, grid)) sequence.append(nn.BatchNorm3d(self.features[(- 2)], affine=True)) sequence.append(nn.ReLU()) self.sequential = nn.Sequential(*sequence) output_features = self.features[(- 2)] self.out_layer = nn.Linear(output_features, self.features[(- 1)]) def forward(self, x): x = self.sequential(x) x = so3_integrate(x) x = self.out_layer(x) return F.log_softmax(x, dim=1)
class Model(nn.Module): def __init__(self, nclasses): super().__init__() self.features = [6, 50, 70, 350, nclasses] self.bandwidths = [128, 32, 22, 7] assert (len(self.bandwidths) == (len(self.features) - 1)) sequence = [] grid = s2_equatorial_grid(max_beta=0, n_alpha=(2 * self.bandwidths[0]), n_beta=1) sequence.append(S2Convolution(self.features[0], self.features[1], self.bandwidths[0], self.bandwidths[1], grid)) for l in range(1, (len(self.features) - 2)): nfeature_in = self.features[l] nfeature_out = self.features[(l + 1)] b_in = self.bandwidths[l] b_out = self.bandwidths[(l + 1)] sequence.append(nn.BatchNorm3d(nfeature_in, affine=True)) sequence.append(nn.ReLU()) grid = so3_equatorial_grid(max_beta=0, max_gamma=0, n_alpha=(2 * b_in), n_beta=1, n_gamma=1) sequence.append(SO3Convolution(nfeature_in, nfeature_out, b_in, b_out, grid)) sequence.append(nn.BatchNorm3d(self.features[(- 2)], affine=True)) sequence.append(nn.ReLU()) self.sequential = nn.Sequential(*sequence) self.out_layer = nn.Sequential(nn.BatchNorm1d(self.features[(- 2)], affine=False), nn.Linear(self.features[(- 2)], self.features[(- 1)])) def forward(self, x): x = self.sequential(x) x = x.view(x.size(0), x.size(1), (- 1)).max((- 1))[0] x = self.out_layer(x) return F.log_softmax(x, dim=1)
class KeepName(): def __init__(self, transform): self.transform = transform def __call__(self, file_name): return (file_name, self.transform(file_name))
def main(log_dir, augmentation, dataset, batch_size, num_workers): print(check_output(['nodejs', '--version']).decode('utf-8')) torch.backends.cudnn.benchmark = True transform = torchvision.transforms.Compose([CacheNPY(prefix='b64_', repeat=augmentation, pick_randomly=False, transform=torchvision.transforms.Compose([ToMesh(random_rotations=True, random_translation=0.1), ProjectOnSphere(bandwidth=64)])), (lambda xs: torch.stack([torch.FloatTensor(x) for x in xs]))]) transform = KeepName(transform) test_set = Shrec17('data', dataset, perturbed=True, download=True, transform=transform) loader = importlib.machinery.SourceFileLoader('model', os.path.join(log_dir, 'model.py')) mod = types.ModuleType(loader.name) loader.exec_module(mod) model = mod.Model(55) model.cuda() model.load_state_dict(torch.load(os.path.join(log_dir, 'state.pkl'))) resdir = os.path.join(log_dir, (dataset + '_perturbed')) if os.path.isdir(resdir): shutil.rmtree(resdir) os.mkdir(resdir) predictions = [] ids = [] loader = torch.utils.data.DataLoader(test_set, batch_size=batch_size, shuffle=False, num_workers=num_workers, pin_memory=True, drop_last=False) for (batch_idx, data) in enumerate(loader): model.eval() if (dataset != 'test'): data = data[0] (file_names, data) = data (batch_size, rep) = data.size()[:2] data = data.view((- 1), *data.size()[2:]) data = data.cuda() with torch.no_grad(): pred = model(data).data pred = pred.view(batch_size, rep, (- 1)) pred = pred.sum(1) predictions.append(pred.cpu().numpy()) ids.extend([x.split('/')[(- 1)].split('.')[0] for x in file_names]) print('[{}/{}] '.format(batch_idx, len(loader))) predictions = np.concatenate(predictions) predictions_class = np.argmax(predictions, axis=1) for i in range(len(ids)): if ((i % 100) == 0): print('{}/{} '.format(i, len(ids)), end='\r') idfile = os.path.join(resdir, ids[i]) retrieved = [(predictions[(j, predictions_class[j])], ids[j]) for j in range(len(ids)) if (predictions_class[j] == predictions_class[i])] retrieved = sorted(retrieved, reverse=True) retrieved = [i for (_, i) in retrieved] with open(idfile, 'w') as f: f.write('\n'.join(retrieved)) url = 'https://shapenet.cs.stanford.edu/shrec17/code/evaluator.zip' file_path = 'evaluator.zip' r = requests.get(url, stream=True) with open(file_path, 'wb') as f: for chunk in r.iter_content(chunk_size=(16 * (1024 ** 2))): if chunk: f.write(chunk) f.flush() zip_ref = zipfile.ZipFile(file_path, 'r') zip_ref.extractall('.') zip_ref.close() print(check_output(['nodejs', 'evaluate.js', (os.path.join('..', log_dir) + '/')], cwd='evaluator').decode('utf-8')) shutil.copy2(os.path.join('evaluator', (log_dir + '.summary.csv')), os.path.join(log_dir, 'summary.csv'))
