Hugging Face's logo

qualcomm
/
QuickSRNetSmall

Image-to-Image
PyTorch
LiteRT
ONNX
android
Model card Files Community
qaihm-bot commited on
Commit
3e67953
·
verified ·
1 Parent(s): 98b214d

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +43 -25
README.md CHANGED
@@ -34,29 +34,31 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.344 ms | 0 - 1 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
38
- | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.003 ms | 0 - 2 MB | FP16 | NPU | [QuickSRNetSmall.so](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.so) |
39
- | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.431 ms | 0 - 1 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
40
- | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.85 ms | 0 - 19 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
41
- | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.621 ms | 0 - 10 MB | FP16 | NPU | [QuickSRNetSmall.so](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.so) |
42
- | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 0.952 ms | 0 - 21 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
43
- | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.915 ms | 0 - 14 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
44
- | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.67 ms | 0 - 8 MB | FP16 | NPU | Use Export Script |
45
- | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.798 ms | 0 - 14 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
46
- | QuickSRNetSmall | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.298 ms | 0 - 8 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
47
- | QuickSRNetSmall | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.839 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
48
- | QuickSRNetSmall | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.326 ms | 1 - 3 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
49
- | QuickSRNetSmall | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.861 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
50
- | QuickSRNetSmall | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 1.297 ms | 0 - 8 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
51
- | QuickSRNetSmall | SA8775 (Proxy) | SA8775P Proxy | QNN | 0.868 ms | 0 - 4 MB | FP16 | NPU | Use Export Script |
52
- | QuickSRNetSmall | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.308 ms | 0 - 1 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
53
- | QuickSRNetSmall | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.852 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
54
- | QuickSRNetSmall | SA8295P ADP | SA8295P | TFLITE | 2.066 ms | 6 - 19 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
55
- | QuickSRNetSmall | SA8295P ADP | SA8295P | QNN | 1.795 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
56
- | QuickSRNetSmall | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.414 ms | 6 - 27 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
57
- | QuickSRNetSmall | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.111 ms | 0 - 13 MB | FP16 | NPU | Use Export Script |
58
- | QuickSRNetSmall | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.985 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
59
- | QuickSRNetSmall | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.445 ms | 8 - 8 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
 
 
60
 
61
 
62
 
@@ -121,7 +123,7 @@ QuickSRNetSmall
121
  Device : Samsung Galaxy S23 (13)
122
  Runtime : TFLITE
123
  Estimated inference time (ms) : 1.3
124
- Estimated peak memory usage (MB): [0, 1]
125
  Total # Ops : 11
126
  Compute Unit(s) : NPU (8 ops) CPU (3 ops)
127
  ```
@@ -142,13 +144,29 @@ in memory using the `jit.trace` and then call the `submit_compile_job` API.
142
  import torch
143
 
144
  import qai_hub as hub
145
- from qai_hub_models.models.quicksrnetsmall import
146
 
147
  # Load the model
 
148
 
149
  # Device
150
  device = hub.Device("Samsung Galaxy S23")
151
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
152
 
153
  ```
154
 
 
34
 
35
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.318 ms | 0 - 4 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
38
+ | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.008 ms | 0 - 26 MB | FP16 | NPU | [QuickSRNetSmall.so](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.so) |
39
+ | QuickSRNetSmall | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.445 ms | 0 - 2 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
40
+ | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.88 ms | 0 - 11 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
41
+ | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.633 ms | 0 - 10 MB | FP16 | NPU | [QuickSRNetSmall.so](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.so) |
42
+ | QuickSRNetSmall | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 0.978 ms | 0 - 21 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
43
+ | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.982 ms | 0 - 8 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
44
+ | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.58 ms | 0 - 8 MB | FP16 | NPU | Use Export Script |
45
+ | QuickSRNetSmall | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.977 ms | 0 - 14 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
46
+ | QuickSRNetSmall | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.34 ms | 0 - 2 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
47
+ | QuickSRNetSmall | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.842 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
48
+ | QuickSRNetSmall | SA7255P ADP | SA7255P | TFLITE | 10.099 ms | 6 - 15 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
49
+ | QuickSRNetSmall | SA7255P ADP | SA7255P | QNN | 8.869 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
50
+ | QuickSRNetSmall | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.357 ms | 0 - 3 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
51
+ | QuickSRNetSmall | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.857 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
52
+ | QuickSRNetSmall | SA8295P ADP | SA8295P | TFLITE | 2.006 ms | 0 - 7 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
53
+ | QuickSRNetSmall | SA8295P ADP | SA8295P | QNN | 1.83 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
54
+ | QuickSRNetSmall | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.364 ms | 0 - 4 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
55
+ | QuickSRNetSmall | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.872 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
56
+ | QuickSRNetSmall | SA8775P ADP | SA8775P | TFLITE | 2.384 ms | 1 - 8 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
57
+ | QuickSRNetSmall | SA8775P ADP | SA8775P | QNN | 1.682 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
58
+ | QuickSRNetSmall | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 2.059 ms | 0 - 12 MB | FP16 | NPU | [QuickSRNetSmall.tflite](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.tflite) |
59
+ | QuickSRNetSmall | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.107 ms | 0 - 13 MB | FP16 | NPU | Use Export Script |
60
+ | QuickSRNetSmall | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.95 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
61
+ | QuickSRNetSmall | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.522 ms | 9 - 9 MB | FP16 | NPU | [QuickSRNetSmall.onnx](https://huggingface.co/qualcomm/QuickSRNetSmall/blob/main/QuickSRNetSmall.onnx) |
62
 
63
 
64
 
 
123
  Device : Samsung Galaxy S23 (13)
124
  Runtime : TFLITE
125
  Estimated inference time (ms) : 1.3
126
+ Estimated peak memory usage (MB): [0, 4]
127
  Total # Ops : 11
128
  Compute Unit(s) : NPU (8 ops) CPU (3 ops)
129
  ```
 
144
  import torch
145
 
146
  import qai_hub as hub
147
+ from qai_hub_models.models.quicksrnetsmall import Model
148
 
149
  # Load the model
150
+ torch_model = Model.from_pretrained()
151
 
152
  # Device
153
  device = hub.Device("Samsung Galaxy S23")
154
 
155
+ # Trace model
156
+ input_shape = torch_model.get_input_spec()
157
+ sample_inputs = torch_model.sample_inputs()
158
+
159
+ pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
160
+
161
+ # Compile model on a specific device
162
+ compile_job = hub.submit_compile_job(
163
+ model=pt_model,
164
+ device=device,
165
+ input_specs=torch_model.get_input_spec(),
166
+ )
167
+
168
+ # Get target model to run on-device
169
+ target_model = compile_job.get_target_model()
170
 
171
  ```
172