Hugging Face's logo

qualcomm
/
EfficientViT-l2-cls

Image Classification
PyTorch
TF Lite
ONNX
backbone
real_time
android
Model card Files Community
qaihm-bot commited on
Commit
4213b15
·
verified ·
1 Parent(s): 2e2f8ee

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +21 -21
README.md CHANGED
@@ -36,35 +36,34 @@ More details on model performance across various devices, can be found
36
 
37
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
- | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 22.227 ms | 0 - 27 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
40
- | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 21.864 ms | 1 - 28 MB | FP16 | NPU | [EfficientViT-l2-cls.so](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.so) |
41
- | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 15.851 ms | 0 - 297 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
42
- | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 15.867 ms | 0 - 89 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
43
- | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 15.564 ms | 1 - 90 MB | FP16 | NPU | [EfficientViT-l2-cls.so](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.so) |
44
- | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 10.89 ms | 0 - 104 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
45
- | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 17.057 ms | 0 - 90 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
46
- | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 16.828 ms | 1 - 91 MB | FP16 | NPU | Use Export Script |
47
- | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 8.305 ms | 0 - 99 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
48
- | EfficientViT-l2-cls | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 22.199 ms | 0 - 27 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
49
- | EfficientViT-l2-cls | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 15.363 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
50
- | EfficientViT-l2-cls | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 20.313 ms | 0 - 66 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
51
- | EfficientViT-l2-cls | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 20.129 ms | 1 - 68 MB | FP16 | NPU | Use Export Script |
52
- | EfficientViT-l2-cls | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 16.661 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
53
- | EfficientViT-l2-cls | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 17.304 ms | 132 - 132 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
54
 
55
 
56
 
57
 
58
  ## Installation
59
 
60
- This model can be installed as a Python package via pip.
61
 
 
62
  ```bash
63
- pip install "qai-hub-models[efficientvit_l2_cls]"
64
  ```
65
 
66
 
67
-
68
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
69
 
70
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -115,7 +114,7 @@ Profiling Results
115
  EfficientViT-l2-cls
116
  Device : Samsung Galaxy S23 (13)
117
  Runtime : TFLITE
118
- Estimated inference time (ms) : 22.2
119
  Estimated peak memory usage (MB): [0, 27]
120
  Total # Ops : 675
121
  Compute Unit(s) : NPU (675 ops)
@@ -143,7 +142,7 @@ from qai_hub_models.models.efficientvit_l2_cls import Model
143
  torch_model = Model.from_pretrained()
144
 
145
  # Device
146
- device = hub.Device("Samsung Galaxy S23")
147
 
148
  # Trace model
149
  input_shape = torch_model.get_input_spec()
@@ -235,7 +234,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
235
 
236
 
237
  ## License
238
- * The license for the original implementation of EfficientViT-l2-cls can be found [here](https://github.com/CVHub520/efficientvit/blob/main/LICENSE).
 
239
  * 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)
240
 
241
 
 
36
 
37
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
+ | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 22.091 ms | 0 - 27 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
40
+ | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 21.879 ms | 0 - 27 MB | FP16 | NPU | [EfficientViT-l2-cls.so](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.so) |
41
+ | EfficientViT-l2-cls | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 15.934 ms | 0 - 291 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
42
+ | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 15.864 ms | 0 - 89 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
43
+ | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 15.585 ms | 1 - 87 MB | FP16 | NPU | [EfficientViT-l2-cls.so](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.so) |
44
+ | EfficientViT-l2-cls | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 10.995 ms | 1 - 104 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
45
+ | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 17.034 ms | 0 - 91 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
46
+ | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 13.817 ms | 1 - 91 MB | FP16 | NPU | Use Export Script |
47
+ | EfficientViT-l2-cls | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 9.708 ms | 0 - 99 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
48
+ | EfficientViT-l2-cls | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 22.166 ms | 0 - 28 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
49
+ | EfficientViT-l2-cls | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 15.234 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
50
+ | EfficientViT-l2-cls | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 20.003 ms | 0 - 62 MB | FP16 | NPU | [EfficientViT-l2-cls.tflite](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.tflite) |
51
+ | EfficientViT-l2-cls | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 19.856 ms | 0 - 68 MB | FP16 | NPU | Use Export Script |
52
+ | EfficientViT-l2-cls | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 16.127 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
53
+ | EfficientViT-l2-cls | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 17.206 ms | 131 - 131 MB | FP16 | NPU | [EfficientViT-l2-cls.onnx](https://huggingface.co/qualcomm/EfficientViT-l2-cls/blob/main/EfficientViT-l2-cls.onnx) |
54
 
55
 
56
 
57
 
58
  ## Installation
59
 
 
60
 
61
+ Install the package via pip:
62
  ```bash
63
+ pip install "qai-hub-models[efficientvit-l2-cls]"
64
  ```
65
 
66
 
 
67
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
68
 
69
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
 
114
  EfficientViT-l2-cls
115
  Device : Samsung Galaxy S23 (13)
116
  Runtime : TFLITE
117
+ Estimated inference time (ms) : 22.1
118
  Estimated peak memory usage (MB): [0, 27]
119
  Total # Ops : 675
120
  Compute Unit(s) : NPU (675 ops)
 
142
  torch_model = Model.from_pretrained()
143
 
144
  # Device
145
+ device = hub.Device("Samsung Galaxy S24")
146
 
147
  # Trace model
148
  input_shape = torch_model.get_input_spec()
 
234
 
235
 
236
  ## License
237
+ * The license for the original implementation of EfficientViT-l2-cls can be found
238
+ [here](https://github.com/CVHub520/efficientvit/blob/main/LICENSE).
239
  * 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)
240
 
241