deeplearn@ML-RefVm-967342:~$ mkdir cifar10 deeplearn@ML-RefVm-967342:~$ cd cifar10 deeplearn@ML-RefVm-967342:~/cifar10$ kaggle competitions download -c ml530-2022-fall-cifar10 Downloading ml530-2022-fall-cifar10.zip to /home/deeplearn/cifar10 93%|█████████████████████████████████████████████████████████████████████████████████████████▌ | 130M/139M [00:01<00:00, 148MB/s] 100%|████████████████████████████████████████████████████████████████████████████████████████████████| 139M/139M [00:01<00:00, 129MB/s] deeplearn@ML-RefVm-967342:~/cifar10$ wget https://cross-entropy.net/ML530/cifar10-tensors.py.txt --2022-10-25 00:54:12-- https://cross-entropy.net/ML530/cifar10-tensors.py.txt Resolving cross-entropy.net (cross-entropy.net)... 107.180.57.14 Connecting to cross-entropy.net (cross-entropy.net)|107.180.57.14|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 1735 (1.7K) [text/plain] Saving to: ‘cifar10-tensors.py.txt’ cifar10-tensors.py.txt 100%[============================================================>] 1.69K --.-KB/s in 0s 2022-10-25 00:54:12 (870 MB/s) - ‘cifar10-tensors.py.txt’ saved [1735/1735] deeplearn@ML-RefVm-967342:~/cifar10$ time python cifar10-tensors.py.txt real 0m14.426s user 0m13.379s sys 0m1.221s deeplearn@ML-RefVm-967342:~/cifar10$ wget https://cross-entropy.net/ML530/cifar10-train.py.txt --2022-10-25 00:55:12-- https://cross-entropy.net/ML530/cifar10-train.py.txt Resolving cross-entropy.net (cross-entropy.net)... 107.180.57.14 Connecting to cross-entropy.net (cross-entropy.net)|107.180.57.14|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 4598 (4.5K) [text/plain] Saving to: ‘cifar10-train.py.txt’ cifar10-train.py.txt 100%[============================================================>] 4.49K --.-KB/s in 0s 2022-10-25 00:55:12 (183 MB/s) - ‘cifar10-train.py.txt’ saved [4598/4598] deeplearn@ML-RefVm-967342:~/cifar10$ time python cifar10-train.py.txt 2022-10-25 00:55:34.367780: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2022-10-25 00:55:46.539211: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1532] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 10794 MB memory: -> device: 0, name: Tesla K80, pci bus id: 0001:00:00.0, compute capability: 3.7 Model: "model" _______________________________________________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ======================================================================================================================================= input_1 (InputLayer) [(None, 32, 32, 3)] 0 [] conv2d (Conv2D) (None, 32, 32, 16) 448 ['input_1[0][0]'] batch_normalization (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d[0][0]'] activation (Activation) (None, 32, 32, 16) 0 ['batch_normalization[0][0]'] conv2d_1 (Conv2D) (None, 32, 32, 16) 272 ['activation[0][0]'] batch_normalization_1 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_1[0][0]'] activation_1 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_1[0][0]'] conv2d_2 (Conv2D) (None, 32, 32, 16) 2320 ['activation_1[0][0]'] batch_normalization_2 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_2[0][0]'] activation_2 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_2[0][0]'] conv2d_4 (Conv2D) (None, 32, 32, 64) 1088 ['activation[0][0]'] conv2d_3 (Conv2D) (None, 32, 32, 64) 1088 ['activation_2[0][0]'] add (Add) (None, 32, 32, 64) 0 ['conv2d_4[0][0]', 'conv2d_3[0][0]'] batch_normalization_3 (BatchNormalization) (None, 32, 32, 64) 256 ['add[0][0]'] activation_3 (Activation) (None, 32, 32, 64) 0 ['batch_normalization_3[0][0]'] conv2d_5 (Conv2D) (None, 32, 32, 16) 1040 ['activation_3[0][0]'] batch_normalization_4 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_5[0][0]'] activation_4 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_4[0][0]'] conv2d_6 (Conv2D) (None, 32, 32, 16) 2320 ['activation_4[0][0]'] batch_normalization_5 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_6[0][0]'] activation_5 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_5[0][0]'] conv2d_7 (Conv2D) (None, 32, 32, 64) 1088 ['activation_5[0][0]'] add_1 (Add) (None, 32, 32, 64) 0 ['add[0][0]', 'conv2d_7[0][0]'] batch_normalization_6 (BatchNormalization) (None, 32, 32, 64) 256 ['add_1[0][0]'] activation_6 (Activation) (None, 32, 32, 64) 0 ['batch_normalization_6[0][0]'] conv2d_8 (Conv2D) (None, 32, 32, 16) 1040 ['activation_6[0][0]'] batch_normalization_7 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_8[0][0]'] activation_7 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_7[0][0]'] conv2d_9 (Conv2D) (None, 32, 32, 16) 2320 ['activation_7[0][0]'] batch_normalization_8 (BatchNormalization) (None, 32, 32, 16) 64 ['conv2d_9[0][0]'] activation_8 (Activation) (None, 32, 32, 16) 0 ['batch_normalization_8[0][0]'] conv2d_10 (Conv2D) (None, 32, 32, 64) 1088 ['activation_8[0][0]'] add_2 (Add) (None, 32, 32, 64) 0 ['add_1[0][0]', 'conv2d_10[0][0]'] batch_normalization_9 (BatchNormalization) (None, 32, 32, 64) 256 ['add_2[0][0]'] activation_9 (Activation) (None, 32, 32, 64) 0 ['batch_normalization_9[0][0]'] conv2d_11 (Conv2D) (None, 16, 16, 64) 4160 ['activation_9[0][0]'] batch_normalization_10 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_11[0][0]'] ) activation_10 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_10[0][0]'] conv2d_12 (Conv2D) (None, 16, 16, 64) 36928 ['activation_10[0][0]'] batch_normalization_11 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_12[0][0]'] ) activation_11 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_11[0][0]'] conv2d_14 (Conv2D) (None, 16, 16, 128) 8320 ['add_2[0][0]'] conv2d_13 (Conv2D) (None, 16, 16, 128) 8320 ['activation_11[0][0]'] add_3 (Add) (None, 16, 16, 128) 0 ['conv2d_14[0][0]', 'conv2d_13[0][0]'] batch_normalization_12 (BatchNormalization (None, 16, 16, 128) 512 ['add_3[0][0]'] ) activation_12 (Activation) (None, 16, 16, 128) 0 ['batch_normalization_12[0][0]'] conv2d_15 (Conv2D) (None, 16, 16, 64) 8256 ['activation_12[0][0]'] batch_normalization_13 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_15[0][0]'] ) activation_13 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_13[0][0]'] conv2d_16 (Conv2D) (None, 16, 16, 64) 36928 ['activation_13[0][0]'] batch_normalization_14 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_16[0][0]'] ) activation_14 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_14[0][0]'] conv2d_17 (Conv2D) (None, 16, 16, 128) 8320 ['activation_14[0][0]'] add_4 (Add) (None, 16, 16, 128) 0 ['add_3[0][0]', 'conv2d_17[0][0]'] batch_normalization_15 (BatchNormalization (None, 16, 16, 128) 512 ['add_4[0][0]'] ) activation_15 (Activation) (None, 16, 16, 128) 0 ['batch_normalization_15[0][0]'] conv2d_18 (Conv2D) (None, 16, 16, 64) 8256 ['activation_15[0][0]'] batch_normalization_16 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_18[0][0]'] ) activation_16 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_16[0][0]'] conv2d_19 (Conv2D) (None, 16, 16, 64) 36928 ['activation_16[0][0]'] batch_normalization_17 (BatchNormalization (None, 16, 16, 64) 256 ['conv2d_19[0][0]'] ) activation_17 (Activation) (None, 16, 16, 64) 0 ['batch_normalization_17[0][0]'] conv2d_20 (Conv2D) (None, 16, 16, 128) 8320 ['activation_17[0][0]'] add_5 (Add) (None, 16, 16, 128) 0 ['add_4[0][0]', 'conv2d_20[0][0]'] batch_normalization_18 (BatchNormalization (None, 16, 16, 128) 512 ['add_5[0][0]'] ) activation_18 (Activation) (None, 16, 16, 128) 0 ['batch_normalization_18[0][0]'] conv2d_21 (Conv2D) (None, 8, 8, 128) 16512 ['activation_18[0][0]'] batch_normalization_19 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_21[0][0]'] ) activation_19 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_19[0][0]'] conv2d_22 (Conv2D) (None, 8, 8, 128) 147584 ['activation_19[0][0]'] batch_normalization_20 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_22[0][0]'] ) activation_20 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_20[0][0]'] conv2d_24 (Conv2D) (None, 8, 8, 256) 33024 ['add_5[0][0]'] conv2d_23 (Conv2D) (None, 8, 8, 256) 33024 ['activation_20[0][0]'] add_6 (Add) (None, 8, 8, 256) 0 ['conv2d_24[0][0]', 'conv2d_23[0][0]'] batch_normalization_21 (BatchNormalization (None, 8, 8, 256) 1024 ['add_6[0][0]'] ) activation_21 (Activation) (None, 8, 8, 256) 0 ['batch_normalization_21[0][0]'] conv2d_25 (Conv2D) (None, 8, 8, 128) 32896 ['activation_21[0][0]'] batch_normalization_22 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_25[0][0]'] ) activation_22 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_22[0][0]'] conv2d_26 (Conv2D) (None, 8, 8, 128) 147584 ['activation_22[0][0]'] batch_normalization_23 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_26[0][0]'] ) activation_23 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_23[0][0]'] conv2d_27 (Conv2D) (None, 8, 8, 256) 33024 ['activation_23[0][0]'] add_7 (Add) (None, 8, 8, 256) 0 ['add_6[0][0]', 'conv2d_27[0][0]'] batch_normalization_24 (BatchNormalization (None, 8, 8, 256) 1024 ['add_7[0][0]'] ) activation_24 (Activation) (None, 8, 8, 256) 0 ['batch_normalization_24[0][0]'] conv2d_28 (Conv2D) (None, 8, 8, 128) 32896 ['activation_24[0][0]'] batch_normalization_25 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_28[0][0]'] ) activation_25 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_25[0][0]'] conv2d_29 (Conv2D) (None, 8, 8, 128) 147584 ['activation_25[0][0]'] batch_normalization_26 (BatchNormalization (None, 8, 8, 128) 512 ['conv2d_29[0][0]'] ) activation_26 (Activation) (None, 8, 8, 128) 0 ['batch_normalization_26[0][0]'] conv2d_30 (Conv2D) (None, 8, 8, 256) 33024 ['activation_26[0][0]'] add_8 (Add) (None, 8, 8, 256) 0 ['add_7[0][0]', 'conv2d_30[0][0]'] batch_normalization_27 (BatchNormalization (None, 8, 8, 256) 1024 ['add_8[0][0]'] ) activation_27 (Activation) (None, 8, 8, 256) 0 ['batch_normalization_27[0][0]'] average_pooling2d (AveragePooling2D) (None, 1, 1, 256) 0 ['activation_27[0][0]'] flatten (Flatten) (None, 256) 0 ['average_pooling2d[0][0]'] dense (Dense) (None, 10) 2570 ['flatten[0][0]'] ======================================================================================================================================= Total params: 849,002 Trainable params: 843,786 Non-trainable params: 5,216 _______________________________________________________________________________________________________________________________________ learning rate: 0.001 Epoch 1/128 2022-10-25 00:56:14.951001: I tensorflow/stream_executor/cuda/cuda_dnn.cc:384] Loaded cuDNN version 8500 352/352 [==============================] - 104s 153ms/step - loss: 2.0279 - accuracy: 0.4538 - val_loss: 1.9168 - val_accuracy: 0.4864 - lr: 0.0010 learning rate: 0.001 Epoch 2/128 352/352 [==============================] - 52s 149ms/step - loss: 1.5718 - accuracy: 0.5897 - val_loss: 1.5072 - val_accuracy: 0.6074 - lr: 0.0010 learning rate: 0.001 Epoch 3/128 352/352 [==============================] - 53s 149ms/step - loss: 1.3621 - accuracy: 0.6513 - val_loss: 1.3162 - val_accuracy: 0.6612 - lr: 0.0010 learning rate: 0.001 Epoch 4/128 352/352 [==============================] - 53s 149ms/step - loss: 1.2152 - accuracy: 0.6950 - val_loss: 1.3486 - val_accuracy: 0.6544 - lr: 0.0010 learning rate: 0.001 Epoch 5/128 352/352 [==============================] - 53s 149ms/step - loss: 1.1095 - accuracy: 0.7293 - val_loss: 1.1990 - val_accuracy: 0.7004 - lr: 0.0010 learning rate: 0.001 Epoch 6/128 352/352 [==============================] - 53s 149ms/step - loss: 1.0246 - accuracy: 0.7551 - val_loss: 1.6897 - val_accuracy: 0.6158 - lr: 0.0010 learning rate: 0.001 Epoch 7/128 352/352 [==============================] - 52s 149ms/step - loss: 0.9693 - accuracy: 0.7689 - val_loss: 1.2786 - val_accuracy: 0.6772 - lr: 0.0010 learning rate: 0.001 Epoch 8/128 352/352 [==============================] - 52s 149ms/step - loss: 0.9120 - accuracy: 0.7870 - val_loss: 1.0321 - val_accuracy: 0.7536 - lr: 0.0010 learning rate: 0.001 Epoch 9/128 352/352 [==============================] - 53s 149ms/step - loss: 0.8779 - accuracy: 0.7980 - val_loss: 1.1817 - val_accuracy: 0.7074 - lr: 0.0010 learning rate: 0.001 Epoch 10/128 352/352 [==============================] - 52s 149ms/step - loss: 0.8414 - accuracy: 0.8085 - val_loss: 1.0932 - val_accuracy: 0.7444 - lr: 0.0010 learning rate: 0.001 Epoch 11/128 352/352 [==============================] - 52s 149ms/step - loss: 0.8066 - accuracy: 0.8166 - val_loss: 