cs231n_assignment1_two_layer_net
2018-03-29 22:27
393 查看
作业内容:实现一个两层的神经网络,包括FP和BP,在CIFAR-10上测试
计算score,score function用的是ReLU:max(0,W*X):
计算loss,用的是softmax+L2正则化:
用BP计算梯度:
利用SGD训练模型:
先是挑选mini batch:
更新:
最后精度达到29%。
很低,所以模型需要debug ,可以画出plot the loss function and the accuracies on the training and validation sets during optimization;visualize the weights that were learned in the first layer of the network。
有几点可以考虑:
1.hidden layer size,
2.learning rate,
3. numer of training epochs,
4.regularization strength.
计算score,score function用的是ReLU:max(0,W*X):
计算loss,用的是softmax+L2正则化:
用BP计算梯度:
利用SGD训练模型:
先是挑选mini batch:
更新:
最后精度达到29%。
很低,所以模型需要debug ,可以画出plot the loss function and the accuracies on the training and validation sets during optimization;visualize the weights that were learned in the first layer of the network。
有几点可以考虑:
1.hidden layer size,
2.learning rate,
3. numer of training epochs,
4.regularization strength.
相关文章推荐
- cs231n-assignment1-SVM/Softmax/two-layer-nets梯度求解
- cs231n作业1--two_layer_net
- cs231n:assignment1——Q4: Two-Layer Neural Network
- [CS231n@Stanford] Assignment1-Q4 (python) Two layer neural network实现
- CS231n课程作业(二) Multi-Layer Net、BN、Dropout、Optimization
- sybase连接问题:ct_connect(): network packet layer: internal net library error: Net-Lib protocol driver call to connect two endpoints
- cs231n assignment(1.4):two_layer_net
- cs231n:assignment2——python文件:layer_utils.py
- cs231n - assignment1 - neural net 梯度推导
- cs231n:assignment2——python文件:fc_net.py
- CS231n课程作业(一) Two-layer Neural Network
- two_layer_net 代码学习笔记
- cs231n assignment1--two_layer_net
- CS231_A1:Two-layer Neural Net
- Ontology Building Basedon Two-layer Ontology Model
- CS231n - CNN for Visual Recognition Assignment1 ---- KNN
- cs231n - assignment1 - softmax 梯度推导
- Caffe傻瓜系列(5):Blob,Layer and Net以及对应配置文件的编写
- H264参数结构二:网络提取层NAL (Net Abstraction Layer) & 视频编码层VCL (Video Coding Layer)
- H264参数结构二:网络提取层NAL (Net Abstraction Layer) & 视频编码层VCL (Video Coding Layer) .