Python mxnet.ndarray.expand_dims() Examples

def face_create_lib(args, npz_embs, npz_emb_len, embedding, item, fd, max_nums):
# 得到当前ID的总个数
    id_sum = npz_embs.shape[0]

    # 为新的人脸入库，创建一个新的文件夹
    new_lib_dir = os.path.join(args.outdir, '%08d' % id_sum)
    os.mkdir(new_lib_dir)

    # 为了统一，扩展一个维度
    embedding = np.expand_dims(embedding, axis=0)

    # 特征向量以及对应的ID图片的数目，都进行垂直拼接
    npz_embs = np.vstack((npz_embs, embedding.reshape(1, -1)))
    npz_emb_len = np.vstack((npz_emb_len, np.array([[1]])))

    new_img_path = os.path.join(new_lib_dir, '00000' + args.encoding)
    old_img_path = os.path.join(args.indir, item[1])

    fd.write(old_img_path + '\t' + new_img_path + '\t' + str(max_nums) + '\n\n')
    shutil.copyfile(old_img_path, new_img_path)

    if args.delete:
        os.remove(old_img_path)
    return npz_embs,npz_emb_len 

def tensor_load_rgbimage(filename, ctx, size=None, scale=None, keep_asp=False):
    img = Image.open(filename).convert('RGB')
    if size is not None:
        if keep_asp:
            size2 = int(size * 1.0 / img.size[0] * img.size[1])
            img = img.resize((size, size2), Image.ANTIALIAS)
        else:
            img = img.resize((size, size), Image.ANTIALIAS)

    elif scale is not None:
        img = img.resize((int(img.size[0] / scale), int(img.size[1] / scale)), Image.ANTIALIAS)
    img = np.array(img).transpose(2, 0, 1).astype(float)
    img = F.expand_dims(mx.nd.array(img, ctx=ctx), 0)
    return img 

def unsqueeze(input, dim):
    return nd.expand_dims(input, axis=dim) 

def batched_l2_dist(a, b):
    a_squared = nd.power(nd.norm(a, axis=-1), 2)
    b_squared = nd.power(nd.norm(b, axis=-1), 2)

    squared_res = nd.add(nd.linalg_gemm(
        a, nd.transpose(b, axes=(0, 2, 1)), nd.broadcast_axes(nd.expand_dims(b_squared, axis=-2), axis=1, size=a.shape[1]), alpha=-2
    ), nd.expand_dims(a_squared, axis=-1))
    res = nd.sqrt(nd.clip(squared_res, 1e-30, np.finfo(np.float32).max))
    return res 

def batched_l1_dist(a, b):
    a = nd.expand_dims(a, axis=-2)
    b = nd.expand_dims(b, axis=-3)
    res = nd.norm(a - b, ord=1, axis=-1)
    return res 

def prepare(self, g, gpu_id, trace=False):
        head_ids, tail_ids = g.all_edges(order='eid')
        projection = self.projection_emb(g.edata['id'], gpu_id, trace)
        projection = projection.reshape(-1, self.entity_dim, self.relation_dim)
        head_emb = g.ndata['emb'][head_ids.as_in_context(g.ndata['emb'].context)].expand_dims(axis=-2)
        tail_emb = g.ndata['emb'][tail_ids.as_in_context(g.ndata['emb'].context)].expand_dims(axis=-2)
        g.edata['head_emb'] = nd.batch_dot(head_emb, projection).squeeze()
        g.edata['tail_emb'] = nd.batch_dot(tail_emb, projection).squeeze() 

def edge_func(self, edges):
        head = edges.src['emb']
        tail = edges.dst['emb'].expand_dims(2)
        rel = edges.data['emb']
        rel = rel.reshape(-1, self.relation_dim, self.entity_dim)
        score = head * mx.nd.batch_dot(rel, tail).squeeze()
        # TODO: check if use self.gamma
        return {'score': mx.nd.sum(score, -1)}
        # return {'score': self.gamma - th.norm(score, p=1, dim=-1)} 

def update(self, gpu_id=-1):
        """ Update embeddings in a sparse manner
        Sparse embeddings are updated in mini batches. we maintains gradient states for 
        each embedding so they can be updated separately.

