code
stringlengths
101
5.91M
def as_dict_handler(obj: Any) -> (dict[(str, Any)] | None): try: return obj.as_dict() except AttributeError: return None
def test_diskdf_sample(): from galpy.df import dehnendf, shudf (ro, vo) = (7.0, 230.0) df = dehnendf(ro=ro, vo=vo) dfnou = dehnendf() dfs = shudf(ro=ro, vo=vo) dfsnou = shudf() numpy.random.seed(1) du = (df.sampledSurfacemassLOS((11.0 * units.deg), n=1, maxd=(10.0 * units.kpc)).to(units.kpc).value / ro) numpy.random.seed(1) dnou = dfnou.sampledSurfacemassLOS(((11.0 * numpy.pi) / 180.0), n=1, maxd=(10.0 / ro)) assert (numpy.fabs((du - dnou)) < (10.0 ** (- 8.0))), 'diskdf sampling method sampledSurfacemassLOS does not return expected Quantity' numpy.random.seed(1) du = (df.sampleVRVT(1.1, n=1).to((units.km / units.s)).value / vo) numpy.random.seed(1) dnou = dfnou.sampleVRVT(1.1, n=1) assert numpy.all((numpy.fabs((du - dnou)) < (10.0 ** (- 8.0)))), 'diskdf sampling method sampleVRVT does not return expected Quantity' numpy.random.seed(1) du = df.sampleLOS((11.0 * units.deg), n=1) numpy.random.seed(1) dnou = dfnou.sampleLOS(11.0, n=1, deg=True) assert numpy.all((numpy.fabs((numpy.array(du[0].vxvv) - numpy.array(dnou[0].vxvv))) < (10.0 ** (- 8.0)))), 'diskdf sampling method sampleLOS does not work as expected with Quantity input' numpy.random.seed(1) du = df.sample(rrange=[(4.0 * units.kpc), (12.0 * units.kpc)], n=1) numpy.random.seed(1) dnou = dfnou.sample(rrange=[(4.0 / ro), (12.0 / ro)], n=1) assert numpy.all((numpy.fabs((numpy.array(du[0].vxvv) - numpy.array(dnou[0].vxvv))) < (10.0 ** (- 8.0)))), 'diskdf sampling method sample does not work as expected with Quantity input' numpy.random.seed(1) du = dfs.sample(rrange=[(4.0 * units.kpc), (12.0 * units.kpc)], n=1) numpy.random.seed(1) dnou = dfsnou.sample(rrange=[(4.0 / ro), (12.0 / ro)], n=1) assert numpy.all((numpy.fabs((numpy.array(du[0].vxvv) - numpy.array(dnou[0].vxvv))) < (10.0 ** (- 8.0)))), 'diskdf sampling method sample does not work as expected with Quantity input' return None
def l2_norm(input, axis=1): norm = torch.norm(input, 2, axis, True) output = torch.div(input, norm) return output
def face_area(V, F): if (type(V).__module__ == np.__name__): (V, F) = (p2e(V), p2e(F)) A = Xd() igl.doublearea(V, F, A) A = e2p(A).flatten() return A
class BasicBlock(nn.Module): expansion = 1 def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1, base_width=64, dilation=1, norm_layer=None): super(BasicBlock, self).__init__() if (norm_layer is None): norm_layer = nn.BatchNorm2d if ((groups != 1) or (base_width != 64)): raise ValueError('BasicBlock only supports groups=1 and base_width=64') self.conv1 = conv3x3(inplanes, planes, stride, dilation=dilation) self.bn1 = norm_layer(planes) self.relu = nn.ReLU(inplace=True) self.conv2 = conv3x3(planes, planes, dilation=dilation) self.bn2 = norm_layer(planes) self.downsample = downsample self.stride = stride def forward(self, x): identity = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) if (self.downsample is not None): identity = self.downsample(x) out += identity out = self.relu(out) return out
def main(): P = parse_args() P.rank = 0 if torch.cuda.is_available(): torch.cuda.set_device(P.rank) device = torch.device((f'cuda' if torch.cuda.is_available() else 'cpu')) P.world_size = torch.cuda.device_count() P.distributed = (P.world_size > 1) assert (not P.distributed) set_random_seed(P.seed) torch.backends.cudnn.deterministic = True torch.backends.cudnn.benchmark = False kwargs = {'batch_size': P.test_batch_size, 'shuffle': True, 'pin_memory': True, 'num_workers': 4} test_set = get_meta_dataset(P, dataset=P.dataset, only_test=True) if P.regression: test_loader = test_set else: test_loader = BatchMetaDataLoader(test_set, **kwargs) ' Initialize model ' model = get_model(P, P.model).to(device) load_model(P, model) from evals import setup as test_setup test_func = test_setup(P.mode, P) if (P.dataset == 'pose'): criterion = nn.MSELoss() elif (P.dataset == 'shapenet'): from data.shapenet1d import AzimuthLoss criterion = AzimuthLoss() else: criterion = nn.CrossEntropyLoss() ' test ' test_func(P, model, test_loader, criterion, 0.0, logger=None)
def create_cmfd_similarity_branch(img_shape=(256, 256, 3), nb_pools=100, name='simiDet'): img_input = Input(shape=img_shape, name=(name + '_in')) bname = (name + '_cnn') x1 = Conv2D(64, (3, 3), activation='relu', padding='same', name=(bname + '_b1c1'))(img_input) x1 = Conv2D(64, (3, 3), activation='relu', padding='same', name=(bname + '_b1c2'))(x1) x1 = MaxPooling2D((2, 2), strides=(2, 2), name=(bname + '_b1p'))(x1) x2 = Conv2D(128, (3, 3), activation='relu', padding='same', name=(bname + '_b2c1'))(x1) x2 = Conv2D(128, (3, 3), activation='relu', padding='same', name=(bname + '_b2c2'))(x2) x2 = MaxPooling2D((2, 2), strides=(2, 2), name=(bname + '_b2p'))(x2) x3 = Conv2D(256, (3, 3), activation='relu', padding='same', name=(bname + '_b3c1'))(x2) x3 = Conv2D(256, (3, 3), activation='relu', padding='same', name=(bname + '_b3c2'))(x3) x3 = Conv2D(256, (3, 3), activation='relu', padding='same', name=(bname + '_b3c3'))(x3) x3 = MaxPooling2D((2, 2), strides=(2, 2), name=(bname + '_b3p'))(x3) x4 = Conv2D(512, (3, 3), activation='relu', padding='same', name=(bname + '_b4c1'))(x3) x4 = Conv2D(512, (3, 3), activation='relu', padding='same', name=(bname + '_b4c2'))(x4) x4 = Conv2D(512, (3, 3), activation='relu', padding='same', name=(bname + '_b4c3'))(x4) x4 = MaxPooling2D((2, 2), strides=(2, 2), name=(bname + '_b4p'))(x4) xx = Activation(std_norm_along_chs, name=(bname + '_sn'))(x4) bname = (name + '_corr') xcorr = SelfCorrelationPercPooling(name=(bname + '_corr'))(xx) xn = BatchNormalization(name=(bname + '_bn'))(xcorr) patch_list = [(1, 1), (3, 3), (5, 5)] bname = (name + '_dconv') f16 = BnInception(xn, 8, patch_list, name=(bname + '_mpf')) f32 = BilinearUpSampling2D(name=(bname + '_bx2'))(f16) dx32 = BnInception(f32, 6, patch_list, name=(bname + '_dx2')) f64a = BilinearUpSampling2D(name=(bname + '_bx4a'))(f32) f64b = BilinearUpSampling2D(name=(bname + '_bx4b'))(dx32) f64 = Concatenate(axis=(- 1), name=(name + '_dx4_m'))([f64a, f64b]) dx64 = BnInception(f64, 4, patch_list, name=(bname + '_dx4')) f128a = BilinearUpSampling2D(name=(bname + '_bx8a'))(f64a) f128b = BilinearUpSampling2D(name=(bname + '_bx8b'))(dx64) f128 = Concatenate(axis=(- 1), name=(name + '_dx8_m'))([f128a, f128b]) dx128 = BnInception(f128, 2, patch_list, name=(bname + '_dx8')) f256a = BilinearUpSampling2D(name=(bname + '_bx16a'))(f128a) f256b = BilinearUpSampling2D(name=(bname + '_bx16b'))(dx128) f256 = Concatenate(axis=(- 1), name=(name + '_dx16_m'))([f256a, f256b]) dx256 = BnInception(f256, 2, patch_list, name=(bname + '_dx16')) fm256 = Concatenate(axis=(- 1), name=(name + '_mfeat'))([f256a, dx256]) masks = BnInception(fm256, 2, [(5, 5), (7, 7), (11, 11)], name=(bname + '_dxF')) pred_mask = Conv2D(1, (3, 3), activation='sigmoid', name=(name + '_pred_mask'), padding='same')(masks) model = Model(inputs=img_input, outputs=pred_mask, name=name) return model
class Logger(): def __init__(self, *args, **kwargs): from expviz.logger import Logger as ExpvizLogger self.expviz = ExpvizLogger(*args, **kwargs) def write(self, scalar_dict, epoch): for (key, value) in scalar_dict.items(): if isinstance(value, torch.Tensor): value = value.item() self.expviz.add_scalar(key, value, epoch)
def linkcode_resolve(domain, info): if (domain != 'py'): return None if (not info['module']): return None filename = info['module'].replace('.', '/') return '{}/{}.py'.format(repo_url, filename)
_model def res2net50_14w_8s(pretrained=False, **kwargs): model_args = dict(block=Bottle2neck, layers=[3, 4, 6, 3], base_width=14, block_args=dict(scale=8), **kwargs) return _create_res2net('res2net50_14w_8s', pretrained, **model_args)
class LRFinder(LearnerCallback): def __init__(self, learn: Learner, start_lr: float=1e-07, end_lr: float=10, num_it: int=100, stop_div: bool=True): super().__init__(learn) (self.data, self.stop_div) = (learn.data, stop_div) self.sched = Scheduler((start_lr, end_lr), num_it, annealing_exp) def on_train_begin(self, pbar, **kwargs: Any) -> None: setattr(pbar, 'clean_on_interrupt', True) self.learn.save('tmp') self.opt = self.learn.opt self.opt.lr = self.sched.start (self.stop, self.best_loss) = (False, 0.0) return {'skip_validate': True} def on_batch_end(self, iteration: int, smooth_loss: TensorOrNumber, **kwargs: Any) -> None: if ((iteration == 0) or (smooth_loss < self.best_loss)): self.best_loss = smooth_loss self.opt.lr = self.sched.step() if (self.sched.is_done or (self.stop_div and ((smooth_loss > (4 * self.best_loss)) or torch.isnan(smooth_loss)))): return {'stop_epoch': True, 'stop_training': True} def on_train_end(self, **kwargs: Any) -> None: self.learn.load('tmp', purge=False) if hasattr(self.learn.model, 'reset'): self.learn.model.reset() for cb in self.callbacks: if hasattr(cb, 'reset'): cb.reset() print('LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.')
class positive_int_or_none(_ParseType): _none def __call__(self, string: str) -> (int | None): return positive_int()(string)
def import_cifar(dataset=10): if (dataset == 10): ((x_train, y_train), (x_test, y_test)) = cifar10.load_data() elif (dataset == 100): ((x_train, y_train), (x_test, y_test)) = cifar100.load_data(label_mode='fine') x_train = x_train.astype('float32') x_test = x_test.astype('float32') (m, st) = (x_train.mean(), x_train.std()) x_train = (x_train - m) x_test = (x_test - m) x_train = (x_train / st) x_test = (x_test / st) y_train = one_hot_encoding(y_train) y_test = one_hot_encoding(y_test) return (x_train, y_train, x_test, y_test)
class GUITopoSim(Sim): def __init__(self, argv, parser=None): super(GUITopoSim, self).__init__(argv) self.topo_file = ((self.ROOT + '/') + self.args.topology_file) self.ns_file = ((self.ROOT + '/') + self.args.net_settings_file) self.net_desc = self.get_net_desc(self.topo_file, self.ns_file) def init_parser(self, parser): super(GUITopoSim, self).init_parser(parser) parser.add_argument('-t', '--topology_file', help='topology file generated by imalse GUI tool') parser.add_argument('-n', '--net_settings_file', help='net_settings file generated by imalse GUI tool') def get_net_desc(self, topo_file, ns_file): new_net_settings_file = fix_fs_addr_prefix_bug(ns_file) net_settings = load_para(f_name=new_net_settings_file, encap=None) net_settings['topo'] = get_inet_adj_mat(topo_file) net_settings['node_type'] = 'NNode' net_settings['node_para'] = {} return net_settings
class Data2VecTextForTokenClassification(metaclass=DummyObject): _backends = ['torch'] def __init__(self, *args, **kwargs): requires_backends(self, ['torch'])
def run(model, logdir, batch_size=50, vanilla=False, custom_steps=None, eta=None, n_samples=50000, nplog=None): if vanilla: print(f'Using Vanilla DDPM sampling with {model.num_timesteps} sampling steps.') else: print(f'Using DDIM sampling with {custom_steps} sampling steps and eta={eta}') tstart = time.time() n_saved = (len(glob.glob(os.path.join(logdir, '*.png'))) - 1) if (model.cond_stage_model is None): all_images = [] print(f'Running unconditional sampling for {n_samples} samples') for _ in trange((n_samples // batch_size), desc='Sampling Batches (unconditional)'): logs = make_convolutional_sample(model, batch_size=batch_size, vanilla=vanilla, custom_steps=custom_steps, eta=eta) n_saved = save_logs(logs, logdir, n_saved=n_saved, key='sample') all_images.extend([custom_to_np(logs['sample'])]) if (n_saved >= n_samples): print(f'Finish after generating {n_saved} samples') break all_img = np.concatenate(all_images, axis=0) all_img = all_img[:n_samples] shape_str = 'x'.join([str(x) for x in all_img.shape]) nppath = os.path.join(nplog, f'{shape_str}-samples.npz') np.savez(nppath, all_img) else: raise NotImplementedError('Currently only sampling for unconditional models supported.') print(f'sampling of {n_saved} images finished in {((time.time() - tstart) / 60.0):.2f} minutes.')
