Hugging Face's logo

qualcomm
/
DDRNet23-Slim

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
04cfc02
·
verified ·
1 Parent(s): 0c9a883

v0.30.5

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.30.5 for changelog.

Files changed (1) hide show
  1. README.md +33 -33
README.md CHANGED
@@ -37,35 +37,35 @@ More details on model performance across various devices, can be found
37
 
38
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
39
  |---|---|---|---|---|---|---|---|---|
40
- | DDRNet23-Slim | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 113.357 ms | 3 - 45 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
41
- | DDRNet23-Slim | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 96.253 ms | 24 - 34 MB | NPU | Use Export Script |
42
- | DDRNet23-Slim | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 49.446 ms | 2 - 53 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
43
- | DDRNet23-Slim | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 60.213 ms | 20 - 75 MB | NPU | Use Export Script |
44
- | DDRNet23-Slim | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 40.708 ms | 2 - 20 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
45
- | DDRNet23-Slim | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 31.825 ms | 24 - 27 MB | NPU | Use Export Script |
46
- | DDRNet23-Slim | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 49.015 ms | 2 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
47
- | DDRNet23-Slim | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 39.42 ms | 24 - 39 MB | NPU | Use Export Script |
48
- | DDRNet23-Slim | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 113.357 ms | 3 - 45 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
49
- | DDRNet23-Slim | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 96.253 ms | 24 - 34 MB | NPU | Use Export Script |
50
- | DDRNet23-Slim | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 40.579 ms | 2 - 23 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
51
- | DDRNet23-Slim | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 32.133 ms | 24 - 27 MB | NPU | Use Export Script |
52
- | DDRNet23-Slim | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 56.216 ms | 2 - 47 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
53
- | DDRNet23-Slim | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 43.951 ms | 24 - 42 MB | NPU | Use Export Script |
54
- | DDRNet23-Slim | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 40.686 ms | 2 - 27 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
55
- | DDRNet23-Slim | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 32.168 ms | 24 - 26 MB | NPU | Use Export Script |
56
- | DDRNet23-Slim | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 49.015 ms | 2 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
57
- | DDRNet23-Slim | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 39.42 ms | 24 - 39 MB | NPU | Use Export Script |
58
- | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 40.839 ms | 2 - 21 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
59
- | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 32.477 ms | 26 - 52 MB | NPU | Use Export Script |
60
- | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 42.108 ms | 26 - 80 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
61
- | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 26.962 ms | 2 - 52 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
62
- | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 23.0 ms | 24 - 84 MB | NPU | Use Export Script |
63
- | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 31.739 ms | 28 - 89 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
64
- | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 27.5 ms | 2 - 48 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
65
- | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 18.908 ms | 24 - 81 MB | NPU | Use Export Script |
66
- | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 17.318 ms | 27 - 72 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
67
- | DDRNet23-Slim | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 32.53 ms | 24 - 24 MB | NPU | Use Export Script |
68
- | DDRNet23-Slim | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 34.462 ms | 24 - 24 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
69
 