        Parameters
        ----------
        gpu_id : int
            Which gpu to accelerate the calculation. if -1 is provided, cpu is used.
        """
        self.state_step += 1
        for idx, data in self.trace:
            grad = data.grad

            clr = self.args.lr
            #clr = self.args.lr / (1 + (self.state_step - 1) * group['lr_decay'])

            # the update is non-linear so indices must be unique
            grad_indices = idx
            grad_values = grad

            grad_sum = (grad_values * grad_values).mean(1)
            ctx = self.state_sum.context
            if ctx != grad_indices.context:
                grad_indices = grad_indices.as_in_context(ctx)
            if ctx != grad_sum.context:
                grad_sum = grad_sum.as_in_context(ctx)
            self.state_sum[grad_indices] += grad_sum
            std = self.state_sum[grad_indices]  # _sparse_mask
            if gpu_id >= 0:
                std = std.as_in_context(mx.gpu(gpu_id))
            std_values = nd.expand_dims(nd.sqrt(std) + 1e-10, 1)
            tmp = (-clr * grad_values / std_values)
            if tmp.context != ctx:
                tmp = tmp.as_in_context(ctx)
            # TODO(zhengda) the overhead is here.
            self.emb[grad_indices] = mx.nd.take(self.emb, grad_indices) + tmp
        self.trace = [] 

def test_quantization(model, args, test_data, size, num_class, pred_offset):
    # output folder
    outdir = 'outdir_int8'
    if not os.path.exists(outdir):
        os.makedirs(outdir)
    print(model)
    metric = gluoncv.utils.metrics.SegmentationMetric(num_class)

    tbar = tqdm(test_data)
    metric.reset()
    tic = time.time()
    for i, (batch, dsts) in enumerate(tbar):
        if args.eval:
            targets = mx.gluon.utils.split_and_load(dsts, ctx_list=args.ctx, even_split=False)
            data = mx.gluon.utils.split_and_load(batch, ctx_list=args.ctx, batch_axis=0, even_split=False)
            outputs = None
            for x in data:
                output = model(x)
                outputs = output if outputs is None else nd.concat(outputs, output, axis=0)
            metric.update(targets, outputs)
            pixAcc, mIoU = metric.get()
            tbar.set_description('pixAcc: %.4f, mIoU: %.4f' % (pixAcc, mIoU))
        else:
            for data, impath in zip(batch, dsts):
                data = data.as_in_context(args.ctx[0])
                if len(data.shape) < 4:
                    data = nd.expand_dims(data, axis=0)
                predict = model(data)[0]
                predict = mx.nd.squeeze(mx.nd.argmax(predict, 1)).asnumpy() + pred_offset
                mask = get_color_pallete(predict, args.dataset)
                outname = os.path.splitext(impath)[0] + '.png'
                mask.save(os.path.join(outdir, outname))
    speed = size / (time.time() - tic)
    print('Inference speed with batchsize %d is %.2f img/sec' % (args.batch_size, speed)) 

def tensor_load_rgbimage(filename, ctx, size=None, scale=None, keep_asp=False):
    img = Image.open(filename).convert('RGB')
    if size is not None:
        if keep_asp:
            size2 = int(size * 1.0 / img.size[0] * img.size[1])
            img = img.resize((size, size2), Image.ANTIALIAS)
        else:
            img = img.resize((size, size), Image.ANTIALIAS)

    elif scale is not None:
        img = img.resize((int(img.size[0] / scale), int(img.size[1] / scale)), Image.ANTIALIAS)
    img = np.array(img).transpose(2, 0, 1).astype(float)
    img = F.expand_dims(mx.nd.array(img, ctx=ctx), 0)
    return img 