def deconv3d_bn_relu(batchNorm, in_planes, out_planes, kernel_size=4, stride=2, padding=1, output_padding=0, bias=True): if batchNorm: return nn.Sequential(nn.ConvTranspose3d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, output_padding=output_padding, bias=bias), nn.BatchNorm3d(out_planes), nn.ReLU(inplace=True)) else: return nn.Sequential(nn.ConvTranspose3d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, output_padding=output_padding, bias=bias), nn.ReLU(inplace=True))
def getPosFromFileName(fileName): return (int(fileName.split('/')[(- 1)].split('_')[(- 3)]) > 200)
class CalibCollector(CollectorBase): def __init__(self, include_tensors_kl, include_tensors_minmax, num_bins=8001): self.min_max_dict = {} self.hist_dict = {} self.num_bins = num_bins self.include_tensors_minmax = include_tensors_minmax self.include_tensors_kl = include_tensors_kl def collect_gluon(self, name, _, arr): if (name in self.include_tensors_kl): alg = 'kl' elif (name in self.include_tensors_minmax): alg = 'minmax' else: return min_range = arr.min().asscalar() max_range = arr.max().asscalar() th = max(abs(min_range), abs(max_range)) if (name in self.min_max_dict): cur_min_max = self.min_max_dict[name] self.min_max_dict[name] = (min(cur_min_max[0], min_range), max(cur_min_max[1], max_range)) else: self.min_max_dict[name] = (min_range, max_range) if (alg == 'kl'): arr = arr.asnumpy() if (name in self.hist_dict): self.hist_dict[name] = self._combine_histogram(self.hist_dict[name], arr, min_range, max_range, th) else: (hist, hist_edges) = np.histogram(arr, bins=self.num_bins, range=((- th), th)) self.hist_dict[name] = (hist, hist_edges, min_range, max_range, th) def _combine_histogram(old_hist, arr, new_min, new_max, new_th): if check_mx_version('2.0.0'): return mx.contrib.quantization._LayerHistogramCollector.combine_histogram(old_hist, arr, new_min, new_max, new_th) else: return mx.contrib.quantization.combine_histogram(old_hist, arr, new_min, new_max, new_th) def calc_kl_th_dict(self, quantized_dtype): if (len(self.hist_dict) > 0): if check_mx_version('2.0.0'): return mx.contrib.quantization._LayerHistogramCollector.get_optimal_thresholds(self.hist_dict, quantized_dtype) else: return mx.contrib.quantization._get_optimal_thresholds(self.hist_dict, quantized_dtype) return {}
class MLDG(ERM): def __init__(self, args): super(MLDG, self).__init__(args) self.args = args def update(self, minibatches, opt, sch): num_mb = len(minibatches) objective = 0 opt.zero_grad() for p in self.network.parameters(): if (p.grad is None): p.grad = torch.zeros_like(p) for ((xi, yi), (xj, yj)) in random_pairs_of_minibatches_by_domainperm(minibatches): (xi, yi, xj, yj) = (xi.cuda().float(), yi.cuda().long(), xj.cuda().float(), yj.cuda().long()) inner_net = copy.deepcopy(self.network) inner_opt = get_optimizer(inner_net, self.args, True) inner_sch = get_scheduler(inner_opt, self.args) inner_obj = F.cross_entropy(inner_net(xi), yi) inner_opt.zero_grad() inner_obj.backward() inner_opt.step() if inner_sch: inner_sch.step() for (p_tgt, p_src) in zip(self.network.parameters(), inner_net.parameters()): if (p_src.grad is not None): p_tgt.grad.data.add_((p_src.grad.data / num_mb)) objective += inner_obj.item() loss_inner_j = F.cross_entropy(inner_net(xj), yj) grad_inner_j = autograd.grad(loss_inner_j, inner_net.parameters(), allow_unused=True) objective += (self.args.mldg_beta * loss_inner_j).item() for (p, g_j) in zip(self.network.parameters(), grad_inner_j): if (g_j is not None): p.grad.data.add_(((self.args.mldg_beta * g_j.data) / num_mb)) objective /= len(minibatches) opt.step() if sch: sch.step() return {'total': objective}
def get_analytics_zoo_classpath(): if os.getenv('BIGDL_CLASSPATH'): for path in os.getenv('BIGDL_CLASSPATH').split(':'): if ((not os.path.exists(path)) and (not os.path.exists(path.split('*')[0]))): invalidInputError(False, 'Path {} specified BIGDL_CLASSPATH does not exist.'.format(path)) return os.environ['BIGDL_CLASSPATH'] jar_dir = os.path.abspath((__file__ + '/../../../')) jar_paths = glob.glob(os.path.join(jar_dir, 'share/orca/lib/*.jar')) if jar_paths: invalidInputError((len(jar_paths) == 1), ('Expecting one jar: %s' % len(jar_paths))) return jar_paths[0] return ''
def is_prime(n): if ((n % 2) == 0): return False sqrt_n = int(math.floor(math.sqrt(n))) for i in range(3, (sqrt_n + 1), 2): if ((n % i) == 0): return False return True
def run(cfg): seeds = ([cfg.seed] if (cfg.seed is not None) else range(cfg.runs)) if (cfg.gnn.model.name == 'RevGAT'): TRAINER = DGLGNNTrainer else: TRAINER = GNNTrainer all_acc = [] start = time.time() for seed in seeds: cfg.seed = seed trainer = TRAINER(cfg, cfg.gnn.train.feature_type) trainer.train() (_, acc) = trainer.eval_and_save() all_acc.append(acc) end = time.time() if (len(all_acc) > 1): df = pd.DataFrame(all_acc) print(f"[{cfg.gnn.model.name} + {cfg.gnn.train.feature_type}] ValACC: {df['val_acc'].mean():.4f} {df['val_acc'].std():.4f}, TestAcc: {df['test_acc'].mean():.4f} {df['test_acc'].std():.4f}") print(f'Running time: {((end - start) / len(seeds)):.2f}s')
def get_checkpoint_fn(): if deepspeed.checkpointing.is_configured(): checkpoint = deepspeed.checkpointing.checkpoint else: checkpoint = torch.utils.checkpoint.checkpoint return checkpoint
def frozen_bn(model): first_bn = True for (name, m) in model.named_modules(): if isinstance(m, (torch.nn.BatchNorm2d, torch.nn.BatchNorm3d)): if first_bn: first_bn = False print(('Skip frozen first bn layer: ' + name)) continue m.eval() m.weight.requires_grad = False m.bias.requires_grad = False
def cli_register(name: str, description: str=''): def _warpper(command): items = name.split('.') com = neuralchat_commands for item in items: com = com[item] com['_command'] = command if description: com['description'] = description return command return _warpper
def copy_bn_params(module, bn_module, remove_bn=True, verbose=False): with torch.no_grad(): if hasattr(bn_module, 'weight'): module.register_parameter('gamma', nn.Parameter(bn_module.weight.data.clone())) if hasattr(bn_module, 'bias'): module.register_parameter('beta', nn.Parameter(bn_module.bias.data.clone()))
.parametrize('space', [Discrete(3), Tuple([Discrete(5), Discrete(10)]), Tuple([Discrete(5), Box(low=np.array([0, 0]), high=np.array([1, 5]))]), Tuple((Discrete(5), Discrete(2), Discrete(2))), MultiDiscrete([2, 2, 100]), Dict({'position': Discrete(5), 'velocity': Box(low=np.array([0, 0]), high=np.array([1, 5]))})]) def test_roundtripping(space): sample_1 = space.sample() sample_2 = space.sample() assert space.contains(sample_1) assert space.contains(sample_2) json_rep = space.to_jsonable([sample_1, sample_2]) json_roundtripped = json.loads(json.dumps(json_rep)) samples_after_roundtrip = space.from_jsonable(json_roundtripped) (sample_1_prime, sample_2_prime) = samples_after_roundtrip s1 = space.to_jsonable([sample_1]) s1p = space.to_jsonable([sample_1_prime]) s2 = space.to_jsonable([sample_2]) s2p = space.to_jsonable([sample_2_prime]) assert (s1 == s1p), 'Expected {} to equal {}'.format(s1, s1p) assert (s2 == s2p), 'Expected {} to equal {}'.format(s2, s2p)
def trainfxn(trainer, model, dataloader, criterion, optimizer, lr_scheduler, epoch, args, num_classes, logger, **kwargs): batch_time = AverageMeter('Time', ':6.3f') data_time = AverageMeter('Data', ':6.3f') losses = AverageMeter('Loss', ':.4f') top1 = AverageMeter('', ':6.2f') if (num_classes >= 5): top5 = AverageMeter('', ':6.2f') else: top5 = AverageMeter('', ':6.2f') if (trainer in ['pgd', 'fgsm', 'trades']): top1_adv = AverageMeter('', ':6.2f') if (num_classes >= 5): top5_adv = AverageMeter('', ':6.2f') else: top5_adv = AverageMeter('', ':6.2f') progress = ProgressMeter(len(dataloader), [batch_time, data_time, losses, top1, top5, top1_adv, top5_adv], prefix='Epoch: [{}]'.format(epoch)) elif (trainer == 'baseline'): progress = ProgressMeter(len(dataloader), [batch_time, data_time, losses, top1, top5], prefix='Epoch: [{}]'.format(epoch)) else: raise ValueError(f'trainer {trainer} not supported') model.train() end = time.time() for (i, (images, targets)) in enumerate(dataloader): data_time.update((time.time() - end)) if ((i == 0) and (args.rank == 0)): logger.info((((f'Batch images shape: {images.shape}, targets shape: {targets.shape}, ' + f'World-size: {args.world_size}, Effective batch size: {(args.world_size * len(images))}, ') + f"Learning rate (epoch {epoch}/{args.epochs}): {optimizer.param_groups[0]['lr']:.5f}, ") + f'pixel range: {[images.min().item(), images.max().item()]}')) (images, targets) = (images.cuda(args.gpu, non_blocking=True), targets.cuda(args.gpu, non_blocking=True)) logits = model(images) (acc1, acc5) = accuracy(logits, targets, topk=(1, (5 if (num_classes >= 5) else 2))) top1.update(acc1[0], images.size(0)) top5.update(acc5[0], images.size(0)) if (trainer in ['fgsm', 'pgd', 'trades']): (logits_adv, loss) = get_adversarial_loss(trainer, model, images, targets, logits, criterion, optimizer, args) (acc1_adv, acc5_adv) = accuracy(logits_adv, targets, topk=(1, (5 if (num_classes >= 5) else 2))) top1_adv.update(acc1_adv[0], images.size(0)) top5_adv.update(acc5_adv[0], images.size(0)) elif (trainer in ['baseline']): loss = criterion(logits, targets) losses.update(loss.item(), images.size(0)) optimizer.zero_grad() loss.backward() optimizer.step() lr_scheduler.step() batch_time.update((time.time() - end)) end = time.time() if (((i % args.print_freq) == 0) and (args.rank == 0)): progress.display(i) if (trainer in ['fgsm', 'pgd', 'trades']): result = {'top1': top1.avg, f'top{(5 if (num_classes >= 5) else 2)}': top5.avg, 'top1_adv': top1_adv.avg, f'top{(5 if (num_classes >= 5) else 2)}_adv': top5_adv.avg} elif (trainer in ['baseline']): result = {'top1': top1.avg, f'top{(5 if (num_classes >= 5) else 2)}': top5.avg} return result
class JTMPN(nn.Module): def __init__(self, hidden_size, depth): super(JTMPN, self).__init__() self.hidden_size = hidden_size self.depth = depth self.W_i = nn.Linear((ATOM_FDIM + BOND_FDIM), hidden_size, bias=False) self.W_h = nn.Linear(hidden_size, hidden_size, bias=False) self.W_o = nn.Linear((ATOM_FDIM + hidden_size), hidden_size) def forward(self, cand_batch, tree_mess): (fatoms, fbonds) = ([], []) (in_bonds, all_bonds) = ([], []) (mess_dict, all_mess) = ({}, [create_var(torch.zeros(self.hidden_size))]) total_atoms = 0 scope = [] for (e, vec) in tree_mess.iteritems(): mess_dict[e] = len(all_mess) all_mess.append(vec) for (mol, all_nodes, ctr_node) in cand_batch: n_atoms = mol.GetNumAtoms() ctr_bid = ctr_node.idx for atom in mol.GetAtoms(): fatoms.append(atom_features(atom)) in_bonds.append([]) for bond in mol.GetBonds(): a1 = bond.GetBeginAtom() a2 = bond.GetEndAtom() x = (a1.GetIdx() + total_atoms) y = (a2.GetIdx() + total_atoms) (x_nid, y_nid) = (a1.GetAtomMapNum(), a2.GetAtomMapNum()) x_bid = (all_nodes[(x_nid - 1)].idx if (x_nid > 0) else (- 1)) y_bid = (all_nodes[(y_nid - 1)].idx if (y_nid > 0) else (- 1)) bfeature = bond_features(bond) b = (len(all_mess) + len(all_bonds)) all_bonds.append((x, y)) fbonds.append(torch.cat([fatoms[x], bfeature], 0)) in_bonds[y].append(b) b = (len(all_mess) + len(all_bonds)) all_bonds.append((y, x)) fbonds.append(torch.cat([fatoms[y], bfeature], 0)) in_bonds[x].append(b) if ((x_bid >= 0) and (y_bid >= 0) and (x_bid != y_bid)): if ((x_bid, y_bid) in mess_dict): mess_idx = mess_dict[(x_bid, y_bid)] in_bonds[y].append(mess_idx) if ((y_bid, x_bid) in mess_dict): mess_idx = mess_dict[(y_bid, x_bid)] in_bonds[x].append(mess_idx) scope.append((total_atoms, n_atoms)) total_atoms += n_atoms total_bonds = len(all_bonds) total_mess = len(all_mess) fatoms = torch.stack(fatoms, 0) fbonds = torch.stack(fbonds, 0) agraph = torch.zeros(total_atoms, MAX_NB).long() bgraph = torch.zeros(total_bonds, MAX_NB).long() tree_message = torch.stack(all_mess, dim=0) for a in xrange(total_atoms): for (i, b) in enumerate(in_bonds[a]): agraph[(a, i)] = b for b1 in xrange(total_bonds): (x, y) = all_bonds[b1] for (i, b2) in enumerate(in_bonds[x]): if ((b2 < total_mess) or (all_bonds[(b2 - total_mess)][0] != y)): bgraph[(b1, i)] = b2 fatoms = create_var(fatoms) fbonds = create_var(fbonds) agraph = create_var(agraph) bgraph = create_var(bgraph) binput = self.W_i(fbonds) graph_message = nn.ReLU()(binput) for i in xrange((self.depth - 1)): message = torch.cat([tree_message, graph_message], dim=0) nei_message = index_select_ND(message, 0, bgraph) nei_message = nei_message.sum(dim=1) nei_message = self.W_h(nei_message) graph_message = nn.ReLU()((binput + nei_message)) message = torch.cat([tree_message, graph_message], dim=0) nei_message = index_select_ND(message, 0, agraph) nei_message = nei_message.sum(dim=1) ainput = torch.cat([fatoms, nei_message], dim=1) atom_hiddens = nn.ReLU()(self.W_o(ainput)) mol_vecs = [] for (st, le) in scope: mol_vec = (atom_hiddens.narrow(0, st, le).sum(dim=0) / le) mol_vecs.append(mol_vec) mol_vecs = torch.stack(mol_vecs, dim=0) return mol_vecs