1.4651 - val_accuracy: 0.6564 - lr: 0.0010 learning rate: 0.001 Epoch 12/128 352/352 [==============================] - 52s 149ms/step - loss: 0.7858 - accuracy: 0.8253 - val_loss: 1.3152 - val_accuracy: 0.7016 - lr: 0.0010 learning rate: 0.001 Epoch 13/128 352/352 [==============================] - 52s 149ms/step - loss: 0.7638 - accuracy: 0.8300 - val_loss: 1.0757 - val_accuracy: 0.7338 - lr: 0.0010 learning rate: 0.001 Epoch 14/128 352/352 [==============================] - 53s 149ms/step - loss: 0.7399 - accuracy: 0.8367 - val_loss: 0.8446 - val_accuracy: 0.8046 - lr: 0.0010 learning rate: 0.001 Epoch 15/128 352/352 [==============================] - 53s 149ms/step - loss: 0.7233 - accuracy: 0.8407 - val_loss: 0.9187 - val_accuracy: 0.7828 - lr: 0.0010 learning rate: 0.001 Epoch 16/128 352/352 [==============================] - 53s 149ms/step - loss: 0.7015 - accuracy: 0.8479 - val_loss: 1.0049 - val_accuracy: 0.7602 - lr: 0.0010 learning rate: 0.001 Epoch 17/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6830 - accuracy: 0.8531 - val_loss: 0.7889 - val_accuracy: 0.8182 - lr: 0.0010 learning rate: 0.001 Epoch 18/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6712 - accuracy: 0.8566 - val_loss: 0.8253 - val_accuracy: 0.8102 - lr: 0.0010 learning rate: 0.001 Epoch 19/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6557 - accuracy: 0.8617 - val_loss: 1.3109 - val_accuracy: 0.7144 - lr: 0.0010 learning rate: 0.001 Epoch 20/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6451 - accuracy: 0.8647 - val_loss: 1.0269 - val_accuracy: 0.7724 - lr: 0.0010 learning rate: 0.001 Epoch 21/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6278 - accuracy: 0.8695 - val_loss: 0.8096 - val_accuracy: 0.8188 - lr: 0.0010 learning rate: 0.001 Epoch 22/128 352/352 [==============================] - 52s 149ms/step - loss: 0.6259 - accuracy: 0.8672 - val_loss: 0.8306 - val_accuracy: 0.8164 - lr: 0.0010 learning rate: 0.001 Epoch 23/128 352/352 [==============================] - 53s 149ms/step - loss: 0.6217 - accuracy: 0.8709 - val_loss: 0.8711 - val_accuracy: 0.8002 - lr: 0.0010 learning rate: 0.001 Epoch 24/128 352/352 [==============================] - 52s 148ms/step - loss: 0.6038 - accuracy: 0.8775 - val_loss: 0.8439 - val_accuracy: 0.8022 - lr: 0.0010 learning rate: 0.001 Epoch 25/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5937 - accuracy: 0.8786 - val_loss: 1.0260 - val_accuracy: 0.7728 - lr: 0.0010 learning rate: 0.001 Epoch 26/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5906 - accuracy: 0.8803 - val_loss: 0.9339 - val_accuracy: 0.7898 - lr: 0.0010 learning rate: 0.001 Epoch 27/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5834 - accuracy: 0.8828 - val_loss: 1.0475 - val_accuracy: 0.7744 - lr: 0.0010 learning rate: 0.001 Epoch 28/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5773 - accuracy: 0.8845 - val_loss: 0.7532 - val_accuracy: 0.8336 - lr: 0.0010 learning rate: 0.001 Epoch 29/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5705 - accuracy: 0.8851 - val_loss: 0.9674 - val_accuracy: 0.7826 - lr: 0.0010 learning rate: 0.001 Epoch 30/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5577 - accuracy: 0.8899 - val_loss: 0.7854 - val_accuracy: 0.8226 - lr: 0.0010 learning rate: 0.001 Epoch 31/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5575 - accuracy: 0.8896 - val_loss: 0.7202 - val_accuracy: 0.8440 - lr: 0.0010 learning rate: 0.001 Epoch 32/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5480 - accuracy: 0.8930 - val_loss: 1.0867 - val_accuracy: 0.7656 - lr: 0.0010 learning rate: 0.001 Epoch 33/128 352/352 [==============================] - 53s 149ms/step - loss: 0.5462 - accuracy: 0.8929 - val_loss: 0.7522 - val_accuracy: 0.8384 - lr: 0.0010 learning rate: 0.001 Epoch 34/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5315 - accuracy: 0.8982 - val_loss: 0.8031 - val_accuracy: 0.8250 - lr: 0.0010 learning rate: 0.001 Epoch 35/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5344 - accuracy: 0.8967 - val_loss: 0.6944 - val_accuracy: 0.8476 - lr: 0.0010 learning rate: 0.001 Epoch 36/128 