Hugging Face's logo

qualcomm
/
ResNeXt50

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
98c2ea2
·
verified ·
1 Parent(s): 71fd91d

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +29 -28
README.md CHANGED
@@ -35,39 +35,39 @@ More details on model performance across various devices, can be found
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.488 ms | 0 - 191 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
39
- | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.563 ms | 1 - 131 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
40
- | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.681 ms | 0 - 159 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
41
- | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.767 ms | 0 - 38 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
42
- | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.864 ms | 0 - 40 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
43
- | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 2.079 ms | 0 - 42 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
44
- | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.702 ms | 0 - 41 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
45
- | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.772 ms | 1 - 41 MB | FP16 | NPU | Use Export Script |
46
- | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.949 ms | 0 - 41 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
47
- | ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.482 ms | 0 - 181 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
48
- | ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.52 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
49
- | ResNeXt50 | SA7255P ADP | SA7255P | TFLITE | 85.661 ms | 0 - 37 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
50
- | ResNeXt50 | SA7255P ADP | SA7255P | QNN | 85.858 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
- | ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.483 ms | 0 - 190 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
52
- | ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.496 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
53
- | ResNeXt50 | SA8295P ADP | SA8295P | TFLITE | 3.967 ms | 0 - 27 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
54
- | ResNeXt50 | SA8295P ADP | SA8295P | QNN | 4.295 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
55
- | ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.481 ms | 0 - 191 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
56
- | ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.512 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
57
- | ResNeXt50 | SA8775P ADP | SA8775P | TFLITE | 4.611 ms | 0 - 37 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
58
- | ResNeXt50 | SA8775P ADP | SA8775P | QNN | 4.667 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
59
  | ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.246 ms | 0 - 27 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
60
- | ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 3.426 ms | 1 - 34 MB | FP16 | NPU | Use Export Script |
61
- | ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.717 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
- | ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.652 ms | 51 - 51 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
69
- This model can be installed as a Python package via pip.
70
 
 
71
  ```bash
72
  pip install qai-hub-models
73
  ```
@@ -124,7 +124,7 @@ ResNeXt50
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
  Estimated inference time (ms) : 2.5
127
- Estimated peak memory usage (MB): [0, 191]
128
  Total # Ops : 79
129
  Compute Unit(s) : NPU (79 ops)
130
  ```
@@ -151,7 +151,7 @@ from qai_hub_models.models.resnext50 import Model
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
- device = hub.Device("Samsung Galaxy S23")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
@@ -243,7 +243,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
243
 
244
 
245
  ## License
246
- * The license for the original implementation of ResNeXt50 can be found [here](https://github.com/pytorch/vision/blob/main/LICENSE).
 
247
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
248
 
249
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.468 ms | 0 - 180 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
39
+ | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.543 ms | 1 - 131 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
40
+ | ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.786 ms | 0 - 163 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
41
+ | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.756 ms | 0 - 38 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
42
+ | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.846 ms | 1 - 40 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
43
+ | ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 2.107 ms | 0 - 40 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
44
+ | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.681 ms | 0 - 42 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
45
+ | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.511 ms | 1 - 41 MB | FP16 | NPU | Use Export Script |
46
+ | ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.99 ms | 1 - 43 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
47
+ | ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.482 ms | 0 - 191 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
48
+ | ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.522 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
49
+ | ResNeXt50 | SA7255P ADP | SA7255P | TFLITE | 85.485 ms | 0 - 38 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
50
+ | ResNeXt50 | SA7255P ADP | SA7255P | QNN | 85.779 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
+ | ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.473 ms | 0 - 191 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
52
+ | ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.53 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
53
+ | ResNeXt50 | SA8295P ADP | SA8295P | TFLITE | 3.969 ms | 0 - 27 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
54
+ | ResNeXt50 | SA8295P ADP | SA8295P | QNN | 4.305 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
55
+ | ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.493 ms | 0 - 181 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
56
+ | ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.51 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
57
+ | ResNeXt50 | SA8775P ADP | SA8775P | TFLITE | 4.599 ms | 0 - 37 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
58
+ | ResNeXt50 | SA8775P ADP | SA8775P | QNN | 4.696 ms | 0 - 10 MB | FP16 | NPU | Use Export Script |
59
  | ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.246 ms | 0 - 27 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
60
+ | ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 3.363 ms | 1 - 28 MB | FP16 | NPU | Use Export Script |
61
+ | ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.63 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
+ | ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.642 ms | 52 - 52 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
 
69
 
70
+ Install the package via pip:
71
  ```bash
72
  pip install qai-hub-models
73
  ```
 
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
  Estimated inference time (ms) : 2.5
127
+ Estimated peak memory usage (MB): [0, 180]
128
  Total # Ops : 79
129
  Compute Unit(s) : NPU (79 ops)
130
  ```
 
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
+ device = hub.Device("Samsung Galaxy S24")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
 
243
 
244
 
245
  ## License
246
+ * The license for the original implementation of ResNeXt50 can be found
247
+ [here](https://github.com/pytorch/vision/blob/main/LICENSE).
248
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
249
 
250