Hugging Face's logo

qualcomm
/
FFNet-122NS-LowRes

Image Segmentation
PyTorch
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
6a4bb53
·
verified ·
1 Parent(s): 0552e03

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +32 -32
README.md CHANGED
@@ -35,45 +35,44 @@ 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
- | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 12.232 ms | 0 - 61 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
39
- | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 14.335 ms | 6 - 27 MB | FP16 | NPU | [FFNet-122NS-LowRes.so](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.so) |
40
- | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 7.343 ms | 2 - 185 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
41
- | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 8.429 ms | 1 - 31 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
42
- | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 9.754 ms | 6 - 35 MB | FP16 | NPU | [FFNet-122NS-LowRes.so](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.so) |
43
- | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 5.934 ms | 6 - 58 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
44
- | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 6.911 ms | 1 - 30 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
45
- | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 10.348 ms | 6 - 31 MB | FP16 | NPU | Use Export Script |
46
- | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 4.447 ms | 6 - 41 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
47
- | FFNet-122NS-LowRes | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 12.208 ms | 0 - 40 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
48
- | FFNet-122NS-LowRes | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 10.758 ms | 6 - 9 MB | FP16 | NPU | Use Export Script |
49
- | FFNet-122NS-LowRes | SA7255P ADP | SA7255P | TFLITE | 215.507 ms | 0 - 23 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
50
- | FFNet-122NS-LowRes | SA7255P ADP | SA7255P | QNN | 211.423 ms | 6 - 16 MB | FP16 | NPU | Use Export Script |
51
- | FFNet-122NS-LowRes | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 12.295 ms | 1 - 20 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
52
- | FFNet-122NS-LowRes | SA8255 (Proxy) | SA8255P Proxy | QNN | 10.72 ms | 6 - 8 MB | FP16 | NPU | Use Export Script |
53
- | FFNet-122NS-LowRes | SA8295P ADP | SA8295P | TFLITE | 18.947 ms | 1 - 26 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
54
- | FFNet-122NS-LowRes | SA8295P ADP | SA8295P | QNN | 15.154 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
55
- | FFNet-122NS-LowRes | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 12.34 ms | 1 - 31 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
56
- | FFNet-122NS-LowRes | SA8650 (Proxy) | SA8650P Proxy | QNN | 10.873 ms | 6 - 9 MB | FP16 | NPU | Use Export Script |
57
- | FFNet-122NS-LowRes | SA8775P ADP | SA8775P | TFLITE | 18.929 ms | 1 - 25 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
58
- | FFNet-122NS-LowRes | SA8775P ADP | SA8775P | QNN | 17.168 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
59
- | FFNet-122NS-LowRes | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 16.271 ms | 0 - 30 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
60
- | FFNet-122NS-LowRes | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 18.095 ms | 6 - 31 MB | FP16 | NPU | Use Export Script |
61
- | FFNet-122NS-LowRes | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 11.142 ms | 6 - 6 MB | FP16 | NPU | Use Export Script |
62
- | FFNet-122NS-LowRes | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.131 ms | 59 - 59 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.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[ffnet_122ns_lowres]"
73
  ```
74
 
75
 
76
-
77
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
78
 
79
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -124,8 +123,8 @@ Profiling Results
124
  FFNet-122NS-LowRes
125
  Device : Samsung Galaxy S23 (13)
126
  Runtime : TFLITE
127
- Estimated inference time (ms) : 12.2
128
- Estimated peak memory usage (MB): [0, 61]
129
  Total # Ops : 218
130
  Compute Unit(s) : NPU (218 ops)
131
  ```
@@ -152,7 +151,7 @@ from qai_hub_models.models.ffnet_122ns_lowres import Model
152
  torch_model = Model.from_pretrained()
153
 
154
  # Device
155
- device = hub.Device("Samsung Galaxy S23")
156
 
157
  # Trace model
158
  input_shape = torch_model.get_input_spec()
@@ -244,7 +243,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
244
 