352/352 [==============================] - 52s 148ms/step - loss: 0.5297 - accuracy: 0.8989 - val_loss: 0.8761 - val_accuracy: 0.8024 - lr: 0.0010 learning rate: 0.001 Epoch 37/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5237 - accuracy: 0.8998 - val_loss: 0.7408 - val_accuracy: 0.8420 - lr: 0.0010 learning rate: 0.001 Epoch 38/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5256 - accuracy: 0.9006 - val_loss: 0.7845 - val_accuracy: 0.8280 - lr: 0.0010 learning rate: 0.001 Epoch 39/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5151 - accuracy: 0.9022 - val_loss: 0.8866 - val_accuracy: 0.8084 - lr: 0.0010 learning rate: 0.001 Epoch 40/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5113 - accuracy: 0.9036 - val_loss: 0.8693 - val_accuracy: 0.8204 - lr: 0.0010 learning rate: 0.001 Epoch 41/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5052 - accuracy: 0.9056 - val_loss: 0.9489 - val_accuracy: 0.7974 - lr: 0.0010 learning rate: 0.001 Epoch 42/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5034 - accuracy: 0.9060 - val_loss: 0.8013 - val_accuracy: 0.8352 - lr: 0.0010 learning rate: 0.001 Epoch 43/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5012 - accuracy: 0.9052 - val_loss: 0.8768 - val_accuracy: 0.8084 - lr: 0.0010 learning rate: 0.001 Epoch 44/128 352/352 [==============================] - 52s 149ms/step - loss: 0.5000 - accuracy: 0.9075 - val_loss: 0.7195 - val_accuracy: 0.8406 - lr: 0.0010 learning rate: 0.001 Epoch 45/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4923 - accuracy: 0.9087 - val_loss: 0.7943 - val_accuracy: 0.8362 - lr: 0.0010 learning rate: 0.001 Epoch 46/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4893 - accuracy: 0.9107 - val_loss: 0.9047 - val_accuracy: 0.8124 - lr: 0.0010 learning rate: 0.001 Epoch 47/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4893 - accuracy: 0.9103 - val_loss: 0.8025 - val_accuracy: 0.8222 - lr: 0.0010 learning rate: 0.001 Epoch 48/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4839 - accuracy: 0.9110 - val_loss: 0.6702 - val_accuracy: 0.8510 - lr: 0.0010 learning rate: 0.001 Epoch 49/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4770 - accuracy: 0.9149 - val_loss: 0.8565 - val_accuracy: 0.8158 - lr: 0.0010 learning rate: 0.001 Epoch 50/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4804 - accuracy: 0.9112 - val_loss: 0.9700 - val_accuracy: 0.8076 - lr: 0.0010 learning rate: 0.001 Epoch 51/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4760 - accuracy: 0.9140 - val_loss: 0.7078 - val_accuracy: 0.8612 - lr: 0.0010 learning rate: 0.001 Epoch 52/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4738 - accuracy: 0.9149 - val_loss: 0.7280 - val_accuracy: 0.8460 - lr: 0.0010 learning rate: 0.001 Epoch 53/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4636 - accuracy: 0.9184 - val_loss: 0.7270 - val_accuracy: 0.8542 - lr: 0.0010 learning rate: 0.001 Epoch 54/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4719 - accuracy: 0.9156 - val_loss: 0.7563 - val_accuracy: 0.8396 - lr: 0.0010 learning rate: 0.001 Epoch 55/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4650 - accuracy: 0.9175 - val_loss: 0.7755 - val_accuracy: 0.8338 - lr: 0.0010 learning rate: 0.001 Epoch 56/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4643 - accuracy: 0.9184 - val_loss: 0.8020 - val_accuracy: 0.8386 - lr: 0.0010 learning rate: 0.001 Epoch 57/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4626 - accuracy: 0.9170 - val_loss: 0.8456 - val_accuracy: 0.8258 - lr: 0.0010 learning rate: 0.001 Epoch 58/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4535 - accuracy: 0.9232 - val_loss: 0.6345 - val_accuracy: 0.8706 - lr: 0.0010 learning rate: 0.001 Epoch 59/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4577 - accuracy: 0.9210 - val_loss: 0.8349 - val_accuracy: 0.8236 - lr: 0.0010 learning rate: 0.001 Epoch 60/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4602 - accuracy: 0.9182 - val_loss: 0.7189 - val_accuracy: 0.8578 - lr: 0.0010 learning rate: 0.001 Epoch 61/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4542 - accuracy: 0.9211 - val_loss: 0.8870 - val_accuracy: 0.8012 - lr: 0.0010 learning rate: 0.001 Epoch 62/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4483 - accuracy: 0.9226 - val_loss: 0.7210 - val_accuracy: 0.8530 - lr: 0.0010 learning rate: 0.001 Epoch 63/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4508 - accuracy: 0.9218 - val_loss: 0.6634 - val_accuracy: 0.8612 - lr: 0.0010 learning rate: 0.001 Epoch 64/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4534 - accuracy: 0.9222 - val_loss: 0.6466 - val_accuracy: 0.8654 - lr: 0.0010 learning rate: 0.001 Epoch 65/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4420 - accuracy: 0.9242 - val_loss: 0.6616 - val_accuracy: 0.8640 - lr: 0.0010 learning rate: 0.001 Epoch 66/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4446 - accuracy: 0.9225 - val_loss: 0.7486 - val_accuracy: 0.8452 - lr: 0.0010 learning rate: 0.001 Epoch 67/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4449 - accuracy: 0.9226 - val_loss: 0.6997 - val_accuracy: 0.8618 - lr: 0.0010 learning rate: 0.001 Epoch 68/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4359 - accuracy: 0.9269 - val_loss: 0.9022 - val_accuracy: 0.8242 - lr: 0.0010 learning rate: 0.001 Epoch 69/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4355 - accuracy: 0.9270 - val_loss: 0.6870 - val_accuracy: 0.8624 - lr: 0.0010 learning rate: 0.001 Epoch 70/128 352/352 [==============================] - 52s 148ms/step - loss: 0.4326 - accuracy: 0.9279 - val_loss: 0.7207 - val_accuracy: 0.8518 - lr: 0.0010 learning rate: 0.001 Epoch 71/128 352/352 [==============================] - 52s 149ms/step - loss: 0.4357 - accuracy: 0.9253 - val_loss: 1.1094 - val_accuracy: 0.7726 - lr: 0.0010 learning rate: 0.001 Epoch 72/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4361 - accuracy: 0.9249 - val_loss: 0.7951 - val_accuracy: 0.8392 - lr: 0.0010 learning rate: 0.001 Epoch 73/128 352/352 [==============================] - 53s 150ms/step - loss: 0.4292 - accuracy: 0.9296 - val_loss: 0.7203 - val_accuracy: 0.8534 - lr: 0.0010 learning rate: 0.001 Epoch 74/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4324 - accuracy: 0.9279 - val_loss: 0.7589 - val_accuracy: 0.8454 - lr: 0.0010 learning rate: 0.001 Epoch 75/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4268 - accuracy: 0.9298 - val_loss: 0.9240 - val_accuracy: 0.8178 - lr: 0.0010 learning rate: 0.001 Epoch 76/128 352/352 [==============================] - 53s 150ms/step - loss: 0.4257 - accuracy: 0.9291 - val_loss: 0.6947 - val_accuracy: 0.8646 - lr: 0.0010 learning rate: 0.001 Epoch 77/128 352/352 [==============================] - 53s 150ms/step - loss: 0.4204 - accuracy: 0.9311 - val_loss: 0.6310 - val_accuracy: 0.8816 - lr: 0.0010 learning rate: 0.001 Epoch 78/128 352/352 [==============================] - 53s 150ms/step - loss: 0.4270 - accuracy: 0.9283 - val_loss: 0.7477 - val_accuracy: 0.8478 - lr: 0.0010 learning rate: 0.001 Epoch 79/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4242 - accuracy: 0.9291 - val_loss: 0.7528 - val_accuracy: 0.8502 - lr: 0.0010 learning rate: 0.001 Epoch 80/128 352/352 [==============================] - 53s 149ms/step - loss: 0.4200 - accuracy: 0.9314 - val_loss: 0.6925 - val_accuracy: 0.8606 - lr: 0.0010 learning rate: 0.0001 Epoch 81/128 352/352 [==============================] - 53s 149ms/step - loss: 0.3477 - accuracy: 0.9570 - val_loss: 0.5094 - val_accuracy: 0.9118 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 82/128 352/352 [==============================] - 53s 149ms/step - loss: 0.3206 - accuracy: 0.9684 - val_loss: 0.5096 - val_accuracy: 0.9130 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 83/128 352/352 [==============================] - 53s 149ms/step - loss: 0.3077 - accuracy: 0.9707 - val_loss: 0.5015 - val_accuracy: 0.9156 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 84/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2975 - accuracy: 0.9735 - val_loss: 0.5108 - val_accuracy: 0.9128 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 85/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2899 - accuracy: 0.9752 - val_loss: 0.5119 - val_accuracy: 0.9134 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 86/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2837 - accuracy: 0.9769 - val_loss: 0.5077 - val_accuracy: 0.9144 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 87/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2737 - accuracy: 0.9795 - val_loss: 0.5185 - val_accuracy: 0.9140 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 88/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2706 - accuracy: 0.9796 - val_loss: 0.5054 - val_accuracy: 0.9182 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 89/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2650 - accuracy: 0.9808 - val_loss: 0.5154 - val_accuracy: 0.9180 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 90/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2624 - accuracy: 0.9807 - val_loss: 0.5029 - val_accuracy: 0.9170 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 91/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2570 - accuracy: 0.9820 - val_loss: 0.5171 - val_accuracy: 0.9174 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 92/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2513 - accuracy: 0.9833 - val_loss: 0.5310 - val_accuracy: 0.9126 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 93/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2471 - accuracy: 0.9842 - val_loss: 0.5114 - val_accuracy: 0.9174 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 94/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2441 - accuracy: 0.9846 - val_loss: 0.5217 - val_accuracy: 0.9148 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 95/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2400 - accuracy: 0.9856 - val_loss: 0.5168 - val_accuracy: 0.9152 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 96/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2372 - accuracy: 0.9856 - val_loss: 0.5113 - val_accuracy: 0.9154 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 97/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2340 - accuracy: 0.9857 - val_loss: 0.5159 - val_accuracy: 0.9166 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 98/128 352/352 [==============================] - 53s 149ms/step - loss: 0.2305 - accuracy: 0.9866 - val_loss: 0.5239 - val_accuracy: 0.9164 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 99/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2303 - accuracy: 0.9863 - val_loss: 0.5168 - val_accuracy: 0.9172 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 100/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2259 - accuracy: 0.9868 - val_loss: 0.5041 - val_accuracy: 0.9170 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 101/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2223 - accuracy: 0.9875 - val_loss: 0.5190 - val_accuracy: 0.9150 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 102/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2172 - accuracy: 0.9892 - val_loss: 0.5198 - val_accuracy: 0.9170 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 103/128 352/352 [==============================] - 52s 149ms/step - loss: 0.2182 - accuracy: 0.9875 - val_loss: 0.5179 - val_accuracy: 0.9194 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 104/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2138 - accuracy: 0.9892 - val_loss: 0.5079 - val_accuracy: 0.9176 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 105/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2118 - accuracy: 0.9889 - val_loss: 0.5779 - val_accuracy: 0.9080 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 106/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2106 - accuracy: 0.9885 - val_loss: 0.5295 - val_accuracy: 0.9154 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 107/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2074 - accuracy: 0.9896 - val_loss: 0.5143 - val_accuracy: 0.9166 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 108/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2067 - accuracy: 0.9895 - val_loss: 0.5156 - val_accuracy: 0.9194 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 109/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2032 - accuracy: 0.9906 - val_loss: 0.5196 - val_accuracy: 0.9162 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 110/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2009 - accuracy: 0.9906 - val_loss: 0.5256 - val_accuracy: 0.9192 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 111/128 352/352 [==============================] - 52s 148ms/step - loss: 0.2008 - accuracy: 0.9898 - val_loss: 0.5383 - val_accuracy: 0.9172 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 112/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1966 - accuracy: 0.9907 - val_loss: 0.5406 - val_accuracy: 0.9150 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 113/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1965 - accuracy: 0.9902 - val_loss: 0.5524 - val_accuracy: 0.9124 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 114/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1952 - accuracy: 0.9900 - val_loss: 0.5392 - val_accuracy: 0.9154 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 115/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1926 - accuracy: 0.9910 - val_loss: 0.5333 - val_accuracy: 0.9126 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 116/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1908 - accuracy: 0.9915 - val_loss: 0.5128 - val_accuracy: 0.9188 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 117/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1884 - accuracy: 0.9914 - val_loss: 0.5089 - val_accuracy: 0.9216 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 118/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1874 - accuracy: 0.9914 - val_loss: 0.5521 - val_accuracy: 0.9118 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 119/128 352/352 [==============================] - 52s 149ms/step - loss: 0.1864 - accuracy: 0.9916 - val_loss: 0.5355 - val_accuracy: 0.9132 - lr: 1.0000e-04 learning rate: 0.0001 Epoch 120/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1844 - accuracy: 0.9916 - val_loss: 0.5442 - val_accuracy: 0.9142 - lr: 1.0000e-04 learning rate: 1e-05 Epoch 121/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1799 - accuracy: 0.9933 - val_loss: 0.5301 - val_accuracy: 0.9184 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 122/128 352/352 [==============================] - 52s 149ms/step - loss: 0.1783 - accuracy: 0.9939 - val_loss: 0.5235 - val_accuracy: 0.9186 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 123/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1773 - accuracy: 0.9944 - val_loss: 0.5264 - val_accuracy: 0.9176 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 124/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1770 - accuracy: 0.9941 - val_loss: 0.5199 - val_accuracy: 0.9188 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 125/128 352/352 [==============================] - 52s 149ms/step - loss: 0.1764 - accuracy: 0.9941 - val_loss: 0.5208 - val_accuracy: 0.9200 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 126/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1759 - accuracy: 0.9941 - val_loss: 0.5205 - val_accuracy: 0.9196 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 127/128 352/352 [==============================] - 52s 149ms/step - loss: 0.1753 - accuracy: 0.9945 - val_loss: 0.5210 - val_accuracy: 0.9210 - lr: 1.0000e-05 learning rate: 1e-05 Epoch 128/128 352/352 [==============================] - 52s 148ms/step - loss: 0.1747 - accuracy: 0.9948 - val_loss: 0.5206 - val_accuracy: 0.9190 - lr: 1.0000e-05 313/313 [==============================] - 5s 15ms/step real 113m26.817s user 132m11.679s sys 2m33.721s deeplearn@ML-RefVm-967342:~/cifar10$ kaggle competitions submit -c ml530-2022-fall-cifar10 -f predictions.csv -m "default" 100%|██████████████████████████████████████████████████████████████████████████████████████████████| 78.1k/78.1k [00:00<00:00, 146kB/s] Successfully submitted to ml530-2022-fall-cifar10deeplearn@ML-RefVm-967342:~/cifar10$