class ResBlock(nn.Module): def __init__(self, in_channels, out_channels, stride=1): super(ResBlock, self).__init__() self.resize_identity = ((in_channels != out_channels) or (stride != 1)) self.body = PreResBottleneck(in_channels=in_channels, out_channels=out_channels, stride=stride) if self.resize_identity: self.identity_conv = conv1x1(in_channels=in_channels, out_channels=out_channels, stride=stride) def forward(self, x): identity = x (x, x_pre_activ) = self.body(x) if self.resize_identity: identity = self.identity_conv(x_pre_activ) x = (x + identity) return x
def PrintModelFree(mfIndi, mbIndi): for i in xrange(len(mfIndi)): if (not np.isnan(mfIndi[i])): print(('[%d]\t%f' % (i, mfIndi[i]))) print('\n')
def init(novel_type, description, request: gr.Request): if (novel_type == ''): novel_type = ('Science Fiction' if ('en' == lang_opt) else '') global _CACHE cookie = request.headers['cookie'] cookie = cookie.split('; _gat_gtag')[0] init_paragraphs = get_init(text=init_prompt(novel_type, description), model=llm_model, tokenizer=llm_tokenizer) start_input_to_human = {'output_paragraph': init_paragraphs['Paragraph 3'], 'input_paragraph': '\n\n'.join([init_paragraphs['Paragraph 1'], init_paragraphs['Paragraph 2'], init_paragraphs['Paragraph 3']]), 'output_memory': init_paragraphs['Summary'], 'output_instruction': [init_paragraphs['Instruction 1'], init_paragraphs['Instruction 2'], init_paragraphs['Instruction 3']]} _CACHE[cookie] = {'start_input_to_human': start_input_to_human, 'init_paragraphs': init_paragraphs} written_paras = (f'''Title: {init_paragraphs['name']} Outline: {init_paragraphs['Outline']} Paragraphs: {start_input_to_human['input_paragraph']}''' if ('en' == lang_opt) else f''': {init_paragraphs['name']} : {init_paragraphs['Outline']} : {start_input_to_human['input_paragraph']}''') long_memory = parse_instructions([init_paragraphs['Paragraph 1'], init_paragraphs['Paragraph 2'], init_paragraphs['Paragraph 3']]) return (start_input_to_human['output_memory'], long_memory, written_paras, init_paragraphs['Instruction 1'], init_paragraphs['Instruction 2'], init_paragraphs['Instruction 3'])
_module class CityscapesDataset(CocoDataset): CLASSES = ('person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle') def _filter_imgs(self, min_size=32): valid_inds = [] ids_with_ann = set((_['image_id'] for _ in self.coco.anns.values())) for (i, img_info) in enumerate(self.img_infos): img_id = img_info['id'] ann_ids = self.coco.getAnnIds(imgIds=[img_id]) ann_info = self.coco.loadAnns(ann_ids) all_iscrowd = all([_['iscrowd'] for _ in ann_info]) if (self.filter_empty_gt and ((self.img_ids[i] not in ids_with_ann) or all_iscrowd)): continue if (min(img_info['width'], img_info['height']) >= min_size): valid_inds.append(i) return valid_inds def _parse_ann_info(self, img_info, ann_info): gt_bboxes = [] gt_labels = [] gt_bboxes_ignore = [] gt_masks_ann = [] for (i, ann) in enumerate(ann_info): if ann.get('ignore', False): continue (x1, y1, w, h) = ann['bbox'] if ((ann['area'] <= 0) or (w < 1) or (h < 1)): continue bbox = [x1, y1, ((x1 + w) - 1), ((y1 + h) - 1)] if ann.get('iscrowd', False): gt_bboxes_ignore.append(bbox) else: gt_bboxes.append(bbox) gt_labels.append(self.cat2label[ann['category_id']]) gt_masks_ann.append(ann['segmentation']) if gt_bboxes: gt_bboxes = np.array(gt_bboxes, dtype=np.float32) gt_labels = np.array(gt_labels, dtype=np.int64) else: gt_bboxes = np.zeros((0, 4), dtype=np.float32) gt_labels = np.array([], dtype=np.int64) if gt_bboxes_ignore: gt_bboxes_ignore = np.array(gt_bboxes_ignore, dtype=np.float32) else: gt_bboxes_ignore = np.zeros((0, 4), dtype=np.float32) ann = dict(bboxes=gt_bboxes, labels=gt_labels, bboxes_ignore=gt_bboxes_ignore, masks=gt_masks_ann, seg_map=img_info['segm_file']) return ann def results2txt(self, results, outfile_prefix): try: import cityscapesscripts.helpers.labels as CSLabels except ImportError: raise ImportError('Please run "pip install citscapesscripts" to install cityscapesscripts first.') result_files = [] os.makedirs(outfile_prefix, exist_ok=True) prog_bar = mmcv.ProgressBar(len(self)) for idx in range(len(self)): result = results[idx] filename = self.img_infos[idx]['filename'] basename = osp.splitext(osp.basename(filename))[0] pred_txt = osp.join(outfile_prefix, (basename + '_pred.txt')) (bbox_result, segm_result) = result bboxes = np.vstack(bbox_result) segms = mmcv.concat_list(segm_result) labels = [np.full(bbox.shape[0], i, dtype=np.int32) for (i, bbox) in enumerate(bbox_result)] labels = np.concatenate(labels) assert (len(bboxes) == len(segms) == len(labels)) num_instances = len(bboxes) prog_bar.update() with open(pred_txt, 'w') as fout: for i in range(num_instances): pred_class = labels[i] classes = self.CLASSES[pred_class] class_id = CSLabels.name2label[classes].id score = bboxes[(i, (- 1))] mask = maskUtils.decode(segms[i]).astype(np.uint8) png_filename = osp.join(outfile_prefix, (basename + '_{}_{}.png'.format(i, classes))) mmcv.imwrite(mask, png_filename) fout.write('{} {} {}\n'.format(osp.basename(png_filename), class_id, score)) result_files.append(pred_txt) return result_files def format_results(self, results, txtfile_prefix=None): assert isinstance(results, list), 'results must be a list' assert (len(results) == len(self)), 'The length of results is not equal to the dataset len: {} != {}'.format(len(results), len(self)) assert isinstance(results, list), 'results must be a list' assert (len(results) == len(self)), 'The length of results is not equal to the dataset len: {} != {}'.format(len(results), len(self)) if (txtfile_prefix is None): tmp_dir = tempfile.TemporaryDirectory() txtfile_prefix = osp.join(tmp_dir.name, 'results') else: tmp_dir = None result_files = self.results2txt(results, txtfile_prefix) return (result_files, tmp_dir) def evaluate(self, results, metric='bbox', logger=None, outfile_prefix=None, classwise=False, proposal_nums=(100, 300, 1000), iou_thrs=np.arange(0.5, 0.96, 0.05)): eval_results = dict() metrics = (metric.copy() if isinstance(metric, list) else [metric]) if ('cityscapes' in metrics): eval_results.update(self._evaluate_cityscapes(results, outfile_prefix, logger)) metrics.remove('cityscapes') if (len(metrics) > 0): self_coco = CocoDataset(self.ann_file, self.pipeline.transforms, self.data_root, self.img_prefix, self.seg_prefix, self.proposal_file, self.test_mode, self.filter_empty_gt) eval_results.update(self_coco.evaluate(results, metrics, logger, outfile_prefix, classwise, proposal_nums, iou_thrs)) return eval_results def _evaluate_cityscapes(self, results, txtfile_prefix, logger): try: import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling as CSEval except ImportError: raise ImportError('Please run "pip install citscapesscripts" to install cityscapesscripts first.') msg = 'Evaluating in Cityscapes style' if (logger is None): msg = ('\n' + msg) print_log(msg, logger=logger) (result_files, tmp_dir) = self.format_results(results, txtfile_prefix) if (tmp_dir is None): result_dir = osp.join(txtfile_prefix, 'results') else: result_dir = osp.join(tmp_dir.name, 'results') eval_results = {} print_log('Evaluating results under {} ...'.format(result_dir), logger=logger) CSEval.args.cityscapesPath = os.path.join(self.img_prefix, '../..') CSEval.args.predictionPath = os.path.abspath(result_dir) CSEval.args.predictionWalk = None CSEval.args.JSONOutput = False CSEval.args.colorized = False CSEval.args.gtInstancesFile = os.path.join(result_dir, 'gtInstances.json') CSEval.args.groundTruthSearch = os.path.join(self.img_prefix.replace('leftImg8bit', 'gtFine'), '*/*_gtFine_instanceIds.png') groundTruthImgList = glob.glob(CSEval.args.groundTruthSearch) assert len(groundTruthImgList), 'Cannot find ground truth images in {}.'.format(CSEval.args.groundTruthSearch) predictionImgList = [] for gt in groundTruthImgList: predictionImgList.append(CSEval.getPrediction(gt, CSEval.args)) CSEval_results = CSEval.evaluateImgLists(predictionImgList, groundTruthImgList, CSEval.args)['averages'] eval_results['mAP'] = CSEval_results['allAp'] eval_results[''] = CSEval_results['allAp50%'] if (tmp_dir is not None): tmp_dir.cleanup() return eval_results
def main(): parser = argparse.ArgumentParser(description='ReID Baseline Inference') parser.add_argument('--config_file', default='./configs/debug.yml', help='path to config file', type=str) parser.add_argument('opts', help='Modify config options using the command-line', default=None, nargs=argparse.REMAINDER) args = parser.parse_args() num_gpus = (int(os.environ['WORLD_SIZE']) if ('WORLD_SIZE' in os.environ) else 1) if (args.config_file != ''): cfg.merge_from_file(args.config_file) cfg.merge_from_list(args.opts) cfg.freeze() output_dir = cfg.OUTPUT_DIR if (output_dir and (not os.path.exists(output_dir))): mkdir(output_dir) logger = setup_logger('reid_baseline', output_dir, 0) logger.info('Using {} GPUS'.format(num_gpus)) logger.info(args) if (args.config_file != ''): logger.info('Loaded configuration file {}'.format(args.config_file)) logger.info('Running with config:\n{}'.format(cfg)) if (cfg.MODEL.DEVICE == 'cuda'): os.environ['CUDA_VISIBLE_DEVICES'] = cfg.MODEL.DEVICE_ID cudnn.benchmark = True (train_loader, val_loader, num_query, num_classes, dataset) = make_data_loader(cfg) model = build_model(cfg, num_classes) model.load_param(cfg.TEST.WEIGHT) inference(cfg, model, val_loader, num_query, dataset)
class ParticleNetWrapper(torch.nn.Module): def __init__(self, **kwargs) -> None: super().__init__() self.mod = ParticleNet(**kwargs) def forward(self, points, features, lorentz_vectors, mask): return self.mod(points, features, mask)
class MetadataCache(Cache): def source(cls, url): print('Getting metadata from source') browser = Browser(os.path.join(os.path.abspath(os.path.dirname(__file__)), 'random_downloads')) browser.get(url) print('URL gotten') metadata = cls.metadata_from_result_page(browser) MHTMLCache.add_from_browser(url, browser) browser.close() return metadata
