Hi Jorge,

Jorge_García_Martíne · ‎11-29-2019

Hello everybody,

when I try to execute the Inference Engine python API with "HETERO:FPGA,CPU" device I have the following error:

exec_net = ie.load_network(network=net, device_name=args.device)
File "ie_api.pyx", line 85, in openvino.inference_engine.ie_api.IECore.load_network
File "ie_api.pyx", line 92, in openvino.inference_engine.ie_api.IECore.load_network
RuntimeError: Failed to call QueryNetwork for FPGA device, error: Failed to create FPGA device handle with ID 0

In CPU the example is OK and when I execute the demos (demo_security_barrier_camera.sh and demo_squeezenet_download_convert_run.sh) with HETERO:FPGA,CPU is OK too.

The FPGA diagnose device is OK.

Any help is welcome!

Thanks...

Sahira_Intel · ‎12-02-2019

Hi Jorge,

What was the command you used to deploy your model? Can you please give more information about your model?

Best Regards,

Sahira

Jorge_García_Martíne · ‎12-02-2019

Hello, I have developed the following model:

Layer (type) Output Shape Param #
=================================================================
conv2d_3 (Conv2D) (None, 24, 24, 32) 832
_________________________________________________________________
max_pooling2d_3 (MaxPooling2 (None, 12, 12, 32) 0
_________________________________________________________________
conv2d_4 (Conv2D) (None, 8, 8, 64) 51264
_________________________________________________________________
max_pooling2d_4 (MaxPooling2 (None, 4, 4, 64) 0
_________________________________________________________________
flatten_2 (Flatten) (None, 1024) 0
_________________________________________________________________
dense_2 (Dense) (None, 10) 10250

I've run the model optimizer for FP16 precision.

The code for Inference Engine tool with HETERO:FPGA,CPU is (similar to classification_sample.py):

#!/usr/bin/env python
"""
 Copyright (C) 2018-2019 Intel Corporation

 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
 You may obtain a copy of the License at

      http://www.apache.org/licenses/LICENSE-2.0

 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
"""
from __future__ import print_function
import sys
import os
from argparse import ArgumentParser, SUPPRESS
import cv2
import numpy as np
import logging as log
from time import time
from openvino.inference_engine import IENetwork, IECore


def build_argparser():
    parser = ArgumentParser(add_help=False)
    args = parser.add_argument_group('Options')
    args.add_argument('-h', '--help', action='help', default=SUPPRESS, help='Show this help message and exit.')
    args.add_argument("-m", "--model", help="Required. Path to an .xml file with a trained model.", required=True,
                      type=str)
    args.add_argument("-i", "--input", help="Required. Path to a folder with images or path to an image files",
                      required=True,
                      type=str, nargs="+")
    args.add_argument("-l", "--cpu_extension",
                      help="Optional. Required for CPU custom layers. "
                           "MKLDNN (CPU)-targeted custom layers. Absolute path to a shared library with the"
                           " kernels implementations.", type=str, default=None)
    args.add_argument("-d", "--device",
                      help="Optional. Specify the target device to infer on; CPU, GPU, FPGA, HDDL, MYRIAD or HETERO: is "
                           "acceptable. The sample will look for a suitable plugin for device specified. Default "
                           "value is CPU",
                      default="CPU", type=str)
    args.add_argument("--labels", help="Optional. Path to a labels mapping file", default=None, type=str)
    args.add_argument("-nt", "--number_top", help="Optional. Number of top results", default=10, type=int)

    return parser


def main():
    log.basicConfig(format="[ %(levelname)s ] %(message)s", level=log.INFO, stream=sys.stdout)
    args = build_argparser().parse_args()
    model_xml = args.model
    model_bin = os.path.splitext(model_xml)[0] + ".bin"

    # Plugin initialization for specified device and load extensions library if specified
    log.info("Creating Inference Engine")
    ie = IECore()
    if args.cpu_extension and 'CPU' in args.device:
        ie.add_extension(args.cpu_extension, "CPU")


    # Read IR
    log.info("Loading network files:\n\t{}\n\t{}".format(model_xml, model_bin))
    net = IENetwork(model=model_xml, weights=model_bin)

    if "CPU" in args.device:
        supported_layers = ie.query_network(net, "CPU")
        not_supported_layers = [l for l in net.layers.keys() if l not in supported_layers]
        if len(not_supported_layers) != 0:
            log.error("Following layers are not supported by the plugin for specified device {}:\n {}".
                      format(args.device, ', '.join(not_supported_layers)))
            log.error("Please try to specify cpu extensions library path in sample's command line parameters using -l "
                      "or --cpu_extension command line argument")
            sys.exit(1)

    assert len(net.inputs.keys()) == 1, "Sample supports only single input topologies"
    assert len(net.outputs) == 1, "Sample supports only single output topologies"

    log.info("Preparing input blobs")
    input_blob = next(iter(net.inputs))
    out_blob = next(iter(net.outputs))
    net.batch_size = len(args.input)
    print("PARAMETROS RED")
    print("net.imputs: ", input_blob)
    print("net.outputs: ", out_blob)
    print("batch_size: ", net.batch_size)

