search

migraphx-benchmark / AMDMIGraphX

AMD's graph optimization engine.
https://rocmsoftwareplatform.github.io/AMDMIGraphX/doc/html/
MIT License
0 stars 1 forks source link

ONNX Model Zoo models #140

Closed gyulaz-htec closed 9 months ago

gyulaz-htec commented 9 months ago

This is a list of MIGraphX errors from ONXX Model Zoo.

The main issues:

Model name Filename Opset version Error description Link
YOLOv3 yolov3-10.onnx 10 concat: all input dimensions should match along axis 2 https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/yolov3/model/yolov3-10.onnx
YOLOv3 yolov3-12.onnx 12 concat: all input dimensions should match along axis 2 https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/yolov3/model/yolov3-12.onnx
Tiny YOLOv3 tiny-yolov3-11.onnx 11 concat: all input dimensions should match along axis 2 https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/tiny-yolov3/model/tiny-yolov3-11.onnx
Udnie udnie-8.onnx 8 CONVOLUTION: mismatched channel numbers https://github.com/onnx/models/blob/main/vision/style_transfer/fast_neural_style/model/udnie-8.onnx
Pointilism pointilism-8.onnx 8 CONVOLUTION: mismatched channel numbers https://github.com/onnx/models/blob/main/vision/style_transfer/fast_neural_style/model/pointilism-8.onnx
Mosaic mosaic-8.onnx 8 CONVOLUTION: mismatched channel numbers https://github.com/onnx/models/blob/main/vision/style_transfer/fast_neural_style/model/mosaic-8.onnx
Candy candy-8.onnx 8 CONVOLUTION: mismatched channel numbers https://github.com/onnx/models/blob/main/vision/style_transfer/fast_neural_style/model/candy-8.onnx
Rain Princess rain-princess-8.onnx 8 CONVOLUTION: mismatched channel numbers https://github.com/onnx/models/blob/main/vision/style_transfer/fast_neural_style/model/rain-princess-8.onnx
ShuffleNet-v2-fp32 shufflenet-v2-12.onnx 12 Error with MLIR enabled:migraphx-driver: /code/AMDMIGraphX/depend/cget/build/tmp-063b48f8f3004e13bb6cf0cd082d884b/rocMLIR-12748a3402c069f733ea7f2ba1f8d8a070b3622a/external/llvm-project/mlir/lib/IR/PatternMatch.cpp:305: virtual void mlir::RewriterBase::eraseOp(Operation *): Assertion `op->use_empty() && "expected 'op' to have no uses"' failed. https://github.com/onnx/models/blob/main/vision/classification/shufflenet/model/shufflenet-v2-12.onnx
VGG 19-caffe2 vgg19-caffe2-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/vgg/model/vgg19-caffe2-3.onnx
CaffeNet caffenet-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/caffenet/model/caffenet-3.onnx
ResNet50-caffe2 resnet50-caffe2-v1-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/resnet/model/resnet50-caffe2-v1-3.onnx
ZFNet-512 zfnet512-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/zfnet-512/model/zfnet512-3.onnx
R-CNN ILSVRC13 rcnn-ilsvrc13-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/rcnn_ilsvrc13/model/rcnn-ilsvrc13-3.onnx
ShuffleNet-v1 shufflenet-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/shufflenet/model/shufflenet-3.onnx
AlexNet bvlcalexnet-3.onnx 3 parse: PARSE_GEMM: A and B should be rank 2, A is rank 4, B is rank 2 https://github.com/onnx/models/blob/main/vision/classification/alexnet/model/bvlcalexnet-3.onnx
GPT-2 with Beam Search Generation gpt2-lm-head-bs-12.onnx 12 parse_inputs: module "Loop_26_loop" has parameter name "vconst_100" existing in parent graph! https://github.com/onnx/models/blob/main/text/machine_comprehension/gpt2-bs/model/gpt2-lm-head-bs-12.onnx
Bidirectional Attention Flow bidaf-11-int8.onnx 11 Prototensor data type 8(string) not supported. Note: The models inputs are tokenized strings of context paragraph and query, this error probably is not relevant https://github.com/onnx/models/blob/main/text/machine_comprehension/bidirectional_attention_flow/model/bidaf-11-int8.onnx
SSD-MobileNetV1 ssd_mobilenet_v1_10.onnx 10 reshape: Wrong number of elements for reshape https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd-mobilenetv1/model/ssd_mobilenet_v1_10.onnx
SSD-MobileNetV1 ssd_mobilenet_v1_12.onnx 12 reshape: Wrong number of elements for reshape https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd-mobilenetv1/model/ssd_mobilenet_v1_12.onnx