class AddSubNode(ExprNode): def __init__(self, left=None, right=None, parse_info=None, raw_text=None, op='+-'): super().__init__(IRNodeType.AddSub, parse_info=parse_info, raw_text=raw_text) self.left = left self.right = right self.op = op def split_node(self): add_node = AddNode(self.left, self.right) add_node.set_parent(self.parent()) sub_node = SubNode(self.left, self.right) sub_node.set_parent(self.parent()) return [add_node, sub_node] def get_child(self, node_type): if self.left.is_node(node_type): child_node = self.left elif self.right.is_node(node_type): child_node = self.right else: child_node = self.left.get_child(node_type) if (child_node is None): child_node = self.right.get_child(node_type) return child_node
def certificate(domain=LOCALHOST, country=None, state=None, city=None, company=None, contact=None, signed=True, **kwargs): s = subprocess.PIPE p = ('openssl', 'genrsa', ('%s' % kwargs.get('encryption', 2048))) p = subprocess.Popen(p, stdin=s, stdout=s, stderr=s) k = (kwargs.get('key') or p.communicate()[0]) f = tempfile.NamedTemporaryFile(delete=False) f.write(k) f.close() p = ('openssl', 'req', '-new', '-key', f.name) p = ((p + ('-x509', '-days', '365')) if signed else p) p = subprocess.Popen(p, stdin=s, stdout=s, stderr=s) x = p.communicate(('%s\n%s\n%s\n%s\n.\n%s\n%s\n\n\n' % ((country or '.'), (state or '.'), (city or '.'), (company or '.'), (domain or LOCALHOST), (contact or '.'))))[0] os.unlink(f.name) return (k, x)
def dice(input, target, ignore_index=None): smooth = 1.0 iflat = input.clone().view((- 1)) tflat = target.clone().view((- 1)) if (ignore_index is not None): mask = (tflat == ignore_index) tflat[mask] = 0 iflat[mask] = 0 intersection = (iflat * tflat).sum() return (((2.0 * intersection) + smooth) / ((iflat.sum() + tflat.sum()) + smooth))
class SubsetSum(BinaryProblem): def __init__(self, C: int, W: list): super(SubsetSum, self).__init__() self.C = C self.W = W self.number_of_bits = len(self.W) self.number_of_objectives = 2 self.number_of_variables = 1 self.number_of_constraints = 0 self.obj_directions = [self.MAXIMIZE, self.MINIMIZE] self.obj_labels = ['Sum', 'No. of Objects'] def evaluate(self, solution: BinarySolution) -> BinarySolution: total_sum = 0.0 number_of_objects = 0 for (index, bits) in enumerate(solution.variables[0]): if bits: total_sum += self.W[index] number_of_objects += 1 if (total_sum > self.C): total_sum = (self.C - (total_sum * 0.1)) if (total_sum < 0.0): total_sum = 0.0 solution.objectives[0] = ((- 1.0) * total_sum) solution.objectives[1] = number_of_objects return solution def create_solution(self) -> BinarySolution: new_solution = BinarySolution(number_of_variables=self.number_of_variables, number_of_objectives=self.number_of_objectives) new_solution.variables[0] = [(True if (random.randint(0, 1) == 0) else False) for _ in range(self.number_of_bits)] return new_solution def name(self) -> str: return 'Subset Sum'
def make_image(tensor): return tensor.detach().clamp_(min=(- 1), max=1).add(1).div_(2).mul(255).type(torch.uint8).permute(0, 2, 3, 1).to('cpu').numpy()
def main(argv): epochs = FLAGS.epochs batch_size = FLAGS.batch_size gru_units = (FLAGS.model_size * FLAGS.model_size_scale) emb_dim = FLAGS.emb_dim max_seq = FLAGS.max_seq patience = FLAGS.patience l2 = FLAGS.l2 lr = FLAGS.lr with open((FLAGS.input + 'train.pkl'), 'rb') as f: (train, train_y) = pickle.load(f) with open((FLAGS.input + 'val.pkl'), 'rb') as f: (val, val_y) = pickle.load(f) with open((FLAGS.input + 'test.pkl'), 'rb') as f: (test, test_y) = pickle.load(f) train = divide_dataset(train, label_fractions=FLAGS.label_fraction) train_y = divide_dataset(train_y, label_fractions=FLAGS.label_fraction) dataset_info = (('# training samples {}\n' + '# validation samples {}\n') + '# test samples {}') print(dataset_info.format(len(train), len(val), len(test))) inputs = tf.keras.layers.Input(shape=(max_seq, emb_dim)) gru_1 = tf.keras.layers.Bidirectional(tf.keras.layers.GRU(gru_units, return_sequences=True, kernel_regularizer=tf.keras.regularizers.l2(l2)))(inputs) gru_2 = tf.keras.layers.Bidirectional(tf.keras.layers.GRU(gru_units, kernel_regularizer=tf.keras.regularizers.l2(l2)))(gru_1) mlp_1 = tf.keras.layers.Dense(gru_units, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(l2))(gru_2) mlp_2 = tf.keras.layers.Dense((gru_units // 2), activation='relu', kernel_regularizer=tf.keras.regularizers.l2(l2))(mlp_1) outputs = tf.keras.layers.Dense(1)(mlp_2) base_model = tf.keras.models.Model(inputs, outputs) base_model.summary() optimizer = tf.keras.optimizers.Adam(lr) loss_object = tf.keras.losses.MeanSquaredLogarithmicError() train_loss = tf.keras.metrics.Mean(name='train_loss') train_msle = tf.keras.metrics.MeanSquaredLogarithmicError(name='train_msle') val_loss = tf.keras.metrics.Mean(name='val_loss') val_msle = tf.keras.metrics.MeanSquaredLogarithmicError(name='train_msle') test_loss = tf.keras.metrics.Mean(name='test_loss') test_msle = tf.keras.metrics.MeanSquaredLogarithmicError(name='test_msle') def train_step(data, labels): with tf.GradientTape() as tape: predictions = base_model(data, training=True) loss = loss_object(labels, predictions) gradients = tape.gradient(loss, base_model.trainable_variables) optimizer.apply_gradients(zip(gradients, base_model.trainable_variables)) train_loss(loss) train_msle(labels, predictions) def val_step(data, labels): predictions = base_model(data, training=False) v_loss = loss_object(labels, predictions) val_loss(v_loss) val_msle(labels, predictions) def test_step(data, labels): predictions = base_model(data, training=False) t_loss = loss_object(labels, predictions) test_loss(t_loss) test_msle(labels, predictions) return predictions best_val_msle = 1000 save_predictions = list() for epoch in range(epochs): (train, train_y) = shuffle_two(train, train_y) time_start = time.time() train_loss.reset_states() train_msle.reset_states() val_loss.reset_states() val_msle.reset_states() test_loss.reset_states() test_msle.reset_states() for i in range(((len(train) // batch_size) + 1)): batch_train = copy.deepcopy(train[(batch_size * i):((batch_size * i) + batch_size)]) batch_train_labels = train_y[(batch_size * i):((batch_size * i) + batch_size)] for batch_cascade in batch_train: while (len(batch_cascade) < max_seq): batch_cascade.append(np.zeros(emb_dim)) train_step(np.array(batch_train), np.array(batch_train_labels)) for i in range(((len(val) // batch_size) + 1)): batch_val = copy.deepcopy(val[(batch_size * i):((batch_size * i) + batch_size)]) batch_val_labels = val_y[(batch_size * i):((batch_size * i) + batch_size)] for batch_cascade in batch_val: while (len(batch_cascade) < max_seq): batch_cascade.append(np.zeros(emb_dim)) val_step(np.array(batch_val), np.array(batch_val_labels)) pred = list() for i in range(((len(test) // batch_size) + 1)): batch_test = copy.deepcopy(test[(batch_size * i):((batch_size * i) + batch_size)]) batch_test_labels = test_y[(batch_size * i):((batch_size * i) + batch_size)] for batch_cascade in batch_test: while (len(batch_cascade) < max_seq): batch_cascade.append(np.zeros(emb_dim)) batch_predictions = test_step(np.array(batch_test), np.array(batch_test_labels)) pred.extend(batch_predictions) pred = [float(pre) for pre in pred] report_loss = np.mean(np.square((np.log2(np.array([(pre if (pre >= 1) else 1) for pre in pred])) - np.log2(np.array([(tru if (tru >= 1) else 1) for tru in list(test_y)]))))) if (val_msle.result() < best_val_msle): best_val_msle = val_msle.result() save_predictions = pred patience = FLAGS.patience template = 'Epoch {:2}, Time: {:.3f}s, Train Loss: {:.3f}, Train MSLE: {:.3f}, Val Loss: {:.3f}, Val MSLE: {:.3f}, Test Loss: {:.3f}, Test MSLE: {:.3f}, LOG2 MSLE: {:.3f}' print(template.format((epoch + 1), (time.time() - time_start), train_loss.result(), train_msle.result(), val_loss.result(), val_msle.result(), test_loss.result(), test_msle.result(), report_loss)) if (patience == 0): report_loss = np.mean(np.square((np.log2(np.array([(pre if (pre >= 1) else 1) for pre in save_predictions])) - np.log2(np.array([(tru if (tru >= 1) else 1) for tru in list(test_y)]))))) print('Predictions saved! Best Test MSLE: {}'.format(report_loss)) break else: patience -= 1 print('Finished! Time used: {:.3f}min'.format(((time.time() - start_time) / 60)))
def get_datamodule(datamodule): datamodule = datamodule.lower() if (datamodule == 'cifar10'): return Cifar10DataModule if (datamodule == 'cifar100'): return Cifar100DataModule elif (datamodule == 'mnist'): return MnistDataModule elif (datamodule == 'imagenet'): return ImagenetDataModule elif (datamodule == 'stl10'): return STL10DataModule elif (datamodule == 'stl10_unlabeled'): return STL10UnlabeledDataModule elif (datamodule == 'coco'): return CocoClipDataModule elif (datamodule == 'food101'): return Food101DataModule elif (datamodule == 'cars196'): return Cars196DataModule elif (datamodule == 'pets37'): return Pets37DataModule elif (datamodule == 'caltech101'): return Caltech101DataModule elif (datamodule == 'pcam'): return PCamDataModule elif (datamodule == 'galaxy'): return GalaxyDataModule elif (datamodule == 'banana'): return BananaDataModule else: raise ValueError(f'Unkown datamodule: {datamodule}')
class ReadWork(): def __init__(self, read_done_condition, key, read_done_flag, cache_lock, cache): self.read_done_condition = read_done_condition self.key = key self.read_done_flag = read_done_flag self.cache_lock = cache_lock self.cache = cache def wait(self): with self.read_done_condition: while (not self.read_done_flag[self.key]): self.read_done_condition.wait() with self.cache_lock: (tensor, ref_count) = self.cache[self.key] ref_count -= 1 if (ref_count == 0): del self.cache[self.key] else: self.cache[self.key] = (tensor, ref_count)
def main(): app = QtGui.QApplication(sys.argv) tool = CityscapesViewer() sys.exit(app.exec_())
class MLPCategoricalActor(Actor): def __init__(self, obs_dim, act_dim, hidden_sizes, activation): super().__init__() self.logits_net = mlp((([obs_dim] + list(hidden_sizes)) + [act_dim]), activation) def _distribution(self, obs): logits = self.logits_net(obs) return Categorical(logits=logits) def _log_prob_from_distribution(self, pi, act): return pi.log_prob(act) def _d_kl(self, obs, old_mu, old_log_std, device): raise NotImplementedError
_operation def div(a: torch.Tensor, b: torch.Tensor): if is_real(b): if (b.dim() >= a.dim()): raise ValueError('Incorrect dimensions.') return div_cplx_real(a, b) return div_cplx_real(mult_conj(a, b), abs_sqr(b))
def test_bytes(doc): assert (m.bytes_from_string().decode() == 'foo') assert (m.bytes_from_str().decode() == 'bar') assert (doc(m.bytes_from_str) == 'bytes_from_str() -> {}'.format(('bytes' if (sys.version_info[0] == 3) else 'str')))
class AlbertPreTrainedModel(): def __init__(self, *args, **kwargs): requires_pytorch(self) def from_pretrained(self, *args, **kwargs): requires_pytorch(self)
def get_coco_api_from_dataset(dataset): for _ in range(10): if isinstance(dataset, torch.utils.data.Subset): dataset = dataset.dataset if isinstance(dataset, LvisDetectionBase): return dataset.lvis if isinstance(dataset, (torchvision.datasets.CocoDetection, CustomCocoDetection)): return dataset.coco
def features_dataset_resizer(features_dataset_class: Type[BaseFeaturesDataset], resize_factor: float): old_get = features_dataset_class.get_features def resizer_get(*args): features = old_get(*args) new_features = copy.deepcopy(features) new_features['skeletons'] *= resize_factor new_features['head_width'] *= resize_factor new_features['midbody_width'] *= resize_factor new_features['tail_width'] *= resize_factor return new_features features_dataset_class.get_features = resizer_get return features_dataset_class
