Create examples.md

Author: bhushan23

examples/examples.md

@@ -0,0 +1,58 @@
+### Converting PyTorch model to CoreML model (Short version)
+Converting PyTorch model into CoreML model is a two step process
+1. Convert PyTorch model into ONNX model
+  - PyTorch model can be converted into ONNX model using `torch.onnx.export`
+  - Reference: https://pytorch.org/docs/stable/onnx.html#id2
+  - Tools required: [PyTorch](https://pytorch.org/get-started/locally/)
+2. Converting ONNX model into CoreML model
+  - We will be using model converted in step 1 using `onnx_coreml.convert`
+  - Tools required: [onnx-coreml](https://pypi.org/project/onnx-coreml/)
+  
+
+### Converting PyTorch model to CoreML model (Long version)
+1. PyTorch to ONNX conversion
+  - In this example, we are creating a small test model
+    ```
+      import torch
+      import torch.nn as nn
+      import torch.nn.functional as F
+
+      # Step 0 - (a) Define ML Model
+      class small_model(nn.Module):
+          def __init__(self):
+              super(small_model, self).__init__()
+              self.fc1 = nn.Linear(768, 256)
+              self.fc2 = nn.Linear(256, 10)
+
+          def forward(self, x):
+              y = F.relu(self.fc1(x))
+              y = F.softmax(self.fc2(y))
+              return y
+    ```
+  - Load model
+    ```
+      # Step 0 - (b) Create model or Load from distk
+      model = small_model()
+      dummy_input = torch.randn(768)
+    ```
+  - Convert From PyTorch to ONNX
+    ```
+      # Step 1 - PyTorch to ONNX model
+      torch.onnx.export(model, dummy_input, './small_model.onnx')
+    ```
+2. ONNX to CoreML
+    ```
+      # Step 2 - ONNX to CoreML model
+      mlmodel = convert(model='./small_model.onnx', target_ios='13')
+      # Save converted CoreML model
+      mlmodel.save('small_model.mlmodel')
+    ```
+
+### What about frameworks other than PyTorch?
+  - Step 1 can be replaced by respective framework to ONNX converter,
+  - Once, you have a ONNX model, you can use onnx_coreml.convert to get CoreML model
+    which can be deployed on device.
+    
+### More examples demonstrating prediction and validating converted model's correctness
+  - BERT: https://github.com/onnx/onnx-coreml/blob/master/examples/BERT.ipynb
+  - All: https://github.com/onnx/onnx-coreml/blob/master/examples/