def generate_text(model, seed, length=512, top_n=10):
    """
    generates text of specified length from trained model
    with given seed character sequence.
    """
    logger.info("generating %s characters from top %s choices.", length, top_n)
    logger.info('generating with seed: "%s".', seed)
    generated = seed
    encoded = mx.nd.array(encode_text(seed))
    seq_len = encoded.shape[0]

    x = F.expand_dims(encoded[:seq_len-1], 1)
    # input shape: [seq_len, 1]
    state = model.begin_state()
    # get rnn state due to seed sequence
    _, state = model(x, state)

    next_index = encoded[seq_len-1].asscalar()
    for i in range(length):
        x = mx.nd.array([[next_index]])
        # input shape: [1, 1]
        logit, state = model(x, state)
        # output shape: [1, vocab_size]
        probs = F.softmax(logit)
        next_index = sample_from_probs(probs.asnumpy().squeeze(), top_n)
        # append to sequence
        generated += ID2CHAR[next_index]

    logger.info("generated text: \n%s\n", generated)
    return generated 

def predict(self,img):
        img = nd.array(img)
        #print(img.shape)
        img = nd.transpose(img, axes=(2, 0, 1)).astype('float32')
        img = nd.expand_dims(img, axis=0)
        #print(img.shape)
        db = mx.io.DataBatch(data=(img,))

        self.model.forward(db, is_train=False)
        net_out = self.model.get_outputs()
        embedding = net_out[0].asnumpy()
        embedding = sklearn.preprocessing.normalize(embedding,axis=1)
        return embedding 

def predict(self, img):
        img = nd.array(img)
        img = nd.transpose(img, axes=(2, 0, 1)).astype('float32')
        img = nd.expand_dims(img, axis=0)
        # print(img.shape)
        db = mx.io.DataBatch(data=(img,))

        self.model.forward(db, is_train=False)
        net_out = self.model.get_outputs()
        embedding = net_out[0].asnumpy()
        embedding = sklearn.preprocessing.normalize(embedding)
        return embedding 

def predict(self,img):
        img = nd.array(img)
        img = nd.transpose(img, axes=(2, 0, 1)).astype('float32')
        img = nd.expand_dims(img, axis=0)
        #print(img.shape)
        db = mx.io.DataBatch(data=(img,))

        self.model.forward(db, is_train=False)
        net_out = self.model.get_outputs()
        embedding = net_out[0].asnumpy()
        embedding = sklearn.preprocessing.normalize(embedding)
        return embedding 

def tensor_load_rgbimage(filename, ctx, size=None, scale=None, keep_asp=False):
    img = Image.open(filename).convert('RGB')
    if size is not None:
        if keep_asp:
            size2 = int(size * 1.0 / img.size[0] * img.size[1])
            img = img.resize((size, size2), Image.ANTIALIAS)
        else:
            img = img.resize((size, size), Image.ANTIALIAS)

    elif scale is not None:
        img = img.resize((int(img.size[0] / scale), int(img.size[1] / scale)), Image.ANTIALIAS)
    img = np.array(img).transpose(2, 0, 1).astype(float)
    img = F.expand_dims(mx.nd.array(img, ctx=ctx), 0)
    return img 

def tensor_load_rgbimage(filename, ctx, size=None, scale=None, keep_asp=False):
    img = Image.open(filename).convert('RGB')
    if size is not None:
        if keep_asp:
            size2 = int(size * 1.0 / img.size[0] * img.size[1])
            img = img.resize((size, size2), Image.ANTIALIAS)
        else:
            img = img.resize((size, size), Image.ANTIALIAS)

    elif scale is not None:
        img = img.resize((int(img.size[0] / scale), int(img.size[1] / scale)), Image.ANTIALIAS)
    img = np.array(img).transpose(2, 0, 1).astype(float)
    img = F.expand_dims(mx.nd.array(img, ctx=ctx), 0)
    return img