def main(log_dir, model_path, augmentation, dataset, batch_size, learning_rate, num_workers): arguments = copy.deepcopy(locals()) os.mkdir(log_dir) shutil.copy2(__file__, os.path.join(log_dir, 'script.py')) shutil.copy2(model_path, os.path.join(log_dir, 'model.py')) logger = logging.getLogger('train') logger.setLevel(logging.DEBUG) logger.handlers = [] ch = logging.StreamHandler() logger.addHandler(ch) fh = logging.FileHandler(os.path.join(log_dir, 'log.txt')) logger.addHandler(fh) logger.info('%s', repr(arguments)) torch.backends.cudnn.benchmark = True loader = importlib.machinery.SourceFileLoader('model', os.path.join(log_dir, 'model.py')) mod = types.ModuleType(loader.name) loader.exec_module(mod) model = mod.Model(55) model.cuda() logger.info('{} paramerters in total'.format(sum((x.numel() for x in model.parameters())))) logger.info('{} paramerters in the last layer'.format(sum((x.numel() for x in model.out_layer.parameters())))) bw = model.bandwidths[0] transform = CacheNPY(prefix='b{}_'.format(bw), repeat=augmentation, transform=torchvision.transforms.Compose([ToMesh(random_rotations=True, random_translation=0.1), ProjectOnSphere(bandwidth=bw)])) def target_transform(x): classes = ['02691156', '02747177', '02773838', '02801938', '02808440', '02818832', '02828884', '02843684', '02871439', '02876657', '02880940', '02924116', '02933112', '02942699', '02946921', '02954340', '02958343', '02992529', '03001627', '03046257', '03085013', '03207941', '03211117', '03261776', '03325088', '03337140', '03467517', '03513137', '03593526', '03624134', '03636649', '03642806', '03691459', '03710193', '03759954', '03761084', '03790512', '03797390', '03928116', '03938244', '03948459', '03991062', '04004475', '04074963', '04090263', '04099429', '04225987', '04256520', '04330267', '04379243', '04401088', '04460130', '04468005', '04530566', '04554684'] return classes.index(x[0]) train_set = Shrec17('data', dataset, perturbed=True, download=True, transform=transform, target_transform=target_transform) train_loader = torch.utils.data.DataLoader(train_set, batch_size=batch_size, shuffle=True, num_workers=num_workers, pin_memory=True, drop_last=True) optimizer = torch.optim.SGD(model.parameters(), lr=0, momentum=0.9) def train_step(data, target): model.train() (data, target) = (data.cuda(), target.cuda()) prediction = model(data) loss = F.nll_loss(prediction, target) optimizer.zero_grad() loss.backward() optimizer.step() correct = prediction.data.max(1)[1].eq(target.data).long().cpu().sum() return (loss.item(), correct.item()) def get_learning_rate(epoch): limits = [100, 200] lrs = [1, 0.1, 0.01] assert (len(lrs) == (len(limits) + 1)) for (lim, lr) in zip(limits, lrs): if (epoch < lim): return (lr * learning_rate) return (lrs[(- 1)] * learning_rate) for epoch in range(300): lr = get_learning_rate(epoch) logger.info('learning rate = {} and batch size = {}'.format(lr, train_loader.batch_size)) for p in optimizer.param_groups: p['lr'] = lr total_loss = 0 total_correct = 0 time_before_load = time.perf_counter() for (batch_idx, (data, target)) in enumerate(train_loader): time_after_load = time.perf_counter() time_before_step = time.perf_counter() (loss, correct) = train_step(data, target) total_loss += loss total_correct += correct logger.info('[{}:{}/{}] LOSS={:.2} <LOSS>={:.2} ACC={:.2} <ACC>={:.2} time={:.2}+{:.2}'.format(epoch, batch_idx, len(train_loader), loss, (total_loss / (batch_idx + 1)), (correct / len(data)), ((total_correct / len(data)) / (batch_idx + 1)), (time_after_load - time_before_load), (time.perf_counter() - time_before_step))) time_before_load = time.perf_counter() torch.save(model.state_dict(), os.path.join(log_dir, 'state.pkl'))
def s2_near_identity_grid(max_beta=(np.pi / 8), n_alpha=8, n_beta=3): '\n :return: rings around the north pole\n size of the kernel = n_alpha * n_beta\n ' beta = ((np.arange(start=1, stop=(n_beta + 1), dtype=np.float) * max_beta) / n_beta) alpha = np.linspace(start=0, stop=(2 * np.pi), num=n_alpha, endpoint=False) (B, A) = np.meshgrid(beta, alpha, indexing='ij') B = B.flatten() A = A.flatten() grid = np.stack((B, A), axis=1) return tuple((tuple(ba) for ba in grid))
def s2_equatorial_grid(max_beta=0, n_alpha=32, n_beta=1): '\n :return: rings around the equator\n size of the kernel = n_alpha * n_beta\n ' beta = np.linspace(start=((np.pi / 2) - max_beta), stop=((np.pi / 2) + max_beta), num=n_beta, endpoint=True) alpha = np.linspace(start=0, stop=(2 * np.pi), num=n_alpha, endpoint=False) (B, A) = np.meshgrid(beta, alpha, indexing='ij') B = B.flatten() A = A.flatten() grid = np.stack((B, A), axis=1) return tuple((tuple(ba) for ba in grid))
def s2_soft_grid(b): beta = (((np.arange((2 * b)) + 0.5) / (2 * b)) * np.pi) alpha = np.linspace(start=0, stop=(2 * np.pi), num=(2 * b), endpoint=False) (B, A) = np.meshgrid(beta, alpha, indexing='ij') B = B.flatten() A = A.flatten() grid = np.stack((B, A), axis=1) return tuple((tuple(ba) for ba in grid))
def s2_mm(x, y): '\n :param x: [l * m, batch, feature_in, complex]\n :param y: [l * m, feature_in, feature_out, complex]\n :return: [l * m * n, batch, feature_out, complex]\n ' from s2cnn.utils.complex import complex_mm assert (y.size(3) == 2) assert (x.size(3) == 2) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) assert (y.size(1) == nfeature_in) nspec = x.size(0) assert (y.size(0) == nspec) if x.is_cuda: return _cuda_S2_mm.apply(x, y) nl = round((nspec ** 0.5)) Fz_list = [] begin = 0 for l in range(nl): L = ((2 * l) + 1) size = L Fx = x[begin:(begin + size)] Fy = y[begin:(begin + size)] Fx = Fx.view((L * nbatch), nfeature_in, 2) Fy = Fy.transpose(0, 1) Fy = Fy.contiguous() Fy = Fy.view(nfeature_in, (L * nfeature_out), 2) Fz = complex_mm(Fx, Fy, conj_y=True) Fz = Fz.view(L, nbatch, L, nfeature_out, 2) Fz = Fz.transpose(1, 2) Fz = Fz.contiguous() Fz = Fz.view((L * L), nbatch, nfeature_out, 2) Fz_list.append(Fz) begin += size z = torch.cat(Fz_list, 0) return z