class MobileNetV2(nn.Module): def __init__(self, channels, init_block_channels, final_block_channels, in_channels=3, in_size=(224, 224), num_classes=1000): super(MobileNetV2, self).__init__() self.in_size = in_size self.num_classes = num_classes self.features = nn.Sequential() self.features.add_module('init_block', conv3x3_block(in_channels=in_channels, out_channels=init_block_channels, stride=2, activation='relu6')) in_channels = init_block_channels for (i, channels_per_stage) in enumerate(channels): stage = nn.Sequential() for (j, out_channels) in enumerate(channels_per_stage): stride = (2 if ((j == 0) and (i != 0)) else 1) expansion = ((i != 0) or (j != 0)) stage.add_module('unit{}'.format((j + 1)), LinearBottleneck(in_channels=in_channels, out_channels=out_channels, stride=stride, expansion=expansion)) in_channels = out_channels self.features.add_module('stage{}'.format((i + 1)), stage) self.features.add_module('final_block', conv1x1_block(in_channels=in_channels, out_channels=final_block_channels, activation='relu6')) in_channels = final_block_channels self.features.add_module('final_pool', nn.AvgPool2d(kernel_size=7, stride=1)) self.output = conv1x1(in_channels=in_channels, out_channels=num_classes, bias=False) self._init_params() def _init_params(self): for (name, module) in self.named_modules(): if isinstance(module, nn.Conv2d): init.kaiming_uniform_(module.weight) if (module.bias is not None): init.constant_(module.bias, 0) def forward(self, x): x = self.features(x) x = self.output(x) x = x.view(x.size(0), (- 1)) return x
def penalties(module, reduction='sum'): for (_, penalty) in named_penalties(module, reduction=reduction): (yield penalty)
def load_model_base(self, model_path: str, from_pretrained_kwargs: dict): revision = from_pretrained_kwargs.get('revision', 'main') print('Customized bigdl-llm loader') tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=self.use_fast_tokenizer, revision=revision) from bigdl.llm.transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained(model_path, low_cpu_mem_usage=True, **from_pretrained_kwargs) return (model, tokenizer)
class p_z(nn.Module): def __init__(self, output_shape, input_shape): super().__init__() (nc_y_in, nc_u_in) = (input_shape[0][0], input_shape[1][0]) nc_out = (2 * output_shape[0]) self.y_nn = nn.Sequential(DenselyEncoder(in_channels=nc_y_in, out_channels=(nc_out // 2), growth_rate=32, steps=5, scale_factor=1), nn.ELU(inplace=True)) self.u_nn = nn.Sequential(DenselyNetwork(in_channels=nc_u_in, out_channels=(nc_out // 2), growth_rate=64, steps=3, blocks=3, act=True)) self.core_nn = nn.Sequential(DenselyNetwork(in_channels=nc_out, out_channels=nc_out, growth_rate=64, steps=3, blocks=3, act=None)) def forward(self, input): (y, u) = (input[0], input[1]) y_out = self.y_nn(y) u_out = self.u_nn(u) joint = torch.cat((y_out, u_out), 1) (mu, logvar) = self.core_nn(joint).chunk(2, 1) return (mu, F.hardtanh(logvar, min_val=(- 7), max_val=7.0))
def run_daemon(bpf): signal.signal(signal.SIGTERM, remove_rt) signal.signal(signal.SIGINT, remove_rt) for (laddr, gaddr) in LOCAL_GLOBAL_MAP.items(): _ = (lambda x: socket.inet_pton(socket.AF_INET6, x)) logger.info('{}:{}'.format(laddr, gaddr)) bpf['link_local_table'][ip_str_to_ct(laddr)] = ip_str_to_ct(gaddr) while 1: bpf.kprobe_poll() sleep(0.01)
class GumbelBatchedGenerator(): def __init__(self, seed=None): if isinstance(seed, random.Random): self.rng = seed else: self.rng = random.Random(seed) def __call__(self): return (- math.log((- math.log(self.rng.random()))))
def import_models(models_dir, namespace): for file in os.listdir(models_dir): path = os.path.join(models_dir, file) if ((not file.startswith('_')) and (not file.startswith('.')) and (file.endswith('.py') or os.path.isdir(path))): model_name = (file[:file.find('.py')] if file.endswith('.py') else file) importlib.import_module(((namespace + '.') + model_name)) if (model_name in MODEL_REGISTRY): parser = argparse.ArgumentParser(add_help=False) group_archs = parser.add_argument_group('Named architectures') group_archs.add_argument('--arch', choices=ARCH_MODEL_INV_REGISTRY[model_name]) group_args = parser.add_argument_group('Additional command-line arguments') MODEL_REGISTRY[model_name].add_args(group_args) globals()[(model_name + '_parser')] = parser
def get_valid_mask(mask: np.ndarray): if (mask.ndim == 3): mask_pil = Image.fromarray(mask).convert('L') mask = np.array(mask_pil) if (mask.max() == 255): mask = (mask / 255) return mask
def test_squeeze_and_excitation_block_2d(): N = 10 C = 128 reduction = 16 data = torch.randn(N, C, 7, 7) model = SqueezeAndExcitationBlock2D(in_channels=C, reduction=reduction) print(model) outputs = model(data) print(outputs.shape) assert (outputs.shape == (N, C, 7, 7))
class FilterResponseNorm2d(FilterResponseNormNd): def __init__(self, num_features, eps=1e-06, learnable_eps=False): super(FilterResponseNorm2d, self).__init__(4, num_features, eps=eps, learnable_eps=learnable_eps)
def Ranger(sync_period=6, slow_step_size=0.5, learning_rate=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-07, weight_decay=0.0, amsgrad=False, sma_threshold=5.0, total_steps=0, warmup_proportion=0.1, min_lr=0.0, name='Ranger'): inner = RectifiedAdam(learning_rate, beta_1, beta_2, epsilon, weight_decay, amsgrad, sma_threshold, total_steps, warmup_proportion, min_lr, name) optim = Lookahead(inner, sync_period, slow_step_size, name) return optim
class DoubleNode(ExprNode): def __init__(self, parse_info=None, raw_text=None): super().__init__(IRNodeType.Double, parse_info=parse_info, raw_text=raw_text) self.value = None
def test_fcm_normalization_nont1w_cli(image: pathlib.Path, mask: pathlib.Path) -> None: args = f'{image} -tm {mask} -mo t2'.split() retval = fcm_main(args) assert (retval == 0)
def dobldobl_ismember(wsys, gpts, dim, point, evatol=1e-06, memtol=1e-06, verbose=True, tasks=0): from phcpy.interface import store_dobldobl_witness_set from phcpy.phcpy2c3 import py2c_witset_dobldobl_ismember as membtest store_dobldobl_witness_set(len(wsys), dim, wsys, gpts) nbc = len(point) nvr = (len(wsys) - dim) if verbose: print('calling dobldobl_ismember with test point :') print(point) result = membtest(int(verbose), tasks, nvr, dim, nbc, evatol, memtol, point) (fail, onpolsys, inwitset) = result return (onpolsys, inwitset)
class FlattenMlp(Mlp): def forward(self, *inputs, **kwargs): flat_inputs = torch.cat(inputs, dim=1) return super().forward(flat_inputs, **kwargs)
class ContextNet(AcousticModel): def __init__(self, num_features: int, num_classes: int, kernel_size: int=3, num_blocks: int=6, num_layers: int=5, conv_out_channels: List[int]=[*([256] * 2), *([512] * 3), 640], subsampling_layers: List[int]=[1, 3], alpha: float=1.5, dropout: int=0.1): super().__init__() self.num_features = num_features self.num_classes = num_classes self.subsampling_factor = (2 * len(subsampling_layers)) conv_channels = ([num_features] + [int((channels * alpha)) for channels in conv_out_channels]) strides = ([1] * num_blocks) for layer in subsampling_layers: strides[layer] = 2 strides[layer] = 2 residuals = [False, *([True] * (num_blocks - 2)), False] blocks_num_layers = [1, *([num_layers] * (num_blocks - 2)), 1] self.block_list = [ContextNetBlock(conv_channels[i], conv_channels[(i + 1)], kernel_size=kernel_size, stride=strides[i], num_layers=blocks_num_layers[i], dropout=dropout, residual=residuals[i]) for i in range(num_blocks)] self.blocks = nn.Sequential(*self.block_list) self.output_layer = nn.Linear(conv_channels[(- 1)], num_classes) def forward(self, x, supervision=None): x = x.transpose(1, (- 1)) x = self.blocks(x) x = self.output_layer(x) x = nn.functional.log_softmax(x, dim=(- 1)).transpose(1, (- 1)) return (x, None, None)
class ObservationModel(nn.Module): def __init__(self, num_ensemble, dim_x, dim_z): super(ObservationModel, self).__init__() self.num_ensemble = num_ensemble self.dim_x = dim_x self.dim_z = dim_z self.linear1 = torch.nn.Linear(self.dim_x, 64) self.linear2 = torch.nn.Linear(64, 128) self.linear3 = torch.nn.Linear(128, 128) self.linear4 = torch.nn.Linear(128, 64) self.linear5 = torch.nn.Linear(64, self.dim_z) def forward(self, state): batch_size = state.shape[0] state = rearrange(state, 'bs k dim -> (bs k) dim', bs=batch_size, k=self.num_ensemble) x = self.linear1(state) x = F.relu(x) x = self.linear2(x) x = F.relu(x) x = self.linear3(x) x = F.relu(x) x = self.linear4(x) x = F.relu(x) z_pred = self.linear5(x) z_pred = rearrange(z_pred, '(bs k) dim -> bs k dim', bs=batch_size, k=self.num_ensemble) return z_pred
def _get_module_macs(module): s = module.__macs__ for child in module.children(): s += _get_module_macs(child) return s
class Grayscale(object): def __call__(self, img): gs = img.clone() gs[0].mul_(0.299).add_(0.587, gs[1]).add_(0.114, gs[2]) gs[1].copy_(gs[0]) gs[2].copy_(gs[0]) return gs
class sampler(Sampler): def __init__(self, train_size, batch_size): self.num_data = train_size self.num_per_batch = int((train_size / batch_size)) self.batch_size = batch_size self.range = torch.arange(0, batch_size).view(1, batch_size).long() self.leftover_flag = False if (train_size % batch_size): self.leftover = torch.arange((self.num_per_batch * batch_size), train_size).long() self.leftover_flag = True def __iter__(self): rand_num = (torch.randperm(self.num_per_batch).view((- 1), 1) * self.batch_size) self.rand_num = (rand_num.expand(self.num_per_batch, self.batch_size) + self.range) self.rand_num_view = self.rand_num.view((- 1)) if self.leftover_flag: self.rand_num_view = torch.cat((self.rand_num_view, self.leftover), 0) return iter(self.rand_num_view) def __len__(self): return self.num_data
_sentencepiece class M2M100TokenizationTest(TokenizerTesterMixin, unittest.TestCase): tokenizer_class = M2M100Tokenizer test_rust_tokenizer = False test_seq2seq = False test_sentencepiece = True def setUp(self): super().setUp() vocab = ['</s>', '<unk>', 'This', 'is', 'a', 't', 'est', 'G', '<pad>'] vocab_tokens = dict(zip(vocab, range(len(vocab)))) save_dir = Path(self.tmpdirname) save_json(vocab_tokens, (save_dir / VOCAB_FILES_NAMES['vocab_file'])) if (not (save_dir / VOCAB_FILES_NAMES['spm_file']).exists()): copyfile(SAMPLE_SP, (save_dir / VOCAB_FILES_NAMES['spm_file'])) tokenizer = M2M100Tokenizer.from_pretrained(self.tmpdirname) tokenizer.save_pretrained(self.tmpdirname) def get_tokenizer(self, **kwargs): return M2M100Tokenizer.from_pretrained(self.tmpdirname, **kwargs) def get_input_output_texts(self, tokenizer): return ('This is a test', 'This is a test') def test_convert_token_and_id(self): token = '</s>' token_id = 0 self.assertEqual(self.get_tokenizer()._convert_token_to_id(token), token_id) self.assertEqual(self.get_tokenizer()._convert_id_to_token(token_id), token) def test_get_vocab(self): tokenizer = self.get_tokenizer() vocab_keys = list(tokenizer.get_vocab().keys()) self.assertEqual(vocab_keys[0], '</s>') self.assertEqual(vocab_keys[1], '<unk>') self.assertEqual(vocab_keys[(- 1)], '<s>') self.assertEqual(len(vocab_keys), (tokenizer.vocab_size + len(tokenizer.get_added_vocab()))) ('Skip this test while all models are still to be uploaded.') def test_pretrained_model_lists(self): pass def test_full_tokenizer(self): tokenizer = self.get_tokenizer() tokens = tokenizer.tokenize('This is a test') self.assertListEqual(tokens, ['This', 'is', 'a', 't', 'est']) self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens), [2, 3, 4, 5, 6]) back_tokens = tokenizer.convert_ids_to_tokens([2, 3, 4, 5, 6]) self.assertListEqual(back_tokens, ['This', 'is', 'a', 't', 'est']) text = tokenizer.convert_tokens_to_string(tokens) self.assertEqual(text, 'This is a test') def test_tokenizer_integration(self): expected_encoding = {'input_ids': [[128022, 110108, 397, 11, 38272, 2247, 124811, 285, 18105, 1586, 207, 7, 39534, 4428, 397, 1019, 18105, 1586, 207, 7, 41337, 16786, 241, 7, 20214, 17, 125690, 10398, 7, 44378, 58069, 68342, 7798, 7343, 11, 299, 33310, 4, 158, 37350, 94077, 4569, 299, 33310, 90, 4, 52840, 290, 4, 31270, 112, 299, 682, 4, 52840, 39953, 14079, 193, 52519, 90894, 17894, 120697, 11, 40445, 551, 17, 1019, 52519, 90894, 17756, 963, 11, 40445, 480, 17, 9792, 1120, 5173, 1393, 6240, 16786, 241, 120996, 28, 1245, 1393, 118240, 11123, 1019, 93612, 2691, 10618, 98058, 120409, 1928, 279, 4, 40683, 367, 178, 207, 1019, 103, 103121, 506, 65296, 5, 2], [128022, 21217, 367, 117, 125450, 128, 719, 7, 7308, 40, 93612, 12669, 1116, 16704, 71, 17785, 3699, 15592, 35, 144, 9584, 241, 11943, 713, 950, 799, 2247, 88427, 150, 149, 118813, 120706, 1019, 106906, 81518, 28, 1224, 22799, 397, 5, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [128022, 1658, 123311, 5155, 5578, 4722, 279, 14947, 2366, 1120, 1197, 14, 1348, 9232, 5, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]} self.tokenizer_integration_test_util(expected_encoding=expected_encoding, model_name='facebook/m2m100_418M', revision='c168bae485c864188cf9aa0e4108b0b6934dc91e')