245
 
246
  ## License
247
- * The license for the original implementation of FFNet-122NS-LowRes can be found [here](https://github.com/Qualcomm-AI-research/FFNet/blob/master/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
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 12.146 ms | 0 - 38 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
39
+ | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 14.329 ms | 7 - 32 MB | FP16 | NPU | [FFNet-122NS-LowRes.so](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.so) |
40
+ | FFNet-122NS-LowRes | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 7.351 ms | 0 - 192 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
41
+ | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 8.4 ms | 1 - 31 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
42
+ | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 9.739 ms | 6 - 36 MB | FP16 | NPU | [FFNet-122NS-LowRes.so](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.so) |
43
+ | FFNet-122NS-LowRes | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 5.937 ms | 6 - 56 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
44
+ | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 6.92 ms | 0 - 28 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
45
+ | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 10.337 ms | 6 - 32 MB | FP16 | NPU | Use Export Script |
46
+ | FFNet-122NS-LowRes | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 4.619 ms | 8 - 41 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
47
+ | FFNet-122NS-LowRes | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 12.227 ms | 0 - 61 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
48
+ | FFNet-122NS-LowRes | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 10.787 ms | 6 - 9 MB | FP16 | NPU | Use Export Script |
49
+ | FFNet-122NS-LowRes | SA7255P ADP | SA7255P | TFLITE | 215.361 ms | 0 - 23 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
50
+ | FFNet-122NS-LowRes | SA7255P ADP | SA7255P | QNN | 211.532 ms | 4 - 14 MB | FP16 | NPU | Use Export Script |
51
+ | FFNet-122NS-LowRes | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 12.215 ms | 0 - 51 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
52
+ | FFNet-122NS-LowRes | SA8255 (Proxy) | SA8255P Proxy | QNN | 10.899 ms | 6 - 8 MB | FP16 | NPU | Use Export Script |
53
+ | FFNet-122NS-LowRes | SA8295P ADP | SA8295P | TFLITE | 18.931 ms | 1 - 26 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
54
+ | FFNet-122NS-LowRes | SA8295P ADP | SA8295P | QNN | 15.399 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
55
+ | FFNet-122NS-LowRes | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 12.311 ms | 0 - 31 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
56
+ | FFNet-122NS-LowRes | SA8650 (Proxy) | SA8650P Proxy | QNN | 10.753 ms | 8 - 10 MB | FP16 | NPU | Use Export Script |
57
+ | FFNet-122NS-LowRes | SA8775P ADP | SA8775P | TFLITE | 18.935 ms | 1 - 24 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
58
+ | FFNet-122NS-LowRes | SA8775P ADP | SA8775P | QNN | 17.147 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
59
+ | FFNet-122NS-LowRes | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 16.375 ms | 1 - 28 MB | FP16 | NPU | [FFNet-122NS-LowRes.tflite](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.tflite) |
60
+ | FFNet-122NS-LowRes | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 18.05 ms | 6 - 29 MB | FP16 | NPU | Use Export Script |
61
+ | FFNet-122NS-LowRes | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 11.205 ms | 6 - 6 MB | FP16 | NPU | Use Export Script |
62
+ | FFNet-122NS-LowRes | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.085 ms | 58 - 58 MB | FP16 | NPU | [FFNet-122NS-LowRes.onnx](https://huggingface.co/qualcomm/FFNet-122NS-LowRes/blob/main/FFNet-122NS-LowRes.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
 
69
 
70
+ Install the package via pip:
71
  ```bash
72
+ pip install "qai-hub-models[ffnet-122ns-lowres]"
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
 
123
  FFNet-122NS-LowRes
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
+ Estimated inference time (ms) : 12.1
127
+ Estimated peak memory usage (MB): [0, 38]
128
  Total # Ops : 218
129
  Compute Unit(s) : NPU (218 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 FFNet-122NS-LowRes can be found
247
+ [here](https://github.com/Qualcomm-AI-research/FFNet/blob/master/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