class _cuda_S2_mm(torch.autograd.Function): @staticmethod def forward(ctx, x, y): ctx.save_for_backward(x, y) return _cuda_s2_mm(x, y) @staticmethod def backward(ctx, gradz): import s2cnn.utils.cuda as cuda_utils (x, y) = ctx.saved_tensors nl = round((x.size(0) ** 0.5)) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) nspec = ((((4 * (nl ** 2)) - 1) * nl) // 3) device = torch.cuda.current_device() gradx_cuda_kernel = _setup_s2mm_gradx_cuda_kernel(nbatch=nbatch, nspec=nspec, nl=nl, nfeature_in=nfeature_in, nfeature_out=nfeature_out, device=device) grady_cuda_kernel = _setup_s2mm_grady_cuda_kernel(nbatch=nbatch, nspec=nspec, nl=nl, nfeature_in=nfeature_in, nfeature_out=nfeature_out, device=device) stream = cuda_utils.Stream(ptr=torch.cuda.current_stream().cuda_stream) gradx = grady = None if ctx.needs_input_grad[0]: gradx = gradz.new_empty(((nl ** 2), nbatch, nfeature_in, 2)) gradx_cuda_kernel(block=(cuda_utils.CUDA_NUM_THREADS, 1, 1), grid=(cuda_utils.get_blocks((((nl ** 2) * nbatch) * nfeature_in), 1024), 1, 1), args=[gradz.contiguous().data_ptr(), y.contiguous().data_ptr(), gradx.data_ptr()], stream=stream) if ctx.needs_input_grad[1]: grady = gradz.new_empty(((nl ** 2), nfeature_in, nfeature_out, 2)) grady_cuda_kernel(block=(cuda_utils.CUDA_NUM_THREADS, 1, 1), grid=(cuda_utils.get_blocks((((nl ** 2) * nfeature_in) * nfeature_out), 1024), 1, 1), args=[gradz.contiguous().data_ptr(), x.contiguous().data_ptr(), grady.data_ptr()], stream=stream) return (gradx, grady)
def _cuda_s2_mm(x, y): '\n :param x: [l * m, batch, feature_in, complex]\n :param y: [l * m, feature_in, feature_out, complex]\n :return: [l * m * n, batch, feature_out, complex]\n ' import s2cnn.utils.cuda as cuda_utils assert (x.is_cuda and (x.dtype == torch.float32)) assert (y.is_cuda and (y.dtype == torch.float32)) assert (y.size(3) == 2) assert (x.size(3) == 2) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) assert (y.size(1) == nfeature_in) assert (y.size(0) == x.size(0)) nl = round((x.size(0) ** 0.5)) nspec = ((((4 * (nl ** 2)) - 1) * nl) // 3) assert (x.size(0) == (nl ** 2)) assert (y.size(0) == (nl ** 2)) device = torch.cuda.current_device() cuda_kernel = _setup_s2mm_cuda_kernel(nbatch=nbatch, nspec=nspec, nfeature_in=nfeature_in, nfeature_out=nfeature_out, device=device) stream = cuda_utils.Stream(ptr=torch.cuda.current_stream().cuda_stream) output = x.new_empty((nspec, nbatch, nfeature_out, 2)) cuda_kernel(block=(cuda_utils.CUDA_NUM_THREADS, 1, 1), grid=(cuda_utils.get_blocks(((nspec * nbatch) * nfeature_out), 1024), 1, 1), args=[x.contiguous().data_ptr(), y.contiguous().data_ptr(), output.data_ptr()], stream=stream) return output
@lru_cache(maxsize=32) def _setup_s2mm_cuda_kernel(nbatch, nspec, nfeature_in, nfeature_out, device=0): kernel = Template('\n#define COMPUTE_LMN(s) int l = powf(3.0/4.0 * s, 1.0/3.0) - 0.5; int L = l * (4 * l * l - 1) / 3; int rest = s - L; if (rest >= (2 * l + 1) * (2 * l + 1)) { ++l; L = l * (4 * l * l - 1) / 3; rest = s - L; } int m = rest / (2 * l + 1) - l; int n = rest % (2 * l + 1) - l;\n\n#define EXTRACT(i1, i2, n2, i3, n3) int i1 = index; int i3 = i1 % (n3); i1 /= n3; int i2 = i1 % (n2); i1 /= n2;\n\n#define CONTRACT1(s1, i2, n2, i3, n3) ( ( (l * l + (l + (s1))) * (n2) + (i2) ) * (n3) + (i3) )\n\n#define CONTRACT2(s1, s2, i2, n2, i3, n3) ( ( (L + (l + (s1)) * (2 * l + 1) + (l + (s2))) * (n2) + (i2) ) * (n3) + (i3) )\n\nextern "C"\n__global__ void main_(const float* in_x, const float* in_y, float* out) {\n for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < ${nspec} * ${nbatch} * ${nfeature_out}; index += blockDim.x * gridDim.x) {\n EXTRACT(s, i, ${nbatch}, f_out, ${nfeature_out})\n\n // compute s -> (l,m,n)\n COMPUTE_LMN(s)\n\n float out_re = 0.0;\n float out_im = 0.0;\n\n for (int f_in = 0; f_in < ${nfeature_in}; ++f_in) {\n float x_re = in_x[CONTRACT1(m, i, ${nbatch}, f_in, ${nfeature_in} ) * 2 + 0];\n float x_im = in_x[CONTRACT1(m, i, ${nbatch}, f_in, ${nfeature_in} ) * 2 + 1];\n float y_re = in_y[CONTRACT1(n, f_in, ${nfeature_in}, f_out, ${nfeature_out}) * 2 + 0];\n float y_im = in_y[CONTRACT1(n, f_in, ${nfeature_in}, f_out, ${nfeature_out}) * 2 + 1];\n\n // x times y conjugate\n out_re += x_re * y_re + x_im * y_im;\n out_im += x_im * y_re - x_re * y_im;\n }\n\n out[index * 2 + 0] = out_re;\n out[index * 2 + 1] = out_im;\n }\n}\n').substitute({'nbatch': nbatch, 'nspec': nspec, 'nfeature_in': nfeature_in, 'nfeature_out': nfeature_out}) import s2cnn.utils.cuda as cuda_utils return cuda_utils.compile_kernel(kernel, 's2mm.cu', 'main_')
