Hugging Face's logo

qualcomm
/
DeepLabV3-ResNet50

Image Segmentation
PyTorch
LiteRT
android
Model card Files Community
qaihm-bot commited on
Commit
8586603
·
verified ·
1 Parent(s): 16e6622

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +16 -15
README.md CHANGED
@@ -35,24 +35,24 @@ 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
- | DeepLabV3-ResNet50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 331.647 ms | 2 - 206 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
39
- | DeepLabV3-ResNet50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 211.881 ms | 23 - 54 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
40
- | DeepLabV3-ResNet50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 221.287 ms | 22 - 42 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
41
- | DeepLabV3-ResNet50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 293.032 ms | 0 - 194 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
42
- | DeepLabV3-ResNet50 | SA7255P ADP | SA7255P | TFLITE | 1187.038 ms | 21 - 41 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
43
- | DeepLabV3-ResNet50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 291.902 ms | 0 - 210 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
44
- | DeepLabV3-ResNet50 | SA8295P ADP | SA8295P | TFLITE | 278.809 ms | 23 - 45 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
45
- | DeepLabV3-ResNet50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 292.163 ms | 0 - 180 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
46
- | DeepLabV3-ResNet50 | SA8775P ADP | SA8775P | TFLITE | 592.15 ms | 23 - 43 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
47
- | DeepLabV3-ResNet50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 773.346 ms | 22 - 54 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
48
 
49
 
50
 
51
 
52
  ## Installation
53
 
54
- This model can be installed as a Python package via pip.
55
 
 
56
  ```bash
57
  pip install qai-hub-models
58
  ```
@@ -108,8 +108,8 @@ Profiling Results
108
  DeepLabV3-ResNet50
109
  Device : Samsung Galaxy S23 (13)
110
  Runtime : TFLITE
111
- Estimated inference time (ms) : 331.6
112
- Estimated peak memory usage (MB): [2, 206]
113
  Total # Ops : 100
114
  Compute Unit(s) : GPU (98 ops) CPU (2 ops)
115
  ```
@@ -136,7 +136,7 @@ from qai_hub_models.models.deeplabv3_resnet50 import Model
136
  torch_model = Model.from_pretrained()
137
 
138
  # Device
139
- device = hub.Device("Samsung Galaxy S23")
140
 
141
  # Trace model
142
  input_shape = torch_model.get_input_spec()
@@ -228,7 +228,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
228
 
229
 
230
  ## License
231
- * The license for the original implementation of DeepLabV3-ResNet50 can be found [here](https://github.com/pytorch/vision/blob/main/LICENSE).
 
232
  * 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)
233
 
234
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | DeepLabV3-ResNet50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 292.941 ms | 0 - 181 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
39
+ | DeepLabV3-ResNet50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 214.81 ms | 22 - 51 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
40
+ | DeepLabV3-ResNet50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 222.393 ms | 22 - 42 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
41
+ | DeepLabV3-ResNet50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 293.817 ms | 0 - 200 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
42
+ | DeepLabV3-ResNet50 | SA7255P ADP | SA7255P | TFLITE | 2148.816 ms | 21 - 41 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
43
+ | DeepLabV3-ResNet50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 294.006 ms | 0 - 190 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
44
+ | DeepLabV3-ResNet50 | SA8295P ADP | SA8295P | TFLITE | 283.104 ms | 23 - 45 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
45
+ | DeepLabV3-ResNet50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 292.801 ms | 0 - 233 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
46
+ | DeepLabV3-ResNet50 | SA8775P ADP | SA8775P | TFLITE | 593.557 ms | 23 - 39 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
47
+ | DeepLabV3-ResNet50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 749.786 ms | 22 - 53 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
48
 
49
 
50
 
51
 
52
  ## Installation
53
 
 
54
 
55
+ Install the package via pip:
56
  ```bash
57
  pip install qai-hub-models
58
  ```
 
108
  DeepLabV3-ResNet50
109
  Device : Samsung Galaxy S23 (13)
110
  Runtime : TFLITE
111
+ Estimated inference time (ms) : 292.9
112
+ Estimated peak memory usage (MB): [0, 181]
113
  Total # Ops : 100
114
  Compute Unit(s) : GPU (98 ops) CPU (2 ops)
115
  ```
 
136
  torch_model = Model.from_pretrained()
137
 
138
  # Device
139
+ device = hub.Device("Samsung Galaxy S24")
140
 
141
  # Trace model
142
  input_shape = torch_model.get_input_spec()
 
228
 
229
 
230
  ## License
231
+ * The license for the original implementation of DeepLabV3-ResNet50 can be found
232
+ [here](https://github.com/pytorch/vision/blob/main/LICENSE).
233
  * 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)
234
 
235