diff --git a/.gitignore b/.gitignore index f3436fe1fd3e8a7064887098b38e50dfda48b27d..098761275e11938a8dee35e4ba12f7e8ebf017d5 100644 --- a/.gitignore +++ b/.gitignore @@ -3,7 +3,7 @@ # Data data/* -transfer_learning/hymenoptera_data/* +hymenoptera_data/* # Torch model *.pt diff --git a/TD2 Deep Learning.ipynb b/TD2 Deep Learning.ipynb index 913375254c0118bacd6136f2fec679e36efaa423..28c25b62ef9b7442c9bb37e9d3d52d5a3577a033 100644 --- a/TD2 Deep Learning.ipynb +++ b/TD2 Deep Learning.ipynb @@ -33,7 +33,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 1, "id": "330a42f5", "metadata": {}, "outputs": [ @@ -41,16 +41,18 @@ "name": "stdout", "output_type": "stream", "text": [ - "Requirement already satisfied: torch in /opt/homebrew/lib/python3.11/site-packages (2.1.0)\n", + "Requirement already satisfied: torch in /opt/homebrew/lib/python3.11/site-packages (2.1.1)\n", "Requirement already satisfied: torchvision in /opt/homebrew/lib/python3.11/site-packages (0.16.0)\n", - "Requirement already satisfied: filelock in /opt/homebrew/lib/python3.11/site-packages (from torch) (3.12.3)\n", - "Requirement already satisfied: typing-extensions in /opt/homebrew/lib/python3.11/site-packages (from torch) (4.5.0)\n", + "Requirement already satisfied: filelock in /opt/homebrew/lib/python3.11/site-packages (from torch) (3.13.1)\n", + "Requirement already satisfied: typing-extensions in /opt/homebrew/lib/python3.11/site-packages (from torch) (4.8.0)\n", "Requirement already satisfied: sympy in /opt/homebrew/lib/python3.11/site-packages (from torch) (1.12)\n", - "Requirement already satisfied: networkx in /opt/homebrew/lib/python3.11/site-packages (from torch) (3.1)\n", + "Requirement already satisfied: networkx in /opt/homebrew/lib/python3.11/site-packages (from torch) (3.2.1)\n", "Requirement already satisfied: jinja2 in /opt/homebrew/lib/python3.11/site-packages (from torch) (3.1.2)\n", "Requirement already satisfied: fsspec in /opt/homebrew/lib/python3.11/site-packages (from torch) (2023.10.0)\n", "Requirement already satisfied: numpy in /opt/homebrew/lib/python3.11/site-packages (from torchvision) (1.24.3)\n", "Requirement already satisfied: requests in /opt/homebrew/lib/python3.11/site-packages (from torchvision) (2.31.0)\n", + "Collecting torch\n", + " Using cached torch-2.1.0-cp311-none-macosx_11_0_arm64.whl.metadata (24 kB)\n", "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /opt/homebrew/lib/python3.11/site-packages (from torchvision) (9.5.0)\n", "Requirement already satisfied: MarkupSafe>=2.0 in /opt/homebrew/lib/python3.11/site-packages (from jinja2->torch) (2.1.3)\n", "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/homebrew/lib/python3.11/site-packages (from requests->torchvision) (3.2.0)\n", @@ -58,6 +60,13 @@ "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/homebrew/lib/python3.11/site-packages (from requests->torchvision) (2.0.4)\n", "Requirement already satisfied: certifi>=2017.4.17 in /opt/homebrew/lib/python3.11/site-packages (from