@lru_cache(maxsize=32) def _setup_s2mm_gradx_cuda_kernel(nbatch, nspec, nl, nfeature_in, nfeature_out, device=0): kernel = Template('\n#define COMPUTE_LM(s) int l = sqrtf(s); int L = (4 * l * l - 1) * l / 3; int m = s - l * l - l;\n\n#define EXTRACT(i1, i2, n2, i3, n3) int i1 = index; int i3 = i1 % (n3); i1 /= n3; int i2 = i1 % (n2); i1 /= n2;\n\n#define CONTRACT1(s1, i2, n2, i3, n3) ( ( (l * l + (l + (s1))) * (n2) + (i2) ) * (n3) + (i3) )\n\n#define CONTRACT2(s1, s2, i2, n2, i3, n3) ( ( (L + (l + (s1)) * (2 * l + 1) + (l + (s2))) * (n2) + (i2) ) * (n3) + (i3) )\n\nextern "C"\n__global__ void main_(const float* grad_z, const float* y, float* grad_x) {\n for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < (${nl} * ${nl}) * ${nbatch} * ${nfeature_in}; index += blockDim.x * gridDim.x) {\n EXTRACT(s, i, ${nbatch}, f_in, ${nfeature_in})\n\n // compute s -> (l,m)\n COMPUTE_LM(s)\n\n float out_re = 0.0;\n float out_im = 0.0;\n\n for (int f_out = 0; f_out < ${nfeature_out}; ++f_out) {\n for (int k = -l; k <= l; ++k) {\n float grad_z_re = grad_z[CONTRACT2(m, k, i, ${nbatch}, f_out, ${nfeature_out}) * 2 + 0];\n float grad_z_im = grad_z[CONTRACT2(m, k, i, ${nbatch}, f_out, ${nfeature_out}) * 2 + 1];\n float y_re = y[CONTRACT1(k, f_in, ${nfeature_in}, f_out, ${nfeature_out}) * 2 + 0];\n float y_im = y[CONTRACT1(k, f_in, ${nfeature_in}, f_out, ${nfeature_out}) * 2 + 1];\n\n // grad_z times y\n out_re += grad_z_re * y_re - grad_z_im * y_im;\n out_im += grad_z_re * y_im + grad_z_im * y_re;\n }\n }\n\n grad_x[index * 2 + 0] = out_re;\n grad_x[index * 2 + 1] = out_im;\n }\n}\n').substitute({'nbatch': nbatch, 'nspec': nspec, 'nl': nl, 'nfeature_in': nfeature_in, 'nfeature_out': nfeature_out}) import s2cnn.utils.cuda as cuda_utils return cuda_utils.compile_kernel(kernel, 's2mm_gradx.cu', 'main_')
@lru_cache(maxsize=32) def _setup_s2mm_grady_cuda_kernel(nbatch, nspec, nl, nfeature_in, nfeature_out, device=0): kernel = Template('\n#define COMPUTE_LM(s) int l = powf(s, 0.5); int L = (4 * l * l - 1) * l / 3; int m = s - l * l - l;\n\n#define EXTRACT(i1, i2, n2, i3, n3) int i1 = index; int i3 = i1 % (n3); i1 /= n3; int i2 = i1 % (n2); i1 /= n2;\n\n#define CONTRACT1(s1, i2, n2, i3, n3) ( ( (l * l + (l + (s1))) * (n2) + (i2) ) * (n3) + (i3) )\n\n#define CONTRACT2(s1, s2, i2, n2, i3, n3) ( ( (L + (l + (s1)) * (2 * l + 1) + (l + (s2))) * (n2) + (i2) ) * (n3) + (i3) )\n\nextern "C"\n__global__ void main_(const float* grad_z, const float* x, float* grad_y) {\n for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < (${nl} * ${nl}) * ${nfeature_in} * ${nfeature_out}; index += blockDim.x * gridDim.x) {\n EXTRACT(s, f_in, ${nfeature_in}, f_out, ${nfeature_out})\n\n // compute s -> (l,m)\n COMPUTE_LM(s)\n\n float out_re = 0.0;\n float out_im = 0.0;\n\n for (int i = 0; i < ${nbatch}; ++i) {\n for (int k = -l; k <= l; ++k) {\n float grad_z_re = grad_z[CONTRACT2(k, m, i, ${nbatch}, f_out, ${nfeature_out}) * 2 + 0];\n float grad_z_im = grad_z[CONTRACT2(k, m, i, ${nbatch}, f_out, ${nfeature_out}) * 2 + 1];\n float x_re = x[CONTRACT1(k, i, ${nbatch}, f_in, ${nfeature_in} ) * 2 + 0];\n float x_im = x[CONTRACT1(k, i, ${nbatch}, f_in, ${nfeature_in} ) * 2 + 1];\n\n // conjugate grad_z times x\n out_re += grad_z_re * x_re + grad_z_im * x_im;\n out_im += grad_z_re * x_im - grad_z_im * x_re;\n }\n }\n\n grad_y[index * 2 + 0] = out_re;\n grad_y[index * 2 + 1] = out_im;\n }\n}\n').substitute({'nbatch': nbatch, 'nspec': nspec, 'nl': nl, 'nfeature_in': nfeature_in, 'nfeature_out': nfeature_out}) import s2cnn.utils.cuda as cuda_utils return cuda_utils.compile_kernel(kernel, 's2mm_grady.cu', 'main_')
def test_compare_cuda_cpu(): x = torch.rand((((1 + 3) + 5) + 7), 2, 3, 2) y = torch.rand((((1 + 3) + 5) + 7), 3, 5, 2) z1 = s2_mm(x, y) z2 = s2_mm(x.cuda(), y.cuda()).cpu() q = ((z1 - z2).abs().max().item() / z1.std().item()) print(q) assert (q < 0.0001)
def so3_rft(x, b, grid): '\n Real Fourier Transform\n :param x: [..., beta_alpha_gamma]\n :param b: output bandwidth signal\n :param grid: tuple of (beta, alpha, gamma) tuples\n :return: [l * m * n, ..., complex]\n ' F = _setup_so3_ft(b, grid, device_type=x.device.type, device_index=x.device.index) assert (x.size((- 1)) == F.size(0)) sz = x.size() x = torch.einsum('ia,afc->fic', (x.view((- 1), x.size((- 1))), F.clone())) x = x.view((- 1), *sz[:(- 1)], 2) return x
@cached_dirpklgz('cache/setup_so3_ft') def __setup_so3_ft(b, grid): from lie_learn.representations.SO3.wigner_d import wigner_D_matrix n_spatial = len(grid) n_spectral = np.sum([(((2 * l) + 1) ** 2) for l in range(b)]) F = np.zeros((n_spatial, n_spectral), dtype=complex) for (i, (beta, alpha, gamma)) in enumerate(grid): Dmats = [wigner_D_matrix(l, alpha, beta, gamma, field='complex', normalization='quantum', order='centered', condon_shortley='cs').conj() for l in range(b)] F[i] = np.hstack([Dl.flatten() for Dl in Dmats]) F = F.view('float').reshape(((- 1), n_spectral, 2)) return F
