560f16bc5eb29fb2c2fff3a0e3db95a282d1002b,tools/test_encoder_decoder.py,,test_net_batch,#Any#Any#Any#Any#,104

Before Change

    saver = tf.train.Saver()

    // Set sess configuration
    sess_config = tf.ConfigProto(device_count={"GPU": 1})
    sess_config.gpu_options.per_process_gpu_memory_fraction = CFG.TEST.GPU_MEMORY_FRACTION
    sess_config.gpu_options.allow_growth = CFG.TRAIN.TF_ALLOW_GROWTH
    sess_config.gpu_options.allocator_type = "BFC"

After Change

    
    image_path_list = glob.glob("{:s}/**/*.jpg".format(src_image_dir), recursive=True)

    with tf.device("/gpu:1"):
        input_tensor = tf.placeholder(dtype=tf.float32, shape=[1, 512, 1024, 3], name="input_tensor")
        phase_tensor = tf.constant("train", dtype=tf.string)

        net = semantic_segmentation.SemanticSeg(net_flag=net_flag, phase=phase_tensor)
        with tf.variable_scope("kitti_loss"):
            net_out = net.build_model(input_tensor=input_tensor, name="inference")

        out_logits = tf.nn.softmax(net_out["logits"])
        out = tf.argmax(out_logits, axis=-1)
        out = tf.squeeze(input=out, axis=0)

    saver = tf.train.Saver()

    // Set sess configuration
    sess_config = tf.ConfigProto(allow_soft_placemeat=True)

Instances