class chamfer_3DFunction(Function): def forward(ctx, xyz1, xyz2): (batchsize, n, _) = xyz1.size() (_, m, _) = xyz2.size() device = xyz1.device dist1 = torch.zeros(batchsize, n) dist2 = torch.zeros(batchsize, m) idx1 = torch.zeros(batchsize, n).type(torch.IntTensor) idx2 = torch.zeros(batchsize, m).type(torch.IntTensor) dist1 = dist1.to(device) dist2 = dist2.to(device) idx1 = idx1.to(device) idx2 = idx2.to(device) torch.cuda.set_device(device) chamfer_3D.forward(xyz1, xyz2, dist1, dist2, idx1, idx2) ctx.save_for_backward(xyz1, xyz2, idx1, idx2) return (dist1, dist2, idx1, idx2) def backward(ctx, graddist1, graddist2, gradidx1, gradidx2): (xyz1, xyz2, idx1, idx2) = ctx.saved_tensors graddist1 = graddist1.contiguous() graddist2 = graddist2.contiguous() device = graddist1.device gradxyz1 = torch.zeros(xyz1.size()) gradxyz2 = torch.zeros(xyz2.size()) gradxyz1 = gradxyz1.to(device) gradxyz2 = gradxyz2.to(device) chamfer_3D.backward(xyz1, xyz2, gradxyz1, gradxyz2, graddist1, graddist2, idx1, idx2) return (gradxyz1, gradxyz2)
def fdmobilenet_wd2_cub(num_classes=200, **kwargs): return get_mobilenet(num_classes=num_classes, version='fd', width_scale=0.5, model_name='fdmobilenet_wd2_cub', **kwargs)
def _get_log_dir(exec_func_name): cwd = pathlib.Path.cwd() return str(cwd.joinpath('data', 'local', 'benchmarks', exec_func_name))
def load_fountain_dataset(): rgbd_images = [] fountain_rgbd_dataset = o3d.data.SampleFountainRGBDImages() for i in range(len(fountain_rgbd_dataset.depth_paths)): depth = o3d.io.read_image(fountain_rgbd_dataset.depth_paths[i]) color = o3d.io.read_image(fountain_rgbd_dataset.color_paths[i]) rgbd_image = o3d.geometry.RGBDImage.create_from_color_and_depth(color, depth, convert_rgb_to_intensity=False) rgbd_images.append(rgbd_image) camera_trajectory = o3d.io.read_pinhole_camera_trajectory(fountain_rgbd_dataset.keyframe_poses_log_path) mesh = o3d.io.read_triangle_mesh(fountain_rgbd_dataset.reconstruction_path) return (mesh, rgbd_images, camera_trajectory)
class PostProcessor(abc.ABC): def __init__(self, tokenizer, ignore_pad_token_for_loss): self.tokenizer = tokenizer self.ignore_pad_token_for_loss = ignore_pad_token_for_loss def process(self, preds, labels, data_info=None): if isinstance(preds, tuple): preds = preds[0] if self.ignore_pad_token_for_loss: labels = np.where((labels != (- 100)), labels, self.tokenizer.pad_token_id) preds = np.where((preds != (- 100)), preds, self.tokenizer.pad_token_id) decoded_preds = self.tokenizer.batch_decode(preds, skip_special_tokens=True) decoded_labels = self.tokenizer.batch_decode(labels, skip_special_tokens=True) decoded_preds = [pred.strip() for pred in decoded_preds] decoded_labels = [label.strip() for label in decoded_labels] return (decoded_preds, decoded_labels)
class WavLMForAudioFrameClassification(metaclass=DummyObject): _backends = ['torch'] def __init__(self, *args, **kwargs): requires_backends(self, ['torch'])
class PSMDispProcessor(nn.Module): def __init__(self, max_disp=192): super().__init__() self.disp_processor = FasterSoftArgmin(max_disp=max_disp, start_disp=0, dilation=1, alpha=1.0, normalize=True) def forward(self, inputs): cost1 = inputs['cost1'] cost2 = inputs['cost2'] cost3 = inputs['cost3'] ref_img = inputs['ref_img'] tgt_img = inputs['tgt_img'] disp1 = self.disp_processor(cost1) disp2 = self.disp_processor(cost2) disp3 = self.disp_processor(cost3) if self.training: disp_gt = inputs['disp_gt'] output = {'training_disp': {'disp': {'disp_ests': [disp1, disp2, disp3], 'disp_gt': inputs['disp_gt'], 'mask': inputs['mask']}}, 'visual_summary': {'image/train/image_c': torch.cat([ref_img[0], tgt_img[0]], dim=1), 'image/train/disp_c': torch.cat([disp_gt[0], disp3[0]], dim=0)}} else: output = {'inference_disp': {'disp_est': disp3}, 'visual_summary': {'image/test/image_c': torch.cat([ref_img[0], tgt_img[0]], dim=1), 'image/test/disp_c': disp3[0]}} if ('disp_gt' in inputs): disp_gt = inputs['disp_gt'] output['visual_summary'] = {'image/val/image_c': torch.cat([ref_img[0], tgt_img[0]], dim=1), 'image/val/disp_c': torch.cat([disp_gt[0], disp3[0]], dim=0)} return output
def stat_coef_diff(X, X_tilde, y, method='lasso_cv', n_splits=5, n_jobs=1, n_lambdas=10, n_iter=1000, group_reg=0.001, l1_reg=0.001, joblib_verbose=0, return_coef=False, solver='liblinear', seed=0): n_features = X.shape[1] X_ko = np.column_stack([X, X_tilde]) lambda_max = (np.max(np.dot(X_ko.T, y)) / (2 * n_features)) lambdas = np.linspace((lambda_max * np.exp((- n_lambdas))), lambda_max, n_lambdas) cv = KFold(n_splits=5, shuffle=True, random_state=seed) estimator = {'lasso_cv': LassoCV(alphas=lambdas, n_jobs=n_jobs, verbose=joblib_verbose, max_iter=10000.0, cv=cv), 'logistic_l1': LogisticRegressionCV(penalty='l1', max_iter=10000.0, solver=solver, cv=cv, n_jobs=n_jobs, tol=1e-08), 'logistic_l2': LogisticRegressionCV(penalty='l2', max_iter=10000.0, n_jobs=n_jobs, verbose=joblib_verbose, cv=cv, tol=1e-08)} try: clf = estimator[method] except KeyError: print('{} is not a valid estimator'.format(method)) clf.fit(X_ko, y) try: coef = np.ravel(clf.coef_) except AttributeError: coef = np.ravel(clf.best_estimator_.coef_) test_score = (np.abs(coef[:n_features]) - np.abs(coef[n_features:])) if return_coef: return (test_score, coef) return test_score
class ResNetV2(nn.Module): def __init__(self, layers, channels=(256, 512, 1024, 2048), num_classes=1000, in_chans=3, global_pool='avg', output_stride=32, width_factor=1, stem_chs=64, stem_type='', avg_down=False, preact=True, act_layer=nn.ReLU, conv_layer=StdConv2d, norm_layer=partial(GroupNormAct, num_groups=32), drop_rate=0.0, drop_path_rate=0.0): super().__init__() self.num_classes = num_classes self.drop_rate = drop_rate wf = width_factor self.feature_info = [] stem_chs = make_div((stem_chs * wf)) self.stem = create_stem(in_chans, stem_chs, stem_type, preact, conv_layer=conv_layer, norm_layer=norm_layer) self.feature_info.append(dict(num_chs=stem_chs, reduction=2, module=('' if preact else 'stem.norm'))) prev_chs = stem_chs curr_stride = 4 dilation = 1 block_dprs = [x.tolist() for x in torch.linspace(0, drop_path_rate, sum(layers)).split(layers)] block_fn = (PreActBottleneck if preact else Bottleneck) self.stages = nn.Sequential() for (stage_idx, (d, c, bdpr)) in enumerate(zip(layers, channels, block_dprs)): out_chs = make_div((c * wf)) stride = (1 if (stage_idx == 0) else 2) if (curr_stride >= output_stride): dilation *= stride stride = 1 stage = ResNetStage(prev_chs, out_chs, stride=stride, dilation=dilation, depth=d, avg_down=avg_down, act_layer=act_layer, conv_layer=conv_layer, norm_layer=norm_layer, block_dpr=bdpr, block_fn=block_fn) prev_chs = out_chs curr_stride *= stride feat_name = f'stages.{stage_idx}' if preact: feat_name = (f'stages.{(stage_idx + 1)}.blocks.0.norm1' if ((stage_idx + 1) != len(channels)) else 'norm') self.feature_info += [dict(num_chs=prev_chs, reduction=curr_stride, module=feat_name)] self.stages.add_module(str(stage_idx), stage) self.num_features = prev_chs self.norm = (norm_layer(self.num_features) if preact else nn.Identity()) self.head = ClassifierHead(self.num_features, num_classes, pool_type=global_pool, drop_rate=self.drop_rate, use_conv=True) for (n, m) in self.named_modules(): if (isinstance(m, nn.Linear) or (('.fc' in n) and isinstance(m, nn.Conv2d))): nn.init.normal_(m.weight, mean=0.0, std=0.01) nn.init.zeros_(m.bias) elif isinstance(m, nn.Conv2d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') def get_classifier(self): return self.head.fc def reset_classifier(self, num_classes, global_pool='avg'): self.head = ClassifierHead(self.num_features, num_classes, pool_type=global_pool, drop_rate=self.drop_rate, use_conv=True) def forward_features(self, x): x = self.stem(x) x = self.stages(x) x = self.norm(x) return x def forward(self, x): x = self.forward_features(x) x = self.head(x) if (not self.head.global_pool.is_identity()): x = x.flatten(1) return x def load_pretrained(self, checkpoint_path, prefix='resnet/'): import numpy as np weights = np.load(checkpoint_path) with torch.no_grad(): stem_conv_w = tf2th(weights[f'{prefix}root_block/standardized_conv2d/kernel']) if (self.stem.conv.weight.shape[1] == 1): self.stem.conv.weight.copy_(stem_conv_w.sum(dim=1, keepdim=True)) else: self.stem.conv.weight.copy_(stem_conv_w) self.norm.weight.copy_(tf2th(weights[f'{prefix}group_norm/gamma'])) self.norm.bias.copy_(tf2th(weights[f'{prefix}group_norm/beta'])) self.head.fc.weight.copy_(tf2th(weights[f'{prefix}head/conv2d/kernel'])) self.head.fc.bias.copy_(tf2th(weights[f'{prefix}head/conv2d/bias'])) for (i, (sname, stage)) in enumerate(self.stages.named_children()): for (j, (bname, block)) in enumerate(stage.blocks.named_children()): convname = 'standardized_conv2d' block_prefix = f'{prefix}block{(i + 1)}/unit{(j + 1):02d}/' block.conv1.weight.copy_(tf2th(weights[f'{block_prefix}a/{convname}/kernel'])) block.conv2.weight.copy_(tf2th(weights[f'{block_prefix}b/{convname}/kernel'])) block.conv3.weight.copy_(tf2th(weights[f'{block_prefix}c/{convname}/kernel'])) block.norm1.weight.copy_(tf2th(weights[f'{block_prefix}a/group_norm/gamma'])) block.norm2.weight.copy_(tf2th(weights[f'{block_prefix}b/group_norm/gamma'])) block.norm3.weight.copy_(tf2th(weights[f'{block_prefix}c/group_norm/gamma'])) block.norm1.bias.copy_(tf2th(weights[f'{block_prefix}a/group_norm/beta'])) block.norm2.bias.copy_(tf2th(weights[f'{block_prefix}b/group_norm/beta'])) block.norm3.bias.copy_(tf2th(weights[f'{block_prefix}c/group_norm/beta'])) if (block.downsample is not None): w = weights[f'{block_prefix}a/proj/{convname}/kernel'] block.downsample.conv.weight.copy_(tf2th(w))
class CLIPVisionCfg(): layers: Union[(Tuple[(int, int, int, int)], int)] width: int head_width: int image_size: int mlp_ratio: float patch_size: int = None timm_model_name: str = None timm_model_pretrained: bool = None timm_pool: str = None timm_proj: str = None
