Hugging Face's logo

qualcomm
/
SESR-M5

Image-to-Image
PyTorch
LiteRT
ONNX
android
Model card Files Community
bhushans commited on
Commit
6d78176
·
verified ·
1 Parent(s): 8cc959b

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +44 -26
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
- | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.225 ms | 0 - 5 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
38
- | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.118 ms | 0 - 4 MB | FP16 | NPU | [SESR-M5.so](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.so) |
39
- | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.693 ms | 0 - 2 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
40
- | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.595 ms | 0 - 27 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
41
- | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.456 ms | 0 - 15 MB | FP16 | NPU | [SESR-M5.so](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.so) |
42
- | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.939 ms | 0 - 30 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
43
- | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.693 ms | 0 - 16 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
44
- | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.289 ms | 0 - 10 MB | FP16 | NPU | Use Export Script |
45
- | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.999 ms | 0 - 16 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
46
- | SESR-M5 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.234 ms | 0 - 3 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
47
- | SESR-M5 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.176 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
48
- | SESR-M5 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.208 ms | 0 - 22 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
49
- | SESR-M5 | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.141 ms | 0 - 5 MB | FP16 | NPU | Use Export Script |
50
- | SESR-M5 | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 2.203 ms | 0 - 8 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
51
- | SESR-M5 | SA8775 (Proxy) | SA8775P Proxy | QNN | 2.146 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
52
- | SESR-M5 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.149 ms | 0 - 6 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
53
- | SESR-M5 | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.151 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
54
- | SESR-M5 | SA8295P ADP | SA8295P | TFLITE | 4.143 ms | 6 - 21 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
55
- | SESR-M5 | SA8295P ADP | SA8295P | QNN | 3.474 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
56
- | SESR-M5 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.4 ms | 6 - 33 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
57
- | SESR-M5 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 3.144 ms | 0 - 15 MB | FP16 | NPU | Use Export Script |
58
- | SESR-M5 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.357 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
59
- | SESR-M5 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.955 ms | 9 - 9 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
 
 
60
 
61
 
62
 
@@ -120,8 +122,8 @@ Profiling Results
120
  SESR-M5
121
  Device : Samsung Galaxy S23 (13)
122
  Runtime : TFLITE
123
- Estimated inference time (ms) : 2.2
124
- Estimated peak memory usage (MB): [0, 5]
125
  Total # Ops : 25
126
  Compute Unit(s) : NPU (22 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.sesr_m5 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
+ | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.29 ms | 0 - 4 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
38
+ | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.146 ms | 0 - 70 MB | FP16 | NPU | [SESR-M5.so](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.so) |
39
+ | SESR-M5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.68 ms | 0 - 2 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
40
+ | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.556 ms | 0 - 12 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
41
+ | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.466 ms | 0 - 13 MB | FP16 | NPU | [SESR-M5.so](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.so) |
42
+ | SESR-M5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.942 ms | 0 - 30 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
43
+ | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.718 ms | 0 - 10 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
44
+ | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.337 ms | 0 - 10 MB | FP16 | NPU | Use Export Script |
45
+ | SESR-M5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 2.002 ms | 0 - 17 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
46
+ | SESR-M5 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.201 ms | 0 - 5 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
47
+ | SESR-M5 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.144 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
48
+ | SESR-M5 | SA7255P ADP | SA7255P | TFLITE | 78.598 ms | 6 - 16 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
49
+ | SESR-M5 | SA7255P ADP | SA7255P | QNN | 77.493 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
50
+ | SESR-M5 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.203 ms | 6 - 74 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
51
+ | SESR-M5 | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.165 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
52
+ | SESR-M5 | SA8295P ADP | SA8295P | TFLITE | 6.026 ms | 6 - 14 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
53
+ | SESR-M5 | SA8295P ADP | SA8295P | QNN | 3.576 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
54
+ | SESR-M5 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.163 ms | 0 - 77 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
55
+ | SESR-M5 | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.179 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
56
+ | SESR-M5 | SA8775P ADP | SA8775P | TFLITE | 4.441 ms | 6 - 15 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
57
+ | SESR-M5 | SA8775P ADP | SA8775P | QNN | 4.51 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
58
+ | SESR-M5 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.661 ms | 0 - 13 MB | FP16 | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) |
59
+ | SESR-M5 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 3.219 ms | 0 - 15 MB | FP16 | NPU | Use Export Script |
60
+ | SESR-M5 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.379 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
61
+ | SESR-M5 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.001 ms | 9 - 9 MB | FP16 | NPU | [SESR-M5.onnx](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx) |
62
 
63
 
64
 
 
122
  SESR-M5
123
  Device : Samsung Galaxy S23 (13)
124
  Runtime : TFLITE
125
+ Estimated inference time (ms) : 2.3
126
+ Estimated peak memory usage (MB): [0, 4]
127
  Total # Ops : 25
128
  Compute Unit(s) : NPU (22 ops) CPU (3 ops)
129
  ```
 
144
  import torch
145
 
146
  import qai_hub as hub
147
+ from qai_hub_models.models.sesr_m5 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