引導式測驗 · 中等 等級
混合精準度訓練測驗
涵蓋 FP16 範圍和損失縮放、bfloat16、主權重、自動投射、Tensor Core 累積以及測量的硬體相關增益。
問題 1 的 8
In an FP16 training recipe that keeps an FP32 master weight, why retain it?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- In an FP16 training recipe that keeps an FP32 master weight, why retain it?
- 為了保持 16 位元可以四捨五入的微小權重更新的精度
- 使磁碟上的模型檔案更小
- 允許在沒有 GPU 的情況下進行訓練
- 加密模型參數
- FP16 訓練中損失縮放解決了什麼問題?
- 模型過度擬合數據
- 由於 FP16 的範圍很窄,梯度下溢到零
- GPU 磁碟空間不足
- 從磁碟載入資料緩慢
- bfloat16相對於FP16有什麼優勢?
- 它使用更少的總位數
- 它存儲文本而不是數字
- 它具有與 FP32 相同的寬指數範圍,減少了損失縮放的需要
- 它僅在 CPU 上運行
- For many supported Tensor Core matrix operations, how are partial sums commonly accumulated?
- In integer arithmetic with no rounding
- In text before converting to FP16
- In FP16 only, which prevents every numerical error
- In FP32, which reduces rounding error compared with low-precision accumulation
- What is a common benefit of using 16-bit values for supported training operations?
- They can reduce tensor memory and improve throughput, with gains depending on hardware and workload
- They guarantee higher model accuracy
- They remove the need to compute gradients
- They always double end-to-end training speed
- Which current PyTorch APIs are commonly used for automatic mixed-precision training?
- torch.cuda.amp only, which is the recommended new namespace
- torch.amp.autocast and, for FP16 gradient scaling, torch.amp.GradScaler
- torch.load and torch.save
- torch.nn.DataParallel and torch.manual_seed
- Why does bfloat16 often need less loss scaling than FP16?
- It has more mantissa bits than FP32
- It stores gradients as integers
- It has an exponent range like FP32, so very small gradients are less likely to underflow
- It never rounds values
- What does autocast do in automatic mixed precision?
- Converts every tensor and operation to FP16 unconditionally
- Replaces the optimizer with a different algorithm
- Guarantees identical bitwise results to FP32
- Selects operation dtypes so supported operations can use lower precision while others retain safer precision