@lru_cache(maxsize=32) def _setup_so3_ft(b, grid, device_type, device_index): F = __setup_so3_ft(b, grid) F = torch.tensor(F.astype(np.float32), dtype=torch.float32, device=torch.device(device_type, device_index)) return F
def so3_mm(x, y): '\n :param x: [l * m * n, batch, feature_in, complex]\n :param y: [l * m * n, feature_in, feature_out, complex]\n :return: [l * m * n, batch, feature_out, complex]\n ' from s2cnn.utils.complex import complex_mm import math assert (y.size(3) == 2) assert (x.size(3) == 2) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) assert (y.size(1) == nfeature_in) nspec = x.size(0) assert (y.size(0) == nspec) nl = math.ceil((((3 / 4) * nspec) ** (1 / 3))) assert (nspec == ((nl * ((4 * (nl ** 2)) - 1)) // 3)) if x.is_cuda: return _cuda_SO3_mm.apply(x, y) Fz_list = [] begin = 0 for l in range(nl): L = ((2 * l) + 1) size = (L ** 2) Fx = x[begin:(begin + size)] Fy = y[begin:(begin + size)] Fx = Fx.view(L, L, nbatch, nfeature_in, 2) Fx = Fx.transpose(0, 1) Fx = Fx.transpose(0, 2) Fx = Fx.transpose(2, 3) Fx = Fx.contiguous() Fx = Fx.view((nbatch * L), (nfeature_in * L), 2) Fy = Fy.view(L, L, nfeature_in, nfeature_out, 2) Fy = Fy.transpose(0, 2) Fy = Fy.contiguous() Fy = Fy.view((nfeature_in * L), (L * nfeature_out), 2) Fz = complex_mm(Fx, Fy, conj_y=True) Fz = Fz.view(nbatch, (L * L), nfeature_out, 2) Fz = Fz.transpose(0, 1) Fz_list.append(Fz) begin += size z = torch.cat(Fz_list, 0) return z
class _cuda_SO3_mm(torch.autograd.Function): @staticmethod def forward(ctx, x, y): '\n :param x: [l * m * n, batch, feature_in, complex]\n :param y: [l * m * n, feature_in, feature_out, complex]\n :return: [l * m * n, batch, feature_out, complex]\n ' assert (x.is_cuda and (x.dtype == torch.float32)) assert (y.is_cuda and (y.dtype == torch.float32)) assert (y.size(3) == 2) assert (x.size(3) == 2) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) assert (y.size(1) == nfeature_in) nspec = x.size(0) assert (y.size(0) == nspec) nl = round((((3 / 4) * nspec) ** (1 / 3))) assert (nspec == ((nl * ((4 * (nl ** 2)) - 1)) // 3)) ctx.save_for_backward(x, y) device = torch.cuda.current_device() cuda_kernel = _setup_so3mm_cuda_kernel(nl=nl, ni=nbatch, nj=nfeature_out, nk=nfeature_in, conj_y=True, trans_y_spec=True, device=device) output = x.new_empty((nspec, nbatch, nfeature_out, 2)) cuda_kernel(x, y, output) return output @staticmethod def backward(ctx, gradz): (x, y) = ctx.saved_tensors nspec = x.size(0) nbatch = x.size(1) nfeature_in = x.size(2) nfeature_out = y.size(2) nl = round((((3 / 4) * nspec) ** (1 / 3))) assert (nspec == ((nl * ((4 * (nl ** 2)) - 1)) // 3)) gradx = grady = None device = torch.cuda.current_device() if ctx.needs_input_grad[0]: gradx_cuda_kernel = _setup_so3mm_cuda_kernel(nl=nl, ni=nbatch, nj=nfeature_in, nk=nfeature_out, trans_y_feature=True, device=device) gradx = gradz.new_empty((nspec, nbatch, nfeature_in, 2)) gradx_cuda_kernel(gradz, y, gradx) if ctx.needs_input_grad[1]: grady_cuda_kernel = _setup_so3mm_cuda_kernel(nl=nl, ni=nfeature_out, nj=nfeature_in, nk=nbatch, trans_out_feature=True, conj_x=True, trans_x_spec=True, trans_x_feature=True, device=device) grady = gradz.new_empty((nspec, nfeature_in, nfeature_out, 2)) grady_cuda_kernel(gradz, x, grady) return (gradx, grady)
@lru_cache(maxsize=32) def _setup_so3mm_cuda_kernel(nl, ni, nj, nk, conj_x=False, conj_y=False, trans_x_spec=False, trans_x_feature=False, trans_y_spec=False, trans_y_feature=False, trans_out_feature=False, device=0): '\n return a function that computes\n out[l*m*n, i, j] = sum_k sum_p x[l*m*p, i, k] y[l*p*n, k, j]\n where out, x, y are complex valued\n\n if conj_x is set to True, x is conjugated\n if conj_y is set to True, y is conjugated\n if trans_x_spec is set to True m and p are permuted in x[...]\n if trans_y_spec is set to True p and n are permuted in y[...]\n if trans_x_feature is set to True i and k are permuted in x[...]\n if trans_y_feature is set to True k and j are permuted in y[...]\n if trans_out_feature is set to True i and j are permuted in out[...]\n ' kernel = '\n#define NI {}\n#define NJ {}\n#define NK {}\n'.format(ni, nj, nk) if ((not trans_x_spec) and (not trans_x_feature)): kernel += '#define INDEX_X (((L0 + m * L + p) * NI + i) * NK + k)\n' if ((not trans_x_spec) and trans_x_feature): kernel += '#define INDEX_X (((L0 + m * L + p) * NK + k) * NI + i)\n' if (trans_x_spec and (not trans_x_feature)): kernel += '#define INDEX_X (((L0 + p * L + m) * NI + i) * NK + k)\n' if (trans_x_spec and trans_x_feature): kernel += '#define INDEX_X (((L0 + p * L + m) * NK + k) * NI + i)\n' if ((not trans_y_spec) and (not trans_y_feature)): kernel += '#define INDEX_Y (((L0 + p * L + n) * NK + k) * NJ + j)\n' if ((not trans_y_spec) and trans_y_feature): kernel += '#define INDEX_Y (((L0 + p * L + n) * NJ + j) * NK + k)\n' if (trans_y_spec and (not trans_y_feature)): kernel += '#define INDEX_Y (((L0 + n * L + p) * NK + k) * NJ + j)\n' if (trans_y_spec and trans_y_feature): kernel += '#define INDEX_Y (((L0 + n * L + p) * NJ + j) * NK + k)\n' if (not trans_out_feature): kernel += '#define INDEX_OUT (((L0 + m * L + n) * NI + i) * NJ + j)\n' if trans_out_feature: kernel += '#define INDEX_OUT (((L0 + m * L + n) * NJ + j) * NI + i)\n' kernel += '\n#define CONJ_X {}\n#define CONJ_Y {}\n'.format(('x_im = -x_im;' if conj_x else ';'), ('y_im = -y_im;' if conj_y else ';')) kernel += '\n#define CEIL_DIV(x, y) (((x) + (y) - 1) / (y))\n\nextern "C"\n__global__ void main_(const float* in_x, const float* in_y, float* out)\n{\n // start of thread independant code\n int l = blockIdx.z;\n int L = 2 * l + 1;\n int L0 = (4 * l*l - 1) * l / 3;\n\n if (blockIdx.y * 32 >= L * NI || blockIdx.x * 32 >= L * NJ) {\n return;\n }\n\n int ntile = CEIL_DIV(L * NK, 32);\n // end of thread independant code\n\n int mi = blockIdx.y * 32 + threadIdx.y;\n int m = mi / NI;\n int i = mi % NI;\n int nj = blockIdx.x * 32 + threadIdx.x;\n int n = nj / NJ;\n int j = nj % NJ;\n\n float sum_re = 0.0;\n float sum_im = 0.0;\n\n for (int tile = 0; tile < ntile; ++tile) {\n __shared__ float tileX[2][32][32];\n __shared__ float tileY[2][32][32];\n\n int pk = tile * 32 + threadIdx.x;\n int p = pk / NK;\n int k = pk % NK;\n int index = INDEX_X * 2;\n tileX[0][threadIdx.y][threadIdx.x] = m < L && p < L ? in_x[index + 0] : 0.0;\n tileX[1][threadIdx.y][threadIdx.x] = m < L && p < L ? in_x[index + 1] : 0.0;\n\n pk = tile * 32 + threadIdx.y;\n p = pk / NK;\n k = pk % NK;\n index = INDEX_Y * 2;\n tileY[0][threadIdx.y][threadIdx.x] = p < L && n < L ? in_y[index + 0] : 0.0;\n tileY[1][threadIdx.y][threadIdx.x] = p < L && n < L ? in_y[index + 1] : 0.0;\n\n __syncthreads();\n\n for (int any = 0; any < 32; ++any) {\n float x_re = tileX[0][threadIdx.y][any];\n float x_im = tileX[1][threadIdx.y][any];\n float y_re = tileY[0][any][threadIdx.x];\n float y_im = tileY[1][any][threadIdx.x];\n\n CONJ_X\n CONJ_Y\n\n sum_re += x_re * y_re - x_im * y_im;\n sum_im += x_re * y_im + x_im * y_re;\n }\n\n __syncthreads();\n }\n\n if (m < L && n < L) {\n int index = INDEX_OUT * 2;\n out[index + 0] = sum_re;\n out[index + 1] = sum_im;\n }\n}\n' import s2cnn.utils.cuda as cuda_utils kernel = cuda_utils.compile_kernel(kernel, 'so3_mm.cu', 'main_') stream = cuda_utils.Stream(ptr=torch.cuda.current_stream().cuda_stream) def fun(x, y, output): assert output.is_contiguous() kernel(block=(32, 32, 1), grid=(math.ceil(((((2 * nl) - 1) * nj) / 32)), math.ceil(((((2 * nl) - 1) * ni) / 32)), nl), args=[x.contiguous().data_ptr(), y.contiguous().data_ptr(), output.data_ptr()], stream=stream) return fun
def test_compare_cuda_cpu(): x = torch.rand((((1 + 9) + 25) + 49), 2, 3, 2) y = torch.rand((((1 + 9) + 25) + 49), 3, 5, 2) z1 = so3_mm(x, y) z2 = so3_mm(x.cuda(), y.cuda()).cpu() q = ((z1 - z2).abs().max().item() / z1.std().item()) print(q) assert (q < 0.0001)
class S2Convolution(Module): def __init__(self, nfeature_in, nfeature_out, b_in, b_out, grid): "\n :param nfeature_in: number of input fearures\n :param nfeature_out: number of output features\n :param b_in: input bandwidth (precision of the input SOFT grid)\n :param b_out: output bandwidth\n :param grid: points of the sphere defining the kernel, tuple of (alpha, beta)'s\n " super(S2Convolution, self).__init__() self.nfeature_in = nfeature_in self.nfeature_out = nfeature_out self.b_in = b_in self.b_out = b_out self.grid = grid self.kernel = Parameter(torch.empty(nfeature_in, nfeature_out, len(grid)).uniform_((- 1), 1)) self.scaling = (1.0 / math.sqrt((((len(self.grid) * self.nfeature_in) * (self.b_out ** 4.0)) / (self.b_in ** 2.0)))) self.bias = Parameter(torch.zeros(1, nfeature_out, 1, 1, 1)) def forward(self, x): '\n :x: [batch, feature_in, beta, alpha]\n :return: [batch, feature_out, beta, alpha, gamma]\n ' assert (x.size(1) == self.nfeature_in) assert (x.size(2) == (2 * self.b_in)) assert (x.size(3) == (2 * self.b_in)) x = S2_fft_real.apply(x, self.b_out) y = s2_rft((self.kernel * self.scaling), self.b_out, self.grid) z = s2_mm(x, y) z = SO3_ifft_real.apply(z) z = (z + self.bias) return z
class SO3Convolution(Module): def __init__(self, nfeature_in, nfeature_out, b_in, b_out, grid): "\n :param nfeature_in: number of input fearures\n :param nfeature_out: number of output features\n :param b_in: input bandwidth (precision of the input SOFT grid)\n :param b_out: output bandwidth\n :param grid: points of the SO(3) group defining the kernel, tuple of (alpha, beta, gamma)'s\n " super(SO3Convolution, self).__init__() self.nfeature_in = nfeature_in self.nfeature_out = nfeature_out self.b_in = b_in self.b_out = b_out self.grid = grid self.kernel = Parameter(torch.empty(nfeature_in, nfeature_out, len(grid)).uniform_((- 1), 1)) self.bias = Parameter(torch.zeros(1, nfeature_out, 1, 1, 1)) self.scaling = (1.0 / math.sqrt((((len(self.grid) * self.nfeature_in) * (self.b_out ** 3.0)) / (self.b_in ** 3.0)))) def forward(self, x): '\n :x: [batch, feature_in, beta, alpha, gamma]\n :return: [batch, feature_out, beta, alpha, gamma]\n ' assert (x.size(1) == self.nfeature_in) assert (x.size(2) == (2 * self.b_in)) assert (x.size(3) == (2 * self.b_in)) assert (x.size(4) == (2 * self.b_in)) x = SO3_fft_real.apply(x, self.b_out) y = so3_rft((self.kernel * self.scaling), self.b_out, self.grid) assert (x.size(0) == y.size(0)) assert (x.size(2) == y.size(1)) z = so3_mm(x, y) assert (z.size(0) == x.size(0)) assert (z.size(1) == x.size(1)) assert (z.size(2) == y.size(2)) z = SO3_ifft_real.apply(z) z = (z + self.bias) return z
