schuetoliver
/
handson-ml
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Compute the losses only when needed

main
Aurélien Geron 2021-03-02 17:26:41 +13:00
parent 6311ef8184
commit 935f26de1c
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
04_training_linear_models.ipynb
View File

@ -1521,10 +1521,10 @@
"for iteration in range(n_iterations):\n", "for iteration in range(n_iterations):\n",
" logits = X_train.dot(Theta)\n", " logits = X_train.dot(Theta)\n",
" Y_proba = softmax(logits)\n", " Y_proba = softmax(logits)\n",
" loss = -np.mean(np.sum(Y_train_one_hot * np.log(Y_proba + epsilon), axis=1))\n",
" error = Y_proba - Y_train_one_hot\n",
" if iteration % 500 == 0:\n", " if iteration % 500 == 0:\n",
" loss = -np.mean(np.sum(Y_train_one_hot * np.log(Y_proba + epsilon), axis=1))\n",
" print(iteration, loss)\n", " print(iteration, loss)\n",
" error = Y_proba - Y_train_one_hot\n",
" gradients = 1/m * X_train.T.dot(error)\n", " gradients = 1/m * X_train.T.dot(error)\n",
" Theta = Theta - eta * gradients" " Theta = Theta - eta * gradients"
] ]
@ -1590,12 +1590,12 @@
"for iteration in range(n_iterations):\n", "for iteration in range(n_iterations):\n",
" logits = X_train.dot(Theta)\n", " logits = X_train.dot(Theta)\n",
" Y_proba = softmax(logits)\n", " Y_proba = softmax(logits)\n",
" xentropy_loss = -np.mean(np.sum(Y_train_one_hot * np.log(Y_proba + epsilon), axis=1))\n",
" l2_loss = 1/2 * np.sum(np.square(Theta[1:]))\n",
" loss = xentropy_loss + alpha * l2_loss\n",
" error = Y_proba - Y_train_one_hot\n",
" if iteration % 500 == 0:\n", " if iteration % 500 == 0:\n",
" xentropy_loss = -np.mean(np.sum(Y_train_one_hot * np.log(Y_proba + epsilon), axis=1))\n",
" l2_loss = 1/2 * np.sum(np.square(Theta[1:]))\n",
" loss = xentropy_loss + alpha * l2_loss\n",
" print(iteration, loss)\n", " print(iteration, loss)\n",
" error = Y_proba - Y_train_one_hot\n",
" gradients = 1/m * X_train.T.dot(error) + np.r_[np.zeros([1, n_outputs]), alpha * Theta[1:]]\n", " gradients = 1/m * X_train.T.dot(error) + np.r_[np.zeros([1, n_outputs]), alpha * Theta[1:]]\n",
" Theta = Theta - eta * gradients" " Theta = Theta - eta * gradients"
] ]
@ -1793,7 +1793,7 @@
"name": "python", "name": "python",
"nbconvert_exporter": "python", "nbconvert_exporter": "python",
"pygments_lexer": "ipython3", "pygments_lexer": "ipython3",
"version": "3.7.8" "version": "3.7.9"
}, },
"nav_menu": {}, "nav_menu": {},
"toc": { "toc": {