70
 
71
 
@@ -129,8 +129,8 @@ Profiling Results
129
  DDRNet23-Slim
130
  Device : cs_8275 (ANDROID 14)
131
  Runtime : TFLITE
132
- Estimated inference time (ms) : 113.4
133
- Estimated peak memory usage (MB): [3, 45]
134
  Total # Ops : 133
135
  Compute Unit(s) : npu (133 ops) gpu (0 ops) cpu (0 ops)
136
  ```
@@ -219,13 +219,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
219
  You can also run the demo on-device.
220
 
221
  ```bash
222
- python -m qai_hub_models.models.ddrnet23_slim.demo --on-device
223
  ```
224
 
225
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
226
  environment, please add the following to your cell (instead of the above).
227
  ```
228
- %run -m qai_hub_models.models.ddrnet23_slim.demo -- --on-device
229
  ```
230
 
231
 
 
37
 
38
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
39
  |---|---|---|---|---|---|---|---|---|
40
+ | DDRNet23-Slim | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 113.097 ms | 3 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
41
+ | DDRNet23-Slim | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 526.808 ms | 24 - 33 MB | NPU | Use Export Script |
42
+ | DDRNet23-Slim | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 48.795 ms | 2 - 58 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
43
+ | DDRNet23-Slim | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 58.99 ms | 24 - 82 MB | NPU | Use Export Script |
44
+ | DDRNet23-Slim | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 40.775 ms | 2 - 28 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
45
+ | DDRNet23-Slim | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 32.242 ms | 24 - 28 MB | NPU | Use Export Script |
46
+ | DDRNet23-Slim | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 49.192 ms | 2 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
47
+ | DDRNet23-Slim | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 40.218 ms | 24 - 37 MB | NPU | Use Export Script |
48
+ | DDRNet23-Slim | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 113.097 ms | 3 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
49
+ | DDRNet23-Slim | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 526.808 ms | 24 - 33 MB | NPU | Use Export Script |
50
+ | DDRNet23-Slim | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 40.744 ms | 2 - 33 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
51
+ | DDRNet23-Slim | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 32.456 ms | 24 - 27 MB | NPU | Use Export Script |
52
+ | DDRNet23-Slim | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 54.336 ms | 2 - 49 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
53
+ | DDRNet23-Slim | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 42.512 ms | 24 - 41 MB | NPU | Use Export Script |
54
+ | DDRNet23-Slim | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 40.871 ms | 2 - 17 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
55
+ | DDRNet23-Slim | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 32.282 ms | 3 - 5 MB | NPU | Use Export Script |
56
+ | DDRNet23-Slim | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 49.192 ms | 2 - 46 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
57
+ | DDRNet23-Slim | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 40.218 ms | 24 - 37 MB | NPU | Use Export Script |
58
+ | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 40.962 ms | 2 - 34 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
59
+ | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 32.75 ms | 24 - 63 MB | NPU | Use Export Script |
60
+ | DDRNet23-Slim | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 25.283 ms | 26 - 70 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
61
+ | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 27.102 ms | 1 - 52 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
62
+ | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 23.269 ms | 24 - 82 MB | NPU | Use Export Script |
63
+ | DDRNet23-Slim | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 16.261 ms | 28 - 86 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
64
+ | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 23.239 ms | 2 - 50 MB | NPU | [DDRNet23-Slim.tflite](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.tflite) |
65
+ | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 15.859 ms | 24 - 86 MB | NPU | Use Export Script |
66
+ | DDRNet23-Slim | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 12.687 ms | 30 - 84 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
67
+ | DDRNet23-Slim | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 32.797 ms | 24 - 24 MB | NPU | Use Export Script |
68
+ | DDRNet23-Slim | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 27.81 ms | 24 - 24 MB | NPU | [DDRNet23-Slim.onnx](https://huggingface.co/qualcomm/DDRNet23-Slim/blob/main/DDRNet23-Slim.onnx) |
69
 
70
 
71
 
 
129
  DDRNet23-Slim
130
  Device : cs_8275 (ANDROID 14)
131
  Runtime : TFLITE
132
+ Estimated inference time (ms) : 113.1
133
+ Estimated peak memory usage (MB): [3, 46]
134
  Total # Ops : 133
135
  Compute Unit(s) : npu (133 ops) gpu (0 ops) cpu (0 ops)
136
  ```
 
219
  You can also run the demo on-device.
220
 
221
  ```bash
222
+ python -m qai_hub_models.models.ddrnet23_slim.demo --eval-mode on-device
223
  ```
224
 
225
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
226
  environment, please add the following to your cell (instead of the above).
227
  ```
228
+ %run -m qai_hub_models.models.ddrnet23_slim.demo -- --eval-mode on-device
229
  ```
230
 
231