SSD-MobileNetV1 ssd_mobilenet_v1_13-qdq.onnx 13 reshape: Wrong number of elements for reshape https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd-mobilenetv1/model/ssd_mobilenet_v1_13-qdq.onnx
Faster RCNN FasterRCNN-10.onnx 10 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/faster-rcnn/model/FasterRCNN-10.onnx
Faster RCNN FasterRCNN-12.onnx 12 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/faster-rcnn/model/FasterRCNN-12.onnx
Faster RCNN FasterRCNN-12-qdq.onnx 12 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/faster-rcnn/model/FasterRCNN-12-qdq.onnx
Mask-RCNN MaskRCNN-10.onnx 10 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/mask-rcnn/model/MaskRCNN-10.onnx
Mask-RCNN MaskRCNN-12.onnx 12 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/mask-rcnn/model/MaskRCNN-12.onnx
Mask-RCNN MaskRCNN-12-qdq.onnx 12 scatter_none: Shapes are not in standard layout https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/mask-rcnn/model/MaskRCNN-12-qdq.onnx
BERT-Squad bertsquad-12-int8.onnx 12 Unknown operator: DynamicQuantizeLinear https://github.com/onnx/models/blob/main/text/machine_comprehension/bert-squad/model/bertsquad-12-int8.onnx
VGG 16-int8 vgg16-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/vgg/model/vgg16-12-int8.onnx
CaffeNet-int8 caffenet-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/caffenet/model/caffenet-12-int8.onnx
EfficientNet-Lite4-int8 efficientnet-lite4-11-int8.onnx 11 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/efficientnet-lite4/model/efficientnet-lite4-11-int8.onnx
MobileNet v2-1.0-int8 mobilenetv2-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/mobilenet/model/mobilenetv2-12-int8.onnx
ResNet50-int8 resnet50-v1-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/resnet/model/resnet50-v1-12-int8.onnx
ZFNet-512-int8 zfnet512-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/zfnet-512/model/zfnet512-12-int8.onnx
MNIST-12-int8 mnist-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/mnist/model/mnist-12-int8.onnx
SqueezeNet 1.0 squeezenet1.0-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/squeezenet/model/squeezenet1.0-12-int8.onnx
ShuffleNet-v2-int8 shufflenet-v2-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/shufflenet/model/shufflenet-v2-12-int8.onnx
AlexNet-int8 bvlcalexnet-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/classification/alexnet/model/bvlcalexnet-12-int8.onnx
SSD ssd-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd/model/ssd-12-int8.onnx
SSD-MobileNetV1 ssd_mobilenet_v1_12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd-mobilenetv1/model/ssd_mobilenet_v1_12-int8.onnx
Faster RCNN FasterRCNN-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/faster-rcnn/model/FasterRCNN-12-int8.onnx
Mask-RCNN MaskRCNN-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/mask-rcnn/model/MaskRCNN-12-int8.onnx
DUC ResNet101-DUC-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/duc/model/ResNet101-DUC-12-int8.onnx
FCN fcn-resnet50-12-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/fcn/model/fcn-resnet50-12-int8.onnx
ArcFace arcfaceresnet100-11-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/body_analysis/arcface/model/arcfaceresnet100-11-int8.onnx
UltraFace version-RFB-320-int8.onnx 12 Unknown operator: QLinearConv https://github.com/onnx/models/blob/main/vision/body_analysis/ultraface/models/version-RFB-320-int8.onnx
YOLOv3 yolov3-12-int8.onnx 12 Unknown operator: QLinearLeakyRelu https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/yolov3/model/yolov3-12-int8.onnx
Emotion FerPlus emotion-ferplus-12-int8.onnx 12 Unknown operator: QLinearMatMul https://github.com/onnx/models/blob/main/vision/body_analysis/emotion_ferplus/model/emotion-ferplus-12-int8.onnx
DenseNet-121-12-int8 densenet-12-int8.onnx 12 Unknown operator: QLinearMul https://github.com/onnx/models/blob/main/vision/classification/densenet-121/model/densenet-12-int8.onnx