class SO3Shortcut(Module): '\n Useful for ResNet\n ' def __init__(self, nfeature_in, nfeature_out, b_in, b_out): super(SO3Shortcut, self).__init__() assert (b_out <= b_in) if ((nfeature_in != nfeature_out) or (b_in != b_out)): self.conv = SO3Convolution(nfeature_in=nfeature_in, nfeature_out=nfeature_out, b_in=b_in, b_out=b_out, grid=((0, 0, 0),)) else: self.conv = None def forward(self, x): '\n :x: [batch, feature_in, beta, alpha, gamma]\n :return: [batch, feature_out, beta, alpha, gamma]\n ' if (self.conv is not None): return self.conv(x) else: return x
def so3_integrate(x): '\n Integrate a signal on SO(3) using the Haar measure\n \n :param x: [..., beta, alpha, gamma] (..., 2b, 2b, 2b)\n :return y: [...] (...)\n ' assert (x.size((- 1)) == x.size((- 2))) assert (x.size((- 2)) == x.size((- 3))) b = (x.size((- 1)) // 2) w = _setup_so3_integrate(b, device_type=x.device.type, device_index=x.device.index) x = torch.sum(x, dim=(- 1)).squeeze((- 1)) x = torch.sum(x, dim=(- 1)).squeeze((- 1)) sz = x.size() x = x.view((- 1), (2 * b)) w = w.view((2 * b), 1) x = torch.mm(x, w).squeeze((- 1)) x = x.view(*sz[:(- 1)]) return x
@lru_cache(maxsize=32) @show_running def _setup_so3_integrate(b, device_type, device_index): import lie_learn.spaces.S3 as S3 return torch.tensor(S3.quadrature_weights(b), dtype=torch.float32, device=torch.device(device_type, device_index))
def so3_rotation(x, alpha, beta, gamma): '\n :param x: [..., beta, alpha, gamma] (..., 2b, 2b, 2b)\n ' b = (x.size()[(- 1)] // 2) x_size = x.size() Us = _setup_so3_rotation(b, alpha, beta, gamma, device_type=x.device.type, device_index=x.device.index) x = SO3_fft_real.apply(x) Fz_list = [] begin = 0 for l in range(b): L = ((2 * l) + 1) size = (L ** 2) Fx = x[begin:(begin + size)] Fx = Fx.view(L, (- 1), 2) U = Us[l].view(L, L, 2) Fz = complex_mm(U, Fx, conj_x=True) Fz = Fz.view(size, (- 1), 2) Fz_list.append(Fz) begin += size Fz = torch.cat(Fz_list, 0) z = SO3_ifft_real.apply(Fz) z = z.contiguous() z = z.view(*x_size) return z
@cached_dirpklgz('cache/setup_so3_rotation') def __setup_so3_rotation(b, alpha, beta, gamma): from lie_learn.representations.SO3.wigner_d import wigner_D_matrix Us = [wigner_D_matrix(l, alpha, beta, gamma, field='complex', normalization='quantum', order='centered', condon_shortley='cs') for l in range(b)] Us = [Us[l].astype(np.complex64).view(np.float32).reshape((((2 * l) + 1), ((2 * l) + 1), 2)) for l in range(b)] return Us
@lru_cache(maxsize=32) def _setup_so3_rotation(b, alpha, beta, gamma, device_type, device_index): Us = __setup_so3_rotation(b, alpha, beta, gamma) Us = [torch.tensor(U, dtype=torch.float32, device=torch.device(device_type, device_index)) for U in Us] return Us
def get_blocks(n, num_threads): n_per_instance = (((n + (num_threads * CUDA_MAX_GRID_DIM)) - 1) // (num_threads * CUDA_MAX_GRID_DIM)) return (((n + (num_threads * n_per_instance)) - 1) // (num_threads * n_per_instance))
def compile_kernel(kernel, filename, functioname): program = Program(kernel, filename) ptx = program.compile() m = function.Module() m.load(bytes(ptx.encode())) f = m.get_function(functioname) return f
class WaitPrint(threading.Thread): def __init__(self, t, message): super().__init__() self.t = t self.message = message self.running = True def stop(self): self.running = False def run(self): for _ in range(int((self.t // 0.1))): time.sleep(0.1) if (not self.running): return print(self.message, end='')
def show_running(func): @wraps(func) def g(*args, **kargs): x = WaitPrint(2, '{}({})... '.format(func.__name__, ', '.join(([repr(x) for x in args] + ['{}={}'.format(key, repr(value)) for (key, value) in kargs.items()])))) x.start() t = time.perf_counter() r = func(*args, **kargs) if x.is_alive(): x.stop() else: print('done in {:.0f} seconds'.format((time.perf_counter() - t))) return r return g
def cached_dirpklgz(dirname): '\n Cache a function with a directory\n ' def decorator(func): '\n The actual decorator\n ' @lru_cache(maxsize=None) @wraps(func) def wrapper(*args): '\n The wrapper of the function\n ' try: os.makedirs(dirname) except FileExistsError: pass indexfile = os.path.join(dirname, 'index.pkl') try: with open(indexfile, 'rb') as file: index = pickle.load(file) except FileNotFoundError: index = {} try: filename = index[args] except KeyError: index[args] = filename = '{}.pkl.gz'.format(len(index)) with open(indexfile, 'wb') as file: pickle.dump(index, file) filepath = os.path.join(dirname, filename) try: with gzip.open(filepath, 'rb') as file: print('load {}... '.format(filename), end='') result = pickle.load(file) except FileNotFoundError: print('compute {}... '.format(filename), end='') sys.stdout.flush() result = func(*args) print('save {}... '.format(filename), end='') with gzip.open(filepath, 'wb') as file: pickle.dump(result, file) print('done') return result return wrapper return decorator
