diff --git a/Rapport.ipynb b/Rapport.ipynb index b2f43f276bada618dec4a32141565447d25e6851..74eb61d78ec26abae81375d9628b386c5aa5789c 100644 --- a/Rapport.ipynb +++ b/Rapport.ipynb @@ -785,7 +785,7 @@ "metadata": {}, "outputs": [], "source": [ - "def run_mlp_cross_entropy_training(data_train, labels_train, data_test, labels_test,d_h,learning_rate,num_epoch):\n", + "def run_mlp_training(data_train, labels_train, data_test, labels_test,d_h,learning_rate,num_epoch):\n", " d_in = data_train.shape[1] # input dimension\n", " d_out = max(labels_train) # output dimension (number of neurons of the output layer)\n", "\n", @@ -808,7 +808,7 @@ }, { "cell_type": "code", - "execution_count": 151, + "execution_count": 153, "metadata": {}, "outputs": [ { @@ -823,7 +823,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "<Figure size 1000x400 with 2 Axes>" ] @@ -835,12 +835,12 @@ "source": [ "if __name__ == \"__main__\":\n", " d, l = read_cifar_batch(\"data/cifar-10-batches-py/data_batch_1\")\n", - " num_epoch=100\n", + " num_epoch=1000\n", " dh=64\n", - " train_param=0.1\n", + " learning_rate=0.1\n", " d_train, l_train, d_test, l_test = split_dataset(d, l, 0.9)\n", "\n", - " train_accuracy,loss,test_accuracy=run_mlp_cross_entropy_training(d_train, l_train, d_test, l_test,dh,train_param,num_epoch)\n", + " train_accuracy,loss,test_accuracy=run_mlp_training(d_train, l_train, d_test, l_test,dh,learning_rate,num_epoch)\n", " fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 4))\n", "\n", " ax1.plot(range(num_epoch),loss,label=\"cross-entropy\")\n", diff --git a/mlp.py b/mlp.py index 6ba4db070224ce0f75c2c21b1fff2e7397a713e9..b2029962519d4c4428b3a0ab4d9b66048fc050c4 100644 --- a/mlp.py +++ b/mlp.py @@ -4,26 +4,28 @@ from read_cifar import read_cifar_batch, split_dataset import matplotlib.pyplot as plt def learning_methode(k,dk,learning_rate): - k=k-learning_rate*dk - return(k) + k2=k-learning_rate*dk + return(k2) def learn_once_mse(w1,b1,w2,b2,data,targets,learning_rate): - a0 = data - z1 = np.matmul(a0, w1) + b1 - a1 = 1 / (1 + np.exp(-z1)) - z2 = np.matmul(a1, w2) + b2 - a2 = 1 / (1 + np.exp(-z2)) - predictions = a2 + a0 = data # the data are the input of the first layer + z1 = np.matmul(a0, w1) + b1 # input of the hidden layer + a1 = 1 / (1 + np.exp(-z1)) # output of the hidden layer (sigmoid activation function) + z2 = np.matmul(a1, w2) + b2 # input of the output layer + a2 = 1 / (1 + np.exp(-z2)) # output of the output layer (sigmoid activation function) + predictions = a2 # the predicted values are the outputs of the output layer - dc_da2=(2/data.shape[0])*(a2-targets) - + # Compute loss (MSE) + loss = np.mean(np.square(predictions - targets)) + + dc_da2=(2/data.shape[0])*(predictions-targets) dc_dz2=dc_da2*(a2*(1-a2)) - dc_dw2=np.matmul(np.transpose(a1), dc_dz2) - dc_db2=np.matmul(np.ones((1,dc_dz2.shape[0])),dc_dz2) - dc_da1=np.matmul(dc_dz2,np.transpose(w2)) + dc_dw2=np.dot(np.transpose(a1), dc_dz2) + dc_db2=np.dot(np.ones((1,dc_dz2.shape[0])),dc_dz2) + dc_da1=np.dot(dc_dz2,np.transpose(w2)) dc_dz1=dc_da1*(a1*(1-a1)) - dc_dw1=np.matmul(np.transpose(a0), dc_dz1) - dc_db1=np.matmul(np.ones((1,dc_dz1.shape[0])),dc_dz1) + dc_dw1=np.dot(np.transpose(a0), dc_dz1) + dc_db1=np.dot(np.ones((1,dc_dz1.shape[0])),dc_dz1) w1=learning_methode(w1,dc_dw1,learning_rate) b1=learning_methode(b1,dc_db1,learning_rate) @@ -31,7 +33,7 @@ def learn_once_mse(w1,b1,w2,b2,data,targets,learning_rate): b2=learning_methode(b2,dc_db2,learning_rate) # Compute loss (MSE) - loss = np.mean(np.square(predictions - targets)) + return(w1,b1,w2,b2,loss) def one_hot(label): @@ -43,50 +45,42 @@ def one_hot(label): def softmax(y): y=np.exp(y) - v=np.sum(y,axis=1) - return(y / v[:, np.newaxis]) + v=np.sum(y, axis=1, keepdims=True) + return(y / v) def learn_once_cross_entropy(w1,b1,w2,b2,data,labels_train,learning_rate): targets = one_hot(labels_train) - targets=targets+1e-15 a0 = data - z1 = np.matmul(a0, w1) + b1 + z1 = np.dot(a0, w1) + b1 a1 = 1 / (1 + np.exp(-z1)) - z2 = np.matmul(a1, w2) + b2 - a2 = 1 / (1 + np.exp(-z2)) - softa2=softmax(a2) - # predictions = softa2 + z2 = np.dot(a1, w2) + b2 + softa2=softmax(z2) predictions=softa2 - # dc_softmax=-(targets/softa2)+((1-targets)/(1-softa2)) - # dc_a2=dc_softmax*(softa2*(1-softa2)) + loss=np.mean(np.sum(-targets*np.log(predictions),axis=1)) - # dc_dz2=dc_a2*(a2*(1-a2)) dc_dz2=predictions-targets - dc_dw2=np.matmul(np.transpose(a1), dc_dz2) - dc_db2=np.matmul(np.ones((1,dc_dz2.shape[0])),dc_dz2) - dc_da1=np.matmul(dc_dz2,np.transpose(w2)) - dc_dz1=dc_da1*(a1*(1-a1)) - dc_dw1=np.matmul(np.transpose(a0), dc_dz1) - dc_db1=np.matmul(np.ones((1,dc_dz1.shape[0])),dc_dz1) + dc_dw2=np.dot(np.transpose(a1), dc_dz2)/data.shape[0] + dc_db2=np.mean(dc_dz2, axis=0, keepdims=True) + + dc_da1=np.dot(dc_dz2,np.transpose(w2)) + dc_dz1=dc_da1*(1-a1)*a1 + dc_dw1=np.dot(np.transpose(a0), dc_dz1)/data.shape[0] + dc_db1=np.mean(dc_dz1, axis=0, keepdims=True) w1=learning_methode(w1,dc_dw1,learning_rate) b1=learning_methode(b1,dc_db1,learning_rate) w2=learning_methode(w2,dc_dw2,learning_rate) b2=learning_methode(b2,dc_db2,learning_rate) - - # binary cross-entropy loss - loss = np.mean(targets*np.log(predictions)-(1-targets)*np.log(1-predictions)) return(w1,b1,w2,b2,loss) def accuracy(w1,b1,w2,b2,data,labels): a0 = data - z1 = np.matmul(a0, w1) + b1 + z1 = np.dot(a0, w1) + b1 a1 = 1 / (1 + np.exp(-z1)) - z2 = np.matmul(a1, w2) + b2 - a2 = 1 / (1 + np.exp(-z2)) - softa2=softmax(a2) + z2 = np.dot(a1, w2) + b2 + softa2=softmax(z2) predictions = softa2 prediction_2 = np.empty(predictions.shape[0], dtype=int) for i, ligne in enumerate(predictions): @@ -95,40 +89,31 @@ def accuracy(w1,b1,w2,b2,data,labels): nombre_indices = len(indices_egalite) return(nombre_indices/len(labels)) -def train_mlp(w1,b1,w2,b2,d_train,labels_train,learning_rate,num_epoch): - train_accuracies=[] +def train_mlp_cross_entropy(w1,b1,w2,b2,d_train,labels_train,learning_rate,num_epoch): + train_accuracies,loss_evo=[],[] for k in range(num_epoch): - w1,b1,w2,b2,loss=learn_once_mse(w1,b1,w2,b2,d_train,labels_train,learning_rate) - train_accuracies.append(accuracy(w1,b1,w2,b2,d_train,labels_train)) - return (w1,b1,w2,b2,train_accuracies) + w1,b1,w2,b2,loss=learn_once_cross_entropy(w1,b1,w2,b2,d_train,labels_train,learning_rate) + t=accuracy(w1,b1,w2,b2,d_train,labels_train) + train_accuracies.append(t) + loss_evo.append(loss) + return (w1,b1,w2,b2,train_accuracies,loss_evo) def test_mlp(w1,b1,w2,b2,d_test,labels_test): - a0 = d_test - z1 = np.matmul(a0, w1) + b1 - a1 = 1 / (1 + np.exp(-z1)) - z2 = np.matmul(a1, w2) + b2 - a2 = 1 / (1 + np.exp(-z2)) - predictions = a2 - prediction_2 = np.empty(predictions.shape[0], dtype=int) - for i, ligne in enumerate(predictions): - prediction_2[i] = np.argmax(ligne)+1 - indices_egalite = np.where(prediction_2 == labels_test)[0] - nombre_indices = len(indices_egalite) - return(nombre_indices/len(labels_test)) + test_accuracy=accuracy(w1,b1,w2,b2,d_test,labels_test) + return(test_accuracy) def run_mlp_training(data_train, labels_train, data_test, labels_test,d_h,learning_rate,num_epoch): - d_in = data_train.shape[1] - d_out = max(labels_train) + d_in = data_train.shape[1] # input dimension + d_out = max(labels_train) # output dimension (number of neurons of the output layer) - w1 = (2*np.random.rand(d_in, d_h)-1) # first layer weights - b1 = 2*np.random.rand(1, d_h)-1 # first layer biaises - w2 = 2*np.random.rand(d_h, d_out)-1 # second layer weights - b2 = 2*np.random.rand(1, d_out) -1 # second layer biaises + w1 = np.random.randn(d_in, d_h) # first layer weights + b1 = np.zeros((1, d_h)) # first layer biaises + w2 = np.random.randn(d_h, d_out) # second layer weights + b2 = np.zeros((1, d_out)) # second layer biaises - w1,b1,w2,b2,loss=train_mlp(w1,b1,w2,b2,data_train, labels_train,learning_rate,num_epoch) + w1,b1,w2,b2,accuratie_train,loss=train_mlp_cross_entropy(w1,b1,w2,b2,data_train, labels_train,learning_rate,num_epoch) test_accuracy=test_mlp(w1,b1,w2,b2,data_test, labels_test) - #test_accuracy2=unit_test(w1,b1,w2,b2,data_test, labels_test) - return(loss,test_accuracy) + return(accuratie_train,loss,test_accuracy) def unit_test(w1,b1,w2,b2,data_test, labels_test): pos=0 @@ -147,12 +132,25 @@ def unit_test(w1,b1,w2,b2,data_test, labels_test): if __name__ == "__main__": d, l = read_cifar_batch("data/cifar-10-batches-py/data_batch_1") num_epoch=100 - d_train, l_train, d_test, l_test = split_dataset(d, l, 0.9) - - loss,test_accuracy=run_mlp_training(d_train, l_train, d_test, l_test,64,0.1,num_epoch) - print(test_accuracy) - plt.plot(range(num_epoch), loss, label='evolution de la fonction loss par epoque') - plt.xlabel('epoque') - plt.ylabel('loss') - plt.legend() - plt.show() + dh=64 + learning_rate=0.1 + fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 4)) + for k in range(10): + d_train, l_train, d_test, l_test = split_dataset(d, l, 0.9) + train_accuracy,loss,test_accuracy=run_mlp_training(d_train, l_train, d_test, l_test,dh,learning_rate,num_epoch) + ax1.plot(range(num_epoch),loss,label=f"cross-entropy cas {k}") + ax2.plot(range(num_epoch),train_accuracy,label=f"cross-entropy cas {k}") + print(k) + + ax1.set_xlabel('epoque') + ax1.set_ylabel('loss') + ax1.set_title('evolution de la fonction loss par epoque') + ax1.legend() + + ax2.set_xlabel('epoque') + ax2.set_ylabel('accuracy') + ax2.set_title('evolution de la accuracy') + ax2.legend() + + plt.tight_layout() + plt.show() \ No newline at end of file diff --git a/result/100epoch_10essaie.png b/result/100epoch_10essaie.png new file mode 100644 index 0000000000000000000000000000000000000000..075e92dd74e5695eb039d1e6ed8b001909ed1c56 Binary files /dev/null and b/result/100epoch_10essaie.png differ