    # Read and pre-process input images
    n, c, h, w = net.inputs[input_blob].shape
    images = np.ndarray(shape=(n, c, h, w))
    print("Valor de n: ", n)
    print("Valor de c: ", c)
    print("Valor de h: ", h)
    print("Valor de w: ", w)
    print("Images", images.shape)

    for i in range(n):
        image = cv2.imread(args.input)
        if image.shape[:-1] != (h, w):
            log.warning("Image {} is resized from {} to {}".format(args.input, image.shape[:-1], (h, w)))
            image = cv2.resize(image, (w, h))
        image = image.transpose((2, 0, 1))  # Change data layout from HWC to CHW
        images = image[2,:,:]
    log.info("Batch size is {}".format(n))


    # Loading model to the plugin
    log.info("Loading model to the plugin")
    exec_net = ie.load_network(network=net, device_name=args.device)

    # Start sync inference
    log.info("Starting inference in synchronous mode")
    res = exec_net.infer(inputs={input_blob: images})

    # Processing output blob
    log.info("Processing output blob")
    res = res[out_blob]
    log.info("Top {} results: ".format(args.number_top))
    if args.labels:
        with open(args.labels, 'r') as f:
            labels_map = [x.split(sep=' ', maxsplit=1)[-1].strip() for x in f]
    else:
        labels_map = None
    classid_str = "classid"
    probability_str = "probability"
    for i, probs in enumerate(res):
        probs = np.squeeze(probs)
        top_ind = np.argsort(probs)[-args.number_top:][::-1]
        print("Image {}\n".format(args.input))
        print(classid_str, probability_str)
        print("{} {}".format('-' * len(classid_str), '-' * len(probability_str)))
        for id in top_ind:
            det_label = labels_map[id] if labels_map else "{}".format(id)
            label_length = len(det_label)
            space_num_before = (len(classid_str) - label_length) // 2
            space_num_after = len(classid_str) - (space_num_before + label_length) + 2
            space_num_before_prob = (len(probability_str) - len(str(probs[id]))) // 2
            print("{}{}{}{}{:.7f}".format(' ' * space_num_before, det_label,
                                          ' ' * space_num_after, ' ' * space_num_before_prob,
                                          probs[id]))
        print("\n")
    log.info("This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool\n")

if __name__ == '__main__':
    sys.exit(main() or 0)

And the comand to run Inference Engine:

python3 cero_IE.py -m /home/cero_red_CNN/Inference_Engine_FPGA/FP16/tf_model0_fp16.xml -i /home/cero_red_CNN/Inference_Engine_FPGA/FP16/Images/cuatro.png -d HETERO:FPGA,CPU

Thanks for all. Jorge.

Sahira_Intel · ‎12-02-2019

Hi Jorge,

When you converted your model, did you use --data_type FP16? If so, you used the conversion to run on Myriad (VPU). If you want to run on the CPU, convert your model with the parameter --data_type FP32.

I hope this information is helpful.

Best,

Sahira

Jorge_García_Martíne · ‎12-03-2019

Hi Sahira,

I've converted my model with FP32 and I have the same problem, the bit stream and the 'aocl diagnose'' are:

Vendor: Intel Corp

Physical Dev Name Status Information

pac_a10_ec00000 Passed PAC Arria 10 Platform (pac_a10_ec00000)
PCIe 59:00.0
FPGA temperature = 51 degrees C.

DIAGNOSTIC_PASSED
--------------------------------------------------------------------

Bitstream --> /opt/intel/openvino_fpga_2019.1.144/bitstreams/a10_dcp_bitstreams/2019R1_RC_FP16_ResNet_SqueezeNet_VGG.aocx

Sahira_Intel · ‎12-13-2019

Hi Jorge,

Can you please run the Benchmark Application with the parameter -d HETERO:FPGA,CPU? Are you getting the same errors?

Best,

Sahira

Jorge_García_Martíne · ‎12-16-2019

Hi Sahira,

I get the same errors. Any idea about the problem?

Thanks, Jorge.

NSVR · ‎01-26-2021

Did you solve this issue. I am also facing the same. Your help is highly appreciated.

E-Hong · ‎05-01-2021

Hi,

Have you solved the problem? I have the same issue on DevCloud too.

Sahira_Intel · ‎12-20-2019

Hi Jorge,

It looks like you're using an older version of OpenVINO - can you please try upgrading to the latest version of OpenVINO?

Best Regards,

Sahira

Inference Engine Python API Error