You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
### The files in this repo (and a description of what they do)
18
18
```
19
-
├── demo
20
-
│ ├── test.html (a minimal test for making sure the generated ONNX model works, uses ONNX.js to load and run the generated ONNX model)
21
-
│ └── onnx_model.onnx (a copy of the generated ONNX model, used by the test code)
22
-
├── convert_to_onnx.py (converts a trained PyTorch model into an ONNX model)
23
-
├── inference_mnist_model.py (the PyTorch model description (without the trained parameters) used by convert_to_onnx.py to generate the ONNX model)
24
-
├── inputs_batch_preview.png (a preview of a batch of augmented input data, generated by preview_mnist_dataset.py)
25
-
├── onnx_model.py (the ONNX model generated by convert_to_onnx.py)
26
-
├── preview_dataset.py (for testing out different types of data augmentation)
27
-
├── pytorch_model.pt (the trained PyTorch model parameters, generated by train_mnist.model.py, and used by convert_to_onnx.py to generate the ONNX model)
28
-
└── train_mnist_model.pt (trains the PyTorch model and saves the trained parameters as pytorch_model.pt)
19
+
├── degug_demo
20
+
│ ├── debug.html (A minimal test for making sure the generated ONNX model works. Uses ONNX.js to load and run the generated ONNX model)
21
+
│ └── onnx_model.onnx (A copy of the generated ONNX model. Used by debug.html.)
22
+
├── full_demo
23
+
│ ├── index.html (The full demo's HTML code.)
24
+
│ ├── onnx_model.onnx (A copy of the generated ONNX model. Used by script.js.)
25
+
│ ├── script.js (The full demos's JS code. Loads the onnx_model.onnx and predicts the drawn numbers.)
26
+
│ └── style.css (The full demo's CSS.)
27
+
├── convert_to_onnx.py (Converts a trained PyTorch model into an ONNX model.)
28
+
├── inference_mnist_model.py (The PyTorch model description (without the trained parameters). Used by convert_to_onnx.py to generate the ONNX model.)
29
+
├── inputs_batch_preview.png (A preview of a batch of augmented input data. Generated by preview_mnist_dataset.py.)
30
+
├── onnx_model.py (The ONNX model generated by convert_to_onnx.py.)
31
+
├── preview_dataset.py (For testing out different types of data augmentation.)
32
+
├── pytorch_model.pt (The trained PyTorch model parameters. Generated by train_mnist.model.py and used by convert_to_onnx.py to generate the ONNX model.)
33
+
└── train_mnist_model.pt (Trains the PyTorch model and saves the trained parameters as pytorch_model.pt.)
29
34
```
30
35
31
36
### The benefits of running a model in the browser:
0 commit comments