Merge

examples/nlp/README.md

+# Natural Language Processing Inference Examples
+
+- [Python BERT-SQuAD](./python_bert_squad)
\ No newline at end of file

examples/python_bert_squad_example/BERT-Squad.ipynb → examples/nlp/python_bert_squad/BERT-Squad.ipynb

@@ -62,7 +62,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!wget -nc https://github.com/onnx/models/raw/master/text/machine_comprehension/bert-squad/model/bertsquad-10.onnx"
+    "!wget -nc https://github.com/onnx/models/raw/main/text/machine_comprehension/bert-squad/model/bertsquad-10.onnx"
    ]
   },
   {

examples/python_bert_squad_example/README.md → examples/nlp/python_bert_squad/README.md

@@ -23,7 +23,7 @@ unzip uncased_L-12_H-768_A-12.zip
 ```
 5) Get BERT ONNX model (bertsquad-10.onnx):
 ```
-wget https://github.com/onnx/models/raw/master/text/machine_comprehension/bert-squad/model/bertsquad-10.onnx
+wget https://github.com/onnx/models/raw/main/text/machine_comprehension/bert-squad/model/bertsquad-10.onnx
 ```
 6) Run the inference, it will compile and run the model on three questions and small data provided in `inputs.json`:
 ```

examples/python_bert_squad_example/requirements_bertsquad.txt → examples/nlp/python_bert_squad/requirements_bertsquad.txt

-tensorflow==2.5.0
+tensorflow==2.7.2
 onnxruntime
 tokenizers
\ No newline at end of file

examples/python_bert_squad_example/run_onnx_squad.py → examples/nlp/python_bert_squad/run_onnx_squad.py

+# Modifications Copyright (C) 2022, Advanced Micro Devices, Inc. All rights reserved
 # Copyright 2018 The Google AI Language Team Authors.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

examples/vision/README.md

+# Vision Inference Examples
+
+- [CPP MNIST](./cpp_mnist)
+- [Python Resnet50](./python_resnet50)
+- [Python Super Resolution](./python_super_resolution)
+- [Python NFNet](./python_nfnet)
+- [Python U-Net](./python_unet)
+- [Python 3D-UNet](./python_3dunet)
\ No newline at end of file

examples/cpp_api_inference/README.md → examples/vision/cpp_mnist/README.md

@@ -60,14 +60,14 @@ migraphx::quantize_int8(prog, targ, quant_opts);
 ## Compilation 
 Network graphs saved in e.g. ONNX or protobuf format are not target-specific. In order to run inference, we must compile the graph into a target-specific program. 
 
-Two options may be turned on (default for both is `false`) when compiling:
-- `bool offload_copy`: For targets with offloaded memory (such as the gpu), this will insert instructions during compilation to copy the input parameters to the offloaded memory and to copy the final result from the offloaded memory back to main memory.
-- `bool fast_math`: Optimize math functions to use faster approximate versions. There may be slight accuracy degredation when enabled. 
+Two options may be turned on when compiling:
+- `set_offload_copy(bool value)`: For targets with offloaded memory (such as the gpu), this will insert instructions during compilation to copy the input parameters to the offloaded memory and to copy the final result from the offloaded memory back to main memory. Default value is `false` for offload_copy.
+- `set_fast_math(bool value)`: Optimize math functions to use faster approximate versions. There may be slight accuracy degredation when enabled. Default value is `true` for fast_math. 
 
 The following snippet assumes `targ` has been set as "gpu", and will compile the program without the fast_math optimization.
 ```
-migraphx_compile_options comp_opts;
-comp_opts.offload_copy = true;
+migraphx::compile_options comp_opts;
+comp_opts.set_offload_copy();
 prog.compile(targ, comp_opts);
 ``` 
 

examples/cpp_api_inference/mnist_inference.cpp → examples/vision/cpp_mnist/mnist_inference.cpp

@@ -99,8 +99,8 @@ int main(int argc, char** argv)
 
     if(GPU)
     {
-        migraphx_compile_options comp_opts;
-        comp_opts.offload_copy = true;
+        migraphx::compile_options comp_opts;
+        comp_opts.set_offload_copy();
         prog.compile(targ, comp_opts);
     }
     else

examples/python_3dunet/3dunet_inference.ipynb → examples/vision/python_3dunet/3dunet_inference.ipynb

@@ -10,6 +10,16 @@
     "https://github.com/naomifridman/Unet_Brain_tumor_segmentation"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "09ceec31",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!pip install SimpleITK matplotlib scikit-image"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,