def test_so3_rfft(b_in, b_out, device): x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), dtype=torch.float, device=device) from s2cnn.soft.so3_fft import so3_rfft y1 = so3_rfft(x, b_out=b_out) from s2cnn import so3_rft, so3_soft_grid import lie_learn.spaces.S3 as S3 weights = torch.tensor(S3.quadrature_weights(b_in), dtype=torch.float, device=device) x2 = torch.einsum('bac,b->bac', (x, weights)) y2 = so3_rft(x2.view((- 1)), b_out, so3_soft_grid(b_in)) assert ((y1 - y2).abs().max().item() < (0.0001 * y1.abs().mean().item()))
def test_inverse(f, g, b_in, b_out, device, complex): if complex: x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), 2, dtype=torch.float, device=device) else: x = torch.randn((2 * b_in), (2 * b_in), (2 * b_in), dtype=torch.float, device=device) x = g(f(x, b_out=b_out), b_out=b_in) y = g(f(x, b_out=b_out), b_out=b_in) assert ((x - y).abs().max().item() < (0.0001 * y.abs().mean().item()))
def test_inverse2(f, g, b_in, b_out, device): x = torch.randn(((b_in * ((4 * (b_in ** 2)) - 1)) // 3), 2, dtype=torch.float, device=device) x = g(f(x, b_out=b_out), b_out=b_in) y = g(f(x, b_out=b_out), b_out=b_in) assert ((x - y).abs().max().item() < (0.0001 * y.abs().mean().item()))
def compare_cpu_gpu(f, x): z1 = f(x.cpu()) z2 = f(x.cuda()).cpu() q = ((z1 - z2).abs().max().item() / z1.std().item()) assert (q < 0.0001)
def get_filenames_of_path(path: pathlib.Path, ext: str='*'): 'Returns a list of files in a directory/path. Uses pathlib.' filenames = [file for file in path.glob(ext) if file.is_file()] return filenames
def compute_max_depth(shape, max_depth=10, print_out=True): shapes = [] shapes.append(shape) for level in range(1, max_depth): if (((shape % (2 ** level)) == 0) and ((shape / (2 ** level)) > 1)): shapes.append((shape / (2 ** level))) if print_out: print(f'Level {level}: {(shape / (2 ** level))}') else: if print_out: print(f'Max-level: {(level - 1)}') break return shapes
def compute_possible_shapes(low, high, depth): possible_shapes = {} for shape in range(low, (high + 1)): shapes = compute_max_depth(shape, max_depth=depth, print_out=False) if (len(shapes) == depth): possible_shapes[shape] = shapes return possible_shapes
def get_filenames_of_path(path: pathlib.Path, ext: str='*'): 'Returns a list of files in a directory/path. Uses pathlib.' filenames = [file for file in path.glob(ext) if file.is_file()] return filenames
def get_filenames_of_path(path: pathlib.Path, ext: str='*'): 'Returns a list of files in a directory/path. Uses pathlib.' filenames = [file for file in path.glob(ext) if file.is_file()] return filenames
def preprocess(img: np.ndarray): img = np.moveaxis(img, (- 1), 0) img = normalize_01(img) img = np.expand_dims(img, axis=0) img = img.astype(np.float32) return img
def postprocess(img: torch.tensor): img = torch.argmax(img, dim=1) img = img.cpu().numpy() img = np.squeeze(img) img = re_normalize(img) return img
def get_filenames_of_path(path: pathlib.Path, ext: str='*'): 'Returns a list of files in a directory/path. Uses pathlib.' filenames = [file for file in path.glob(ext) if file.is_file()] return filenames
def get_filenames_of_path(path: pathlib.Path, ext: str='*'): 'Returns a list of files in a directory/path. Uses pathlib.' filenames = [file for file in path.glob(ext) if file.is_file()] return filenames
class DatasetViewerExtra(DatasetViewer): def show_sample(self): sample = self.get_sample_dataset(self.index) (x, y, x_name, y_name) = (sample['x'], sample['y'], sample['x_name'], sample['y_name']) x = self.transform_x(x) y = self.transform_y(y) if (self.image_layer not in self.viewer.layers): self.image_layer = self.create_image_layer(x, x_name) else: self.update_image_layer(self.image_layer, x, x_name) if (self.label_layer not in self.viewer.layers): self.label_layer = self.create_label_layer(y, y_name) else: self.update_label_layer(self.label_layer, y, y_name) self.viewer.reset_view()
def get_k_highest_values(scores, k): return np.argpartition(np.array(scores), (- k))[(- k):]
def get_k_lowest_values(scores, k): return np.argpartition(np.array(scores), k)[:k]
def log_k_worst_best_scores(metric_obj, k): import itertools scores = metric_obj.get_metrics_epoch(last=True, transpose=False).numpy() names = np.array(list(itertools.chain.from_iterable(metric_obj.last_names))) k_lowest = get_k_lowest_values(scores, k=k) k_highest = get_k_highest_values(scores, k=k) df_lowest = pd.DataFrame({f'{metric_obj}': scores[k_lowest], 'name': names[k_lowest]}) df_highest = pd.DataFrame({f'{metric_obj}': scores[k_highest], 'name': names[k_highest]}) log_table(name=f'{metric_obj}-lowest', table=df_lowest, experiment=neptune_logger.experiment) log_table(name=f'{metric_obj}-highest', table=df_highest, experiment=neptune_logger.experiment) return (df_lowest, df_highest)