Hugging Face's logo

qualcomm
/
DETR-ResNet50-DC5

Object Detection
PyTorch
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
1a522a9
·
verified ·
1 Parent(s): aa8af6f

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +31 -31
README.md CHANGED
@@ -34,45 +34,44 @@ 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
- | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 84.57 ms | 0 - 45 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
38
- | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 86.746 ms | 3 - 48 MB | FP16 | NPU | [DETR-ResNet50-DC5.so](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.so) |
39
- | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 75.578 ms | 1 - 210 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
40
- | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 60.231 ms | 0 - 149 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
41
- | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 61.935 ms | 0 - 149 MB | FP16 | NPU | [DETR-ResNet50-DC5.so](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.so) |
42
- | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 51.971 ms | 2 - 187 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
43
- | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 56.498 ms | 0 - 155 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
44
- | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 57.393 ms | 3 - 156 MB | FP16 | NPU | Use Export Script |
45
- | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 45.682 ms | 3 - 172 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
46
- | DETR-ResNet50-DC5 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 84.921 ms | 0 - 45 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
47
- | DETR-ResNet50-DC5 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 94.141 ms | 3 - 6 MB | FP16 | NPU | Use Export Script |
48
- | DETR-ResNet50-DC5 | SA7255P ADP | SA7255P | TFLITE | 631.414 ms | 0 - 154 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
49
- | DETR-ResNet50-DC5 | SA7255P ADP | SA7255P | QNN | 630.56 ms | 0 - 9 MB | FP16 | NPU | Use Export Script |
50
- | DETR-ResNet50-DC5 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 84.161 ms | 0 - 45 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
51
- | DETR-ResNet50-DC5 | SA8255 (Proxy) | SA8255P Proxy | QNN | 98.32 ms | 3 - 5 MB | FP16 | NPU | Use Export Script |
52
- | DETR-ResNet50-DC5 | SA8295P ADP | SA8295P | TFLITE | 95.042 ms | 0 - 136 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
53
- | DETR-ResNet50-DC5 | SA8295P ADP | SA8295P | QNN | 75.531 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
54
- | DETR-ResNet50-DC5 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 85.415 ms | 0 - 48 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
55
- | DETR-ResNet50-DC5 | SA8650 (Proxy) | SA8650P Proxy | QNN | 98.007 ms | 3 - 5 MB | FP16 | NPU | Use Export Script |
56
- | DETR-ResNet50-DC5 | SA8775P ADP | SA8775P | TFLITE | 95.743 ms | 0 - 154 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
57
- | DETR-ResNet50-DC5 | SA8775P ADP | SA8775P | QNN | 98.602 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
58
- | DETR-ResNet50-DC5 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 98.161 ms | 0 - 135 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
59
- | DETR-ResNet50-DC5 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 96.055 ms | 3 - 138 MB | FP16 | NPU | Use Export Script |
60
- | DETR-ResNet50-DC5 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 78.6 ms | 3 - 3 MB | FP16 | NPU | Use Export Script |
61
- | DETR-ResNet50-DC5 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 65.667 ms | 84 - 84 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
62
 