class Host(): def __init__(self, ID, IPS, RAM, Disk, Bw, Latency, Powermodel, Environment): self.id = ID self.ipsCap = IPS self.ramCap = RAM self.diskCap = Disk self.bwCap = Bw self.latency = Latency self.powermodel = Powermodel self.powermodel.allocHost(self) self.powermodel.host = self self.env = Environment def getPower(self): return self.powermodel.power() def getPowerFromIPS(self, ips): return self.powermodel.powerFromCPU(min(100, (100 * (ips / self.ipsCap)))) def getCPU(self): ips = self.getApparentIPS() return (100 * (ips / self.ipsCap)) def getBaseIPS(self): ips = 0 containers = self.env.getContainersOfHost(self.id) for containerID in containers: ips += self.env.getContainerByID(containerID).getBaseIPS() return ips def getApparentIPS(self): ips = 0 containers = self.env.getContainersOfHost(self.id) for containerID in containers: ips += self.env.getContainerByID(containerID).getApparentIPS() return int(ips) def getIPSAvailable(self): return (self.ipsCap - self.getBaseIPS()) def getCurrentRAM(self): (size, read, write) = (0, 0, 0) containers = self.env.getContainersOfHost(self.id) for containerID in containers: (s, r, w) = self.env.getContainerByID(containerID).getRAM() size += s read += r write += w return (size, read, write) def getRAMAvailable(self): (size, read, write) = self.getCurrentRAM() return ((self.ramCap.size - size), (self.ramCap.read - read), (self.ramCap.write - write)) def getCurrentDisk(self): (size, read, write) = (0, 0, 0) containers = self.env.getContainersOfHost(self.id) for containerID in containers: (s, r, w) = self.env.getContainerByID(containerID).getDisk() size += s read += r write += w assert (size <= self.diskCap.size) assert (read <= self.diskCap.read) assert (write <= self.diskCap.write) return (size, read, write) def getDiskAvailable(self): (size, read, write) = self.getCurrentDisk() return ((self.diskCap.size - size), (self.diskCap.read - read), (self.diskCap.write - write))
class ClassAssocationRule(): id = 0 def __init__(self, antecedent, consequent, support, confidence): self.antecedent = antecedent self.consequent = consequent self.support = support self.confidence = confidence self.rulelen = (len(antecedent) + 1) self.rid = ClassAssocationRule.id ClassAssocationRule.id += 1 self.support_count = 0 self.marked = False self.class_cases_covered = collections.Counter() self.replace = set() def __gt__(self, other): if (self.confidence > other.confidence): return True elif ((self.confidence == other.confidence) and (self.support > other.support)): return True elif ((self.confidence == other.confidence) and (self.support == other.support) and (self.rulelen < other.rulelen)): return True elif ((self.confidence == other.confidence) and (self.support == other.support) and (self.rulelen == other.rulelen) and (self.rid < other.rid)): return True else: return False def __lt__(self, other): return (not (self > other)) def __len__(self): return (len(self.antecedent) + len(self.consequent)) def __repr__(self): args = [self.antecedent.string(), (('{' + self.consequent.string()) + '}'), self.support, self.confidence, self.rulelen, self.rid] text = 'CAR {} => {} sup: {:.2f} conf: {:.2f} len: {}, id: {}'.format(*args) return text
def add_prefix(inputs, prefix): outputs = dict() for (name, value) in inputs.items(): outputs[f'{prefix}.{name}'] = value return outputs
def get_symbol_edges(node: Union[(str, ast.AST)]) -> List[Tuple[(Union[(str, ast.AST)], ast.AST)]]: if isinstance(node, str): node = ast.parse(node).body[0] return GetSymbolEdges()(node)
def retrieve_data_cfg(config_path, skip_type): cfg = Config.fromfile(config_path) train_data_cfg = cfg.data.train if hasattr(train_data_cfg, 'pipeline'): train_data_cfg['pipeline'] = [x for x in train_data_cfg.pipeline if (x['type'] not in skip_type)] else: train_data_cfg['dataset']['pipeline'] = [x for x in train_data_cfg.dataset.pipeline if (x['type'] not in skip_type)] return cfg
def get_dataframes_model(all_results, datasets, model_name, divide=False): dataset_keys = list(all_results.keys()) if divide: df_avg_self = pd.DataFrame() df_avg_mt = pd.DataFrame() else: df_avg = {} for average in (['avg'] + list(languages.keys())): df_avg[average] = pd.DataFrame() for dataset in datasets: dfs = [] for dataset_key in dataset_keys: if dataset_key.startswith(dataset): if (model_name == 'open_llama_v2'): if (not any([(model.lower() in models['open_llama_v2']) for model in all_results[dataset_key]])): continue elif (not any([(model_name in model.lower()) for model in all_results[dataset_key]])): continue if (('600M' in dataset_key) or ('1.3B' in dataset_key)): continue results = pd.DataFrame(all_results[dataset_key]).T results['dataset'] = dataset_key results['model'] = [models_reverse[model] for model in results.index] results = results[(results['model'] == model_name)] results.drop(columns=['model'], inplace=True) results['model'] = model_name results['size'] = model_sizes[model_name][:len(results)] dfs.append(results) df_concat = pd.concat(dfs) df_concat = df_concat.sort_values(['size', 'dataset']) df_concat = df_concat.reindex(columns=(['model', 'size', 'dataset'] + [col for col in df_concat.columns if (col not in ['model', 'size', 'dataset'])])) df_concat['dataset'] = df_concat['dataset'].str.replace(dataset, 'Direct') df_concat['dataset'] = df_concat['dataset'].str.replace('Direct-mt_few-shot', 'Self-translate') df_concat['dataset'] = df_concat['dataset'].str.replace('Direct-mt_nllb-200-3.3B', 'MT (NLLB)') if divide: df_concat_self = df_concat[df_concat['dataset'].isin(['Direct', 'Self-MT'])] df_concat_mt = df_concat[df_concat['dataset'].isin(['Self-MT', 'MT'])] df_avg_self['model'] = df_concat_self['model'] df_avg_self['size'] = df_concat_self['size'] df_avg_self['dataset'] = df_concat_self['dataset'] df_avg_self[dataset] = df_concat_self['avg'] df_avg_mt['model'] = df_concat_mt['model'] df_avg_mt['size'] = df_concat_mt['size'] df_avg_mt['dataset'] = df_concat_mt['dataset'] df_avg_mt[dataset] = df_concat_mt['avg'] (yield df_concat_self) (yield df_concat_mt) else: for average in (['avg'] + list(languages.keys())): df_avg[average]['model'] = df_concat['model'] df_avg[average]['size'] = df_concat['size'] df_avg[average]['dataset'] = df_concat['dataset'] df_avg[average][dataset] = df_concat[average] (yield df_concat) if divide: size = df_avg_self['size'] df_avg_self = df_avg_self.drop(columns=['size']) size = df_avg_mt['size'] df_avg_mt = df_avg_mt.drop(columns=['size']) df_avg_self['avg'] = df_avg_self.mean(axis=1).round(1) df_avg_mt['avg'] = df_avg_mt.mean(axis=1).round(1) df_avg_self['size'] = size df_avg_mt['size'] = size (yield df_avg_self) (yield df_avg_mt) else: for average in (['avg'] + list(languages.keys())): size = df_avg[average]['size'] df_avg[average] = df_avg[average].drop(columns=['size']) df_avg[average][average] = df_avg[average].mean(axis=1).round(1) df_avg[average]['size'] = size (yield df_avg[average])
class VGGLoss_ESRGAN(nn.Module): def __init__(self): super().__init__() vgg19_model = models.vgg19(pretrained=True) self.vgg19_54 = nn.Sequential(*list(vgg19_model.features.children())[:35]) self.criterion = nn.L1Loss() def forward(self, real, fake): return self.criterion(self.vgg19_54(real), self.vgg19_54(fake))
def submit_training(**kws): dims = (N_CLUSTERS, kws['dim_l1'], kws['dim__adj'], kws['dim__v']) scenario_key = (kws['aid'], kws['cid']) model_key = (kws['model_type'], kws['spill_v2adj'], kws['ov'], kws['sampling_id']) if ((scenario_key, model_key) in kws['gathered']): history = None else: history = client.submit(make_train_vae, dims, kws['q_overlap'], kws['p__slices'], scenario_key, with_progress=False, key=('make_train_vae', id(make_train_vae), scenario_key, model_key)) return (scenario_key, model_key, history)
class ActionRepeat(object): def __init__(self, env, amount): self._env = env self._amount = amount def __getattr__(self, name): return getattr(self._env, name) def step(self, action): done = False total_reward = 0 current_step = 0 while ((current_step < self._amount) and (not done)): (observ, reward, done, info) = self._env.step(action) total_reward += reward current_step += 1 return (observ, total_reward, done, info)
def load_embedding(VOCAB, path, embedding_dim=300): with open(path) as f: weights = np.random.rand(VOCAB.get_vocab_size(), embedding_dim) counter = 0 for line in f.readlines(): try: line = line.strip().split() v = list(map(float, line[1:])) word = line[0] wid = VOCAB.get_index(word) if (wid != VOCAB.get_index('<unk>')): counter += 1 weights[wid] = np.array(v) except Exception as e: print(e) ipdb.set_trace() print(f'[!] Loading the weights {round((counter / VOCAB.get_vocab_size()), 4)}') return weights
def lanczos_generalized(operator, metric_operator=None, metric_inv_operator=None, num_eigenthings=10, which='LM', max_steps=20, tol=1e-06, num_lanczos_vectors=None, init_vec=None, use_gpu=False): if isinstance(operator.size, int): size = operator.size else: size = operator.size[0] shape = (size, size) if (num_lanczos_vectors is None): num_lanczos_vectors = min((2 * num_eigenthings), (size - 1)) if (num_lanczos_vectors < (2 * num_eigenthings)): warn('[lanczos] number of lanczos vectors should usually be > 2*num_eigenthings') def _scipy_apply(x): x = torch.from_numpy(x) if use_gpu: x = x.cuda() return operator.apply(x.float()).cpu().numpy() scipy_op = ScipyLinearOperator(shape, _scipy_apply) if (isinstance(metric_operator, np.ndarray) or isinstance(metric_operator, ScipyLinearOperator)): metric_op = metric_operator else: def _scipy_apply_metric(x): x = torch.from_numpy(x) if use_gpu: x = x.cuda() return metric_operator.apply(x.float()).cpu().numpy() metric_op = ScipyLinearOperator(shape, _scipy_apply_metric) if (isinstance(metric_inv_operator, np.ndarray) or isinstance(metric_inv_operator, ScipyLinearOperator)): metric_inv_op = metric_inv_operator else: def _scipy_apply_metric_inv(x): x = torch.from_numpy(x) if use_gpu: x = x.cuda() return metric_inv_operator.apply(x.float()).cpu().numpy() metric_inv_op = ScipyLinearOperator(shape, _scipy_apply_metric_inv) if (init_vec is None): init_vec = np.random.rand(size) elif isinstance(init_vec, torch.Tensor): init_vec = init_vec.cpu().numpy() (eigenvals, eigenvecs) = eigsh(A=scipy_op, k=num_eigenthings, M=metric_op, Minv=metric_inv_op, which=which, maxiter=max_steps, tol=tol, ncv=num_lanczos_vectors, return_eigenvectors=True) return (eigenvals, eigenvecs.T)
def train_data_loader(config, batch_size): train_dataset = DataSetMap[config.get('dataset', 'LinearDataset')](size=config.get('data_size', 1000), nested_input=config.get('nested_input', False)) train_loader = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size) return train_loader
class BasicBlock(nn.Module): expansion = 1 def __init__(self, inplanes, planes, stride=1, downsample=None, last=False): super(BasicBlock, self).__init__() self.conv1 = conv3x3(inplanes, planes, stride) self.bn1 = nn.BatchNorm2d(planes) self.relu = nn.ReLU(inplace=True) self.conv2 = conv3x3(planes, planes) self.bn2 = nn.BatchNorm2d(planes) self.downsample = downsample self.stride = stride self.last = last def forward(self, x): residual = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) if (self.downsample is not None): residual = self.downsample(x) out += residual out = self.relu(out) return out
