知识卡片

结构图卡 · 6694.i · 生成模型:GPT-6

训练循环包含四种不同动作

理解等级:中级

结构图:

flowchart TD
 A[取同序输入标签及可选权重] --> B[清理无需累积的旧梯度]
 B --> C[前向计算预测]
 C --> D[按定义计算损失]
 D --> E[反向链式求梯度]
 E --> F[优化器更新所管理参数]
 F --> G[记录固定时点训练及验证指标]
 G --> H{是否继续训练}
 H -->|是| A

采样、求导、更新及评价各有职责,选择SGD类不代表已写出小批采样。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      >>> for epoch in pbar_epochs: ... optimizer.zero_grad() #1 ... outputs = model(X_train) ... loss_train = loss_func_train(outputs, y_train) #2 ... loss_train.backward() #3 ... optimizer.step() #4 Epoch:: 100%|================================| 200/200 [00:04<00:00, 42.84it/s] ↪ 96.26it/s] #1 Step 1: Sets the stored gradients to 0 #2 Step 2: Calculates the training loss #3 Step 3: Computes the gradient on the training set #4 Step 4: Updates weights and biases with

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第05章《神经网络基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-5.html) - 结论依据:摘录中关于训练循环包含四种不同动作的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:>>> for epoch in pbar_epochs: ... optimizer.zero_grad() #1 ... outputs = model(X_train) ... loss_train = loss_func_train(outputs, y_train) #2 ... loss_train.backward() #3 ... optimizer.step() #4 Epoch:: 100%|================================| 200/200 [00:04<00:00, 42.84it/s] ↪ 96.26it/s] #1 Step 1: Sets the stored gradients to 0 #2 Step 2: Calculates the training loss #3 Step 3: Computes the gradient on the training set #4 Step 4: Updates weights and biases with