63
 
64
 
65
 
66
  ## Installation
67
 
68
- This model can be installed as a Python package via pip.
69
 
 
70
  ```bash
71
- pip install "qai-hub-models[detr_resnet50_dc5]"
72
  ```
73
 
74
 
75
-
76
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
77
 
78
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -124,7 +123,7 @@ DETR-ResNet50-DC5
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
  Estimated inference time (ms) : 84.6
127
- Estimated peak memory usage (MB): [0, 45]
128
  Total # Ops : 789
129
  Compute Unit(s) : NPU (789 ops)
130
  ```
@@ -151,7 +150,7 @@ from qai_hub_models.models.detr_resnet50_dc5 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 +242,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
243
 
244
 
245
  ## License
246
- * The license for the original implementation of DETR-ResNet50-DC5 can be found [here](https://github.com/facebookresearch/detr/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
 
 
34
 
35
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 84.554 ms | 0 - 46 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
38
+ | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 87.135 ms | 3 - 48 MB | FP16 | NPU | [DETR-ResNet50-DC5.so](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.so) |
39
+ | DETR-ResNet50-DC5 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 76.799 ms | 1 - 249 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
40
+ | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 60.359 ms | 0 - 148 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
41
+ | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 60.483 ms | 183 - 330 MB | FP16 | NPU | [DETR-ResNet50-DC5.so](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.so) |
42
+ | DETR-ResNet50-DC5 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 52.476 ms | 3 - 185 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
43
+ | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 49.686 ms | 0 - 157 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
44
+ | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 58.148 ms | 3 - 157 MB | FP16 | NPU | Use Export Script |
45
+ | DETR-ResNet50-DC5 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 42.806 ms | 3 - 171 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
46
+ | DETR-ResNet50-DC5 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 85.22 ms | 0 - 45 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
47
+ | DETR-ResNet50-DC5 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 87.958 ms | 3 - 5 MB | FP16 | NPU | Use Export Script |
48
+ | DETR-ResNet50-DC5 | SA7255P ADP | SA7255P | TFLITE | 631.56 ms | 0 - 154 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
49
+ | DETR-ResNet50-DC5 | SA7255P ADP | SA7255P | QNN | 631.581 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
50
+ | DETR-ResNet50-DC5 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 87.617 ms | 0 - 46 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
51
+ | DETR-ResNet50-DC5 | SA8255 (Proxy) | SA8255P Proxy | QNN | 94.499 ms | 3 - 6 MB | FP16 | NPU | Use Export Script |
52
+ | DETR-ResNet50-DC5 | SA8295P ADP | SA8295P | TFLITE | 94.937 ms | 0 - 136 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
53
+ | DETR-ResNet50-DC5 | SA8295P ADP | SA8295P | QNN | 75.408 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
54
+ | DETR-ResNet50-DC5 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 84.113 ms | 0 - 40 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
55
+ | DETR-ResNet50-DC5 | SA8650 (Proxy) | SA8650P Proxy | QNN | 99.625 ms | 3 - 6 MB | FP16 | NPU | Use Export Script |
56
+ | DETR-ResNet50-DC5 | SA8775P ADP | SA8775P | TFLITE | 95.76 ms | 0 - 154 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
57
+ | DETR-ResNet50-DC5 | SA8775P ADP | SA8775P | QNN | 98.526 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
58
+ | DETR-ResNet50-DC5 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 104.266 ms | 0 - 131 MB | FP16 | NPU | [DETR-ResNet50-DC5.tflite](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.tflite) |
59
+ | DETR-ResNet50-DC5 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 104.678 ms | 4 - 137 MB | FP16 | NPU | Use Export Script |
60
+ | DETR-ResNet50-DC5 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 76.04 ms | 3 - 3 MB | FP16 | NPU | Use Export Script |
61
+ | DETR-ResNet50-DC5 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 65.409 ms | 84 - 84 MB | FP16 | NPU | [DETR-ResNet50-DC5.onnx](https://huggingface.co/qualcomm/DETR-ResNet50-DC5/blob/main/DETR-ResNet50-DC5.onnx) |
62
 
63
 
64
 
65
 
66
  ## Installation
67
 
 
68
 
69
+ Install the package via pip:
70
  ```bash
71
+ pip install "qai-hub-models[detr-resnet50-dc5]"
72
  ```
73
 
74
 
 
75
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
76
 
77
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
 
123
  Device : Samsung Galaxy S23 (13)
124
  Runtime : TFLITE
125
  Estimated inference time (ms) : 84.6
126
+ Estimated peak memory usage (MB): [0, 46]
127
  Total # Ops : 789
128
  Compute Unit(s) : NPU (789 ops)
129
  ```
 
150
  torch_model = Model.from_pretrained()
151
 
152
  # Device
153
+ device = hub.Device("Samsung Galaxy S24")
154
 
155
  # Trace model
156
  input_shape = torch_model.get_input_spec()
 
242
 
243
 
244
  ## License
245
+ * The license for the original implementation of DETR-ResNet50-DC5 can be found
246
+ [here](https://github.com/facebookresearch/detr/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