class FINNExampleOverlay(Overlay): def __init__(self, bitfile_name, platform, io_shape_dict, batch_size=1, fclk_mhz=100.0, device=None, download=True, runtime_weight_dir='runtime_weights/'): super().__init__(bitfile_name, download=download, device=device) self.runtime_weight_dir = runtime_weight_dir self._io_shape_dict = io_shape_dict self.ibuf_packed_device = None self.obuf_packed_device = None self.platform = platform self.batch_size = batch_size self.fclk_mhz = fclk_mhz self.idma = [] self.odma = [] self.odma_handle = [] if ('input_dma_name' in io_shape_dict.keys()): for idma_name in io_shape_dict['input_dma_name']: self.idma.append(getattr(self, idma_name)) else: self.idma = [self.idma0] if ('output_dma_name' in io_shape_dict.keys()): for odma_name in io_shape_dict['output_dma_name']: self.odma.append(getattr(self, odma_name)) if (self.platform == 'alveo'): self.odma_handle.append(None) else: self.odma = [self.odma0] if (self.platform == 'alveo'): self.odma_handle.append(None) if (self.platform == 'zynq-iodma'): if (self.fclk_mhz > 0): Clocks.fclk0_mhz = self.fclk_mhz self.load_external_weights() self.load_runtime_weights() def load_external_weights(self): self.external_weights = [] w_filenames = [] if (not os.path.isdir(self.runtime_weight_dir)): return for (dirpath, dirnames, filenames) in os.walk(self.runtime_weight_dir): w_filenames.extend(filenames) tmp_weight_dict = {} for w_filename in w_filenames: if w_filename.endswith('.npy'): weight_tensor = np.load(((self.runtime_weight_dir + '/') + w_filename)) else: continue idma_name = w_filename.split('.')[0] tmp_weight_dict[idma_name] = weight_tensor for idma_name in tmp_weight_dict.keys(): if (idma_name in self.ip_dict.keys()): iwdma = getattr(self, idma_name) weight_tensor = tmp_weight_dict[idma_name] weight_buf = allocate(weight_tensor.shape, dtype=np.uint8) weight_buf[:] = weight_tensor weight_buf.flush() self.external_weights += [(iwdma, weight_buf, idma_name)] if ('number_of_external_weights' in self._io_shape_dict): hw_ext_weights = self._io_shape_dict['number_of_external_weights'] assert (len(self.external_weights) == hw_ext_weights), (('Number of hardware external weights and number of external ' + 'weight tensors available do not match. \n') + 'Is runtime_weight_dir pointing to the correct folder?') def load_runtime_weights(self, flush_accel=True, verify=True): w_filenames = [] if (not os.path.isdir(self.runtime_weight_dir)): return for (dirpath, dirnames, filenames) in os.walk(self.runtime_weight_dir): w_filenames.extend(filenames) rt_weight_dict = {} for w_filename in w_filenames: if w_filename.endswith('.dat'): with open(((self.runtime_weight_dir + '/') + w_filename), 'r') as f: dat = f.read() else: continue layer_w = np.fromiter([int(x, 16) for x in dat.strip().split()], dtype=np.uint32) sdp_ind = int(w_filename.split('_')[0]) layer_ind = int(w_filename.split('_')[1]) rt_weight_dict[(sdp_ind, layer_ind)] = layer_w for (sdp_ind, layer_ind) in rt_weight_dict.keys(): cand_if_name = ('StreamingDataflowPartition_%d/s_axilite_%d' % (sdp_ind, layer_ind)) if (cand_if_name in self.ip_dict.keys()): layer_mmio = getattr(getattr(self, ('StreamingDataflowPartition_%d' % sdp_ind)), ('s_axilite_%d' % layer_ind)).mmio layer_w = rt_weight_dict[(sdp_ind, layer_ind)] layer_mmio.write_mm(0, layer_w.tobytes()) if verify: new_w = np.copy(layer_mmio.array[:layer_w.shape[0]]) assert (layer_w == new_w).all() if flush_accel: self.execute_on_buffers() def idt(self, ind=0): return self._io_shape_dict['idt'][ind] def odt(self, ind=0): return self._io_shape_dict['odt'][ind] def ishape_normal(self, ind=0): ret = list(self._io_shape_dict['ishape_normal'][ind]) ret[0] = self.batch_size return tuple(ret) def oshape_normal(self, ind=0): ret = list(self._io_shape_dict['oshape_normal'][ind]) ret[0] = self.batch_size return tuple(ret) def ishape_folded(self, ind=0): ret = list(self._io_shape_dict['ishape_folded'][ind]) ret[0] = self.batch_size return tuple(ret) def oshape_folded(self, ind=0): ret = list(self._io_shape_dict['oshape_folded'][ind]) ret[0] = self.batch_size return tuple(ret) def ishape_packed(self, ind=0): ret = list(self._io_shape_dict['ishape_packed'][ind]) ret[0] = self.batch_size return tuple(ret) def oshape_packed(self, ind=0): ret = list(self._io_shape_dict['oshape_packed'][ind]) ret[0] = self.batch_size return tuple(ret) def num_inputs(self): return self._io_shape_dict['num_inputs'] def num_outputs(self): return self._io_shape_dict['num_outputs'] def batch_size(self): return self._batch_size _size.setter def batch_size(self, value): self._batch_size = value if (self.ibuf_packed_device is not None): self.ibuf_packed_device = None if (self.obuf_packed_device is not None): self.obuf_packed_device = None cacheable = {'alveo': False, 'zynq-iodma': True}[self.platform] self.ibuf_packed_device = [] self.obuf_packed_device = [] self.obuf_packed = [] for i in range(self.num_inputs): new_packed_ibuf = allocate(shape=self.ishape_packed(i), dtype=np.uint8, cacheable=cacheable) self.ibuf_packed_device.append(new_packed_ibuf) for o in range(self.num_outputs): new_packed_obuf = allocate(shape=self.oshape_packed(o), dtype=np.uint8, cacheable=cacheable) self.obuf_packed_device.append(new_packed_obuf) self.obuf_packed.append(np.empty_like(new_packed_obuf)) def fold_input(self, ibuf_normal, ind=0): assert (ibuf_normal.shape == self.ishape_normal(ind)) ibuf_folded = ibuf_normal.reshape(self.ishape_folded(ind)) return ibuf_folded def pack_input(self, ibuf_folded, ind=0): ibuf_packed = finnpy_to_packed_bytearray(ibuf_folded, self.idt(ind), reverse_endian=True, reverse_inner=True, fast_mode=True) return ibuf_packed def unpack_output(self, obuf_packed, ind=0): obuf_folded = packed_bytearray_to_finnpy(obuf_packed, self.odt(ind), self.oshape_folded(ind), reverse_endian=True, reverse_inner=True, fast_mode=True) return obuf_folded def unfold_output(self, obuf_folded, ind=0): obuf_normal = obuf_folded.reshape(self.oshape_normal(ind)) return obuf_normal def copy_input_data_to_device(self, data, ind=0): np.copyto(self.ibuf_packed_device[ind], data) self.ibuf_packed_device[ind].flush() def copy_output_data_from_device(self, data, ind=0): self.obuf_packed_device[ind].invalidate() np.copyto(data, self.obuf_packed_device[ind]) def execute_on_buffers(self, asynch=False, batch_size=None): if (batch_size is None): batch_size = self.batch_size assert (batch_size <= self.batch_size), 'Specified batch_size is too large.' if (self.platform == 'zynq-iodma'): for o in range(self.num_outputs): assert ((self.odma[o].read(0) & 4) != 0), ('Output DMA %d is not idle' % o) for (iwdma, iwbuf, iwdma_name) in self.external_weights: iwdma.write(16, iwbuf.device_address) iwdma.write(28, batch_size) iwdma.write(0, 1) for o in range(self.num_outputs): self.odma[o].write(16, self.obuf_packed_device[o].device_address) self.odma[o].write(28, batch_size) self.odma[o].write(0, 1) for i in range(self.num_inputs): self.idma[i].write(16, self.ibuf_packed_device[i].device_address) self.idma[i].write(28, batch_size) self.idma[i].write(0, 1) elif (self.platform == 'alveo'): for o in range(self.num_outputs): assert (self.odma_handle[o] is None), ('Output DMA %d is already running' % o) for i in range(self.num_inputs): self.idma[i].start(self.ibuf_packed_device[i], batch_size) for (iwdma, iwbuf, iwdma_name) in self.external_weights: iwdma.start(iwbuf, batch_size) for o in range(self.num_outputs): self.odma_handle[o] = self.odma[o].start(self.obuf_packed_device[o], batch_size) else: raise Exception(('Unrecognized platform: %s' % self.platform)) if (asynch is False): self.wait_until_finished() def wait_until_finished(self): if (self.platform == 'zynq-iodma'): for o in range(self.num_outputs): status = self.odma[o].read(0) while ((status & 2) == 0): status = self.odma[o].read(0) elif (self.platform == 'alveo'): assert all([(x is not None) for x in self.odma_handle]), 'No odma_handle to wait on' for o in range(self.num_outputs): self.odma_handle[o].wait() self.odma_handle[o] = None else: raise Exception(('Unrecognized platform: %s' % self.platform)) def execute(self, input_npy): if (not (type(input_npy) is list)): input_npy = [input_npy] assert (self.num_inputs == len(input_npy)), 'Not all accelerator inputs are specified.' for i in range(self.num_inputs): ibuf_folded = self.fold_input(input_npy[i], ind=i) ibuf_packed = self.pack_input(ibuf_folded, ind=i) self.copy_input_data_to_device(ibuf_packed, ind=i) self.execute_on_buffers() outputs = [] for o in range(self.num_outputs): self.copy_output_data_from_device(self.obuf_packed[o], ind=o) obuf_folded = self.unpack_output(self.obuf_packed[o], ind=o) obuf_normal = self.unfold_output(obuf_folded, ind=o) outputs.append(obuf_normal) if (self.num_outputs == 1): return outputs[0] else: return outputs def throughput_test(self): res = {} start = time.time() self.execute_on_buffers() end = time.time() runtime = (end - start) res['runtime[ms]'] = (runtime * 1000) res['throughput[images/s]'] = (self.batch_size / runtime) total_in = 0 for i in range(self.num_inputs): total_in += np.prod(self.ishape_packed(i)) res['DRAM_in_bandwidth[MB/s]'] = ((total_in * 1e-06) / runtime) total_out = 0 for o in range(self.num_outputs): total_out += np.prod(self.oshape_packed(o)) res['DRAM_out_bandwidth[MB/s]'] = ((total_out * 1e-06) / runtime) for (iwdma, iwbuf, iwdma_name) in self.external_weights: res[('DRAM_extw_%s_bandwidth[MB/s]' % iwdma_name)] = (((self.batch_size * np.prod(iwbuf.shape)) * 1e-06) / runtime) if (self.platform == 'zynq-iodma'): res['fclk[mhz]'] = Clocks.fclk0_mhz elif (self.platform == 'alveo'): res['fclk[mhz]'] = self.clock_dict['clock0']['frequency'] res['batch_size'] = self.batch_size input_npy = gen_finn_dt_tensor(self.idt(), self.ishape_normal()) if (self.idt() == DataType['UINT8']): input_npy = input_npy.astype(np.uint8) elif (self.idt() == DataType['INT8']): input_npy = input_npy.astype(np.int8) start = time.time() ibuf_folded = self.fold_input(input_npy) end = time.time() runtime = (end - start) res['fold_input[ms]'] = (runtime * 1000) start = time.time() ibuf_packed = self.pack_input(ibuf_folded) end = time.time() runtime = (end - start) res['pack_input[ms]'] = (runtime * 1000) start = time.time() self.copy_input_data_to_device(ibuf_packed) end = time.time() runtime = (end - start) res['copy_input_data_to_device[ms]'] = (runtime * 1000) start = time.time() self.copy_output_data_from_device(self.obuf_packed[0]) end = time.time() runtime = (end - start) res['copy_output_data_from_device[ms]'] = (runtime * 1000) start = time.time() obuf_folded = self.unpack_output(self.obuf_packed[0]) end = time.time() runtime = (end - start) res['unpack_output[ms]'] = (runtime * 1000) start = time.time() self.unfold_output(obuf_folded) end = time.time() runtime = (end - start) res['unfold_output[ms]'] = (runtime * 1000) return res
def mean_absolute_error(y_true, y_pred): result = mean(abs((y_true - y_pred)), axis=1) return result
_module() class CascadeRCNN(TwoStageDetector): 'Implementation of `Cascade R-CNN: Delving into High Quality Object\n Detection < def __init__(self, backbone: ConfigType, neck: OptConfigType=None, rpn_head: OptConfigType=None, roi_head: OptConfigType=None, train_cfg: OptConfigType=None, test_cfg: OptConfigType=None, data_preprocessor: OptConfigType=None, init_cfg: OptMultiConfig=None) -> None: super().__init__(backbone=backbone, neck=neck, rpn_head=rpn_head, roi_head=roi_head, train_cfg=train_cfg, test_cfg=test_cfg, data_preprocessor=data_preprocessor, init_cfg=init_cfg)
class BackendPytorchNative(backend.Backend): def __init__(self): super(BackendPytorchNative, self).__init__() self.sess = None self.model = None self.device = ('cuda:0' if torch.cuda.is_available() else 'cpu') def version(self): return torch.__version__ def name(self): return 'pytorch-native' def image_format(self): return 'NCHW' def load(self, model_path, inputs=None, outputs=None): self.model = torch.load(model_path, map_location=(lambda storage, loc: storage)) self.model.eval() if inputs: self.inputs = inputs else: self.inputs = [] initializers = set() for i in self.model.graph.initializer: initializers.add(i.name) for i in self.model.graph.input: if (i.name not in initializers): self.inputs.append(i.name) if outputs: self.outputs = outputs else: self.outputs = [] for i in self.model.graph.output: self.outputs.append(i.name) self.model = self.model.to(self.device) return self def predict(self, feed): key = [key for key in feed.keys()][0] feed[key] = torch.tensor(feed[key]).float().to(self.device) with torch.no_grad(): output = self.model(feed[key]) return output