Hugging Face's logo

qualcomm
/
ConvNext-Base

Image Classification
PyTorch
LiteRT
bu_auto
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
d5895bd
·
verified ·
1 Parent(s): 0d116a2

v0.30.5

Browse files

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

Files changed (2) hide show
  1. ConvNext-Base_w8a16.onnx +2 -2
  2. README.md +36 -33
ConvNext-Base_w8a16.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:70839ab7aab1c72e34239f7967e369532403b9a0e9d5e250a89bbfc26228fe57
3
- size 355196720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49589d49c943b3ba4a2777f4e9cc214cdb5bb472716b4253f53bffbb05096aec
3
+ size 355235397
README.md CHANGED
@@ -34,35 +34,38 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | ConvNext-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 41.798 ms | 0 - 258 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
38
- | ConvNext-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 41.66 ms | 0 - 9 MB | NPU | Use Export Script |
39
- | ConvNext-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 19.208 ms | 0 - 272 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
40
- | ConvNext-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 23.154 ms | 0 - 275 MB | NPU | Use Export Script |
41
- | ConvNext-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.537 ms | 0 - 19 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
42
- | ConvNext-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 8.067 ms | 1 - 4 MB | NPU | Use Export Script |
43
- | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.52 ms | 0 - 12 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
44
- | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 8.047 ms | 0 - 18 MB | NPU | Use Export Script |
45
- | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 7.866 ms | 0 - 416 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
46
- | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.561 ms | 0 - 269 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
47
- | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 6.147 ms | 1 - 272 MB | NPU | Use Export Script |
48
- | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.082 ms | 1 - 283 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
49
- | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 5.041 ms | 0 - 262 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
50
- | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 4.54 ms | 1 - 268 MB | NPU | Use Export Script |
51
- | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.677 ms | 0 - 267 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
52
- | ConvNext-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 8.17 ms | 1 - 1 MB | NPU | Use Export Script |
53
- | ConvNext-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.871 ms | 176 - 176 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
54
- | ConvNext-Base | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 14.487 ms | 0 - 10 MB | NPU | Use Export Script |
55
- | ConvNext-Base | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 10.081 ms | 0 - 130 MB | NPU | Use Export Script |
56
- | ConvNext-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 5.914 ms | 0 - 3 MB | NPU | Use Export Script |
57
- | ConvNext-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 39.288 ms | 0 - 14 MB | NPU | Use Export Script |
58
- | ConvNext-Base | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 5.9 ms | 0 - 30 MB | NPU | Use Export Script |
59
- | ConvNext-Base | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 244.602 ms | 575 - 965 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
60
- | ConvNext-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 4.186 ms | 0 - 128 MB | NPU | Use Export Script |
61
- | ConvNext-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 182.381 ms | 685 - 1282 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
62
- | ConvNext-Base | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 3.087 ms | 0 - 122 MB | NPU | Use Export Script |
63
- | ConvNext-Base | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 179.182 ms | 673 - 1239 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
64
- | ConvNext-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 6.284 ms | 0 - 0 MB | NPU | Use Export Script |
65
- | ConvNext-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 224.64 ms | 926 - 926 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
 
 
 
66
 
67
 
68
 
@@ -126,8 +129,8 @@ Profiling Results
126
  ConvNext-Base
127
  Device : cs_8275 (ANDROID 14)
128
  Runtime : TFLITE
129
- Estimated inference time (ms) : 41.8
130
- Estimated peak memory usage (MB): [0, 258]
131
  Total # Ops : 598
132
  Compute Unit(s) : npu (598 ops) gpu (0 ops) cpu (0 ops)
133
  ```
@@ -216,13 +219,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
216
  You can also run the demo on-device.
217
 
218
  ```bash
219
- python -m qai_hub_models.models.convnext_base.demo --on-device
220
  ```
221
 
222
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
223
  environment, please add the following to your cell (instead of the above).
224
  ```
225
- %run -m qai_hub_models.models.convnext_base.demo -- --on-device
226
  ```
227
 
228
 
 
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | ConvNext-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 41.646 ms | 0 - 283 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
38
+ | ConvNext-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 302.999 ms | 1 - 10 MB | NPU | Use Export Script |
39
+ | ConvNext-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 18.275 ms | 0 - 296 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
40
+ | ConvNext-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 21.029 ms | 1 - 298 MB | NPU | Use Export Script |
41
+ | ConvNext-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.467 ms | 0 - 18 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
42
+ | ConvNext-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 7.731 ms | 1 - 3 MB | NPU | Use Export Script |
43
+ | ConvNext-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.43 ms | 0 - 283 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
44
+ | ConvNext-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 11.633 ms | 1 - 10 MB | NPU | Use Export Script |
45
+ | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.491 ms | 0 - 32 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
46
+ | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 7.782 ms | 0 - 27 MB | NPU | Use Export Script |
47
+ | ConvNext-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 7.667 ms | 0 - 399 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
48
+ | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.573 ms | 0 - 295 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
49
+ | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 5.678 ms | 1 - 291 MB | NPU | Use Export Script |
50
+ | ConvNext-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 5.535 ms | 1 - 299 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
51
+ | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 4.163 ms | 0 - 283 MB | NPU | [ConvNext-Base.tflite](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.tflite) |
52
+ | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 5.065 ms | 1 - 286 MB | NPU | Use Export Script |
53
+ | ConvNext-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.985 ms | 1 - 284 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
54
+ | ConvNext-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 8.114 ms | 1 - 1 MB | NPU | Use Export Script |
55
+ | ConvNext-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.883 ms | 176 - 176 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base.onnx) |
56
+ | ConvNext-Base | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 13.77 ms | 0 - 10 MB | NPU | Use Export Script |
57
+ | ConvNext-Base | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 8.739 ms | 0 - 136 MB | NPU | Use Export Script |
58
+ | ConvNext-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 5.574 ms | 0 - 3 MB | NPU | Use Export Script |
59
+ | ConvNext-Base | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 5.84 ms | 0 - 10 MB | NPU | Use Export Script |
60
+ | ConvNext-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 32.998 ms | 0 - 12 MB | NPU | Use Export Script |
61
+ | ConvNext-Base | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 5.543 ms | 0 - 30 MB | NPU | Use Export Script |
62
+ | ConvNext-Base | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 227.755 ms | 446 - 884 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
63
+ | ConvNext-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.893 ms | 0 - 135 MB | NPU | Use Export Script |
64
+ | ConvNext-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 201.925 ms | 645 - 1429 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
65
+ | ConvNext-Base | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 3.552 ms | 0 - 127 MB | NPU | Use Export Script |
66
+ | ConvNext-Base | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 148.722 ms | 649 - 1340 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
67
+ | ConvNext-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.897 ms | 0 - 0 MB | NPU | Use Export Script |
68
+ | ConvNext-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 212.316 ms | 920 - 920 MB | NPU | [ConvNext-Base.onnx](https://huggingface.co/qualcomm/ConvNext-Base/blob/main/ConvNext-Base_w8a16.onnx) |
69
 
70
 
71
 
 
129
  ConvNext-Base
130
  Device : cs_8275 (ANDROID 14)
131
  Runtime : TFLITE
132
+ Estimated inference time (ms) : 41.6
133
+ Estimated peak memory usage (MB): [0, 283]
134
  Total # Ops : 598
135
  Compute Unit(s) : npu (598 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.convnext_base.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.convnext_base.demo -- --eval-mode on-device
229
  ```
230
 
231