本项目将 LangGPT/facial_emotions_detection 模型适配到昇腾 NPU (Ascend910B) 上运行。
| 项目 | 版本/信息 |
|---|---|
| NPU 驱动 | npu-smi 25.5.2 |
| NPU 设备 | Ascend910_9362 |
| Python | 3.11.14 |
| PyTorch | (环境自带) |
| torch_npu | (环境自带) |
| transformers | 4.50.0 |
NPU 健康状态:OK
pip install -r requirements.txt
python inference.py推理结果 (NPU Top-5):
完整推理日志 (logs/inference.log):
=== LangGPT/facial_emotions_detection NPU Inference ===
Model: LangGPT/facial_emotions_detection
Loader type: transformers
Input shape: torch.Size([1, 3, 224, 224])
Output shape: torch.Size([1, 7])
Num classes: 7
NPU Top-5 Predictions:
Top-1: neutral (0.184739)
Top-2: sad (0.169405)
Top-3: happy (0.152713)
Top-4: disgust (0.147103)
Top-5: fear (0.130241)
All class probabilities:
sad: 0.169405
disgust: 0.147103
angry: 0.126599
neutral: 0.184739
fear: 0.130241
surprise: 0.089200
happy: 0.152713对单张测试图片进行 CPU 与 NPU 一致性验证:
| 指标 | 数值 |
|---|---|
| max_abs_error | 0.000305 |
| mean_abs_error | 0.000171 |
| relative_error | 0.1234% |
| cosine_similarity | 0.999985 |
| threshold | 1.0% |
| 结果 | PASS |
| 指标 | 数值 |
|---|---|
| avg latency | 6.1507 ms |
| min latency | 6.1041 ms |
| max latency | 6.2148 ms |
| p50 latency | 6.1746 ms |
| p90 latency | 6.2148 ms |
| p95 latency | 6.2148 ms |
| images/sec | 162.58 |
测试配置: warmup 2 次 + 正式 10 次,单卡 NPU。
本项目包含单图 smoke consistency 验证,非官方完整验证集评测。详细指标见第 4 节。
见 screenshots/self_verification.png。
| 日志 | 说明 |
|---|---|
logs/inference.log | NPU 推理输出 |
logs/accuracy.log | CPU-NPU 精度一致性 |
logs/benchmark.log | NPU 性能基准测试 |
snapshot_download 下载,不提交到仓库。local_files_only=True 避免 HuggingFace 自动下载。#NPU #Ascend #Ascend910 #ViT #ImageClassification #FacialEmotion