requests->torchvision) (2023.7.22)\n", "Requirement already satisfied: mpmath>=0.19 in /opt/homebrew/lib/python3.11/site-packages (from sympy->torch) (1.3.0)\n", + "Using cached torch-2.1.0-cp311-none-macosx_11_0_arm64.whl (59.6 MB)\n", + "Installing collected packages: torch\n", + " Attempting uninstall: torch\n", + " Found existing installation: torch 2.1.1\n", + " Uninstalling torch-2.1.1:\n", + " Successfully uninstalled torch-2.1.1\n", + "Successfully installed torch-2.1.0\n", "Note: you may need to restart the kernel to use updated packages.\n" ] } @@ -77,7 +86,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 7, "id": "b1950f0a", "metadata": {}, "outputs": [ @@ -85,34 +94,34 @@ "name": "stdout", "output_type": "stream", "text": [ - "tensor([[-2.1639, 0.8244, 1.5943, 2.2299, 1.0955, 0.0066, 0.7362, 0.9062,\n", - " 0.2100, -0.2733],\n", - " [ 0.5875, -0.5496, 0.0346, 0.7855, -0.2050, -1.2647, -0.2694, -0.8206,\n", - " 0.9529, 0.1084],\n", - " [ 0.3888, -0.4262, -0.7277, -1.7934, 0.5649, 0.5087, 0.0708, -1.6240,\n", - " 0.3957, -1.3972],\n", - " [-0.3963, 1.0606, 0.3235, -1.9662, 1.4115, -1.5677, -0.6780, -0.2157,\n", - " -0.4618, -2.2241],\n", - " [-0.6263, -1.0853, -0.4133, 0.3194, -0.6319, -0.0079, 0.6466, 0.0214,\n", - " 1.9381, 1.1991],\n", - " [-0.3298, 1.2924, -0.4258, -0.1929, -0.8135, -0.3615, 0.5608, -0.9598,\n", - " -0.5712, 0.1913],\n", - " [ 0.3499, 1.3629, 0.6441, -0.0488, -0.3175, 0.4435, 0.0610, 0.7947,\n", - " -0.6018, -0.0878],\n", - " [ 2.5382, 2.0874, 1.2169, 0.5647, 2.0643, -0.8136, -0.4945, -0.6485,\n", - " 0.9139, -0.7381],\n", - " [ 0.3832, -0.3594, 1.4491, 0.1338, 0.4270, 2.5786, 1.0514, 1.5069,\n", - " 0.6788, -1.2688],\n", - " [ 0.2101, 1.0140, 2.6259, -0.4087, -0.7627, -1.1328, -0.2586, 0.3925,\n", - " 0.4864, -1.2308],\n", - " [ 0.6330, 2.2641, -0.5665, 1.3809, -0.7827, -2.1725, 0.0656, 0.9261,\n", - " -1.4323, 0.4885],\n", - " [-0.2806, 1.1392, -0.5848, -0.8053, -0.2428, 0.5051, 0.7809, 2.5740,\n", - " -1.1393, 0.7766],\n", - " [ 0.8259, -1.0614, -0.9382, -1.7892, 0.7500, 1.1515, 0.2378, 0.7476,\n", - " -0.3611, -0.7560],\n", - " [-0.5055, -0.7351, -0.1483, -1.6377, 1.9209, 1.2716, 1.9331, -2.2570,\n", - " -0.7971, -0.2331]])\n", + "tensor([[ 0.6040, 1.1735, -0.3281, 0.0421, -1.1511, 2.2371, 1.0696, -1.0485,\n", + " 1.6036, 0.3566],\n", + " [ 0.4657, -2.0252, 1.0225, 0.6135, -0.0736, -0.3264, -0.8538, 0.8715,\n", + " 0.7152, -1.6525],\n", + " [-0.2131, -0.5202, -0.6285, 0.3349, 0.4621, 1.1021, -0.4378, 0.1772,\n", + " 0.4466, -0.4951],\n", + " [ 0.6991, -0.7424, -0.4973, -0.3371, 1.2757, -0.9223, 1.8115, -0.0630,\n", + " 1.2063, -0.9551],\n", + " [ 1.0789, 0.7348, 0.1372, 2.0986, 1.8549, 0.4573, 1.1224, 0.0214,\n", + " -1.3154, 0.1054],\n", + " [-0.9701, -1.5672, 0.9513, -1.3851, 0.5345, 1.3727, 0.9487, 0.2369,\n", + " -0.9851, -1.0116],\n", + " [-0.8436, -1.7018, -1.5359, -1.1312, 1.1230, 1.1783, 1.3175, -1.4586,\n", + " -0.7515, -0.2875],\n", + " [ 0.3251, 0.8702, 0.2339, 0.5709, 1.5246, 1.6322, 0.0345, -0.0938,\n", + " -0.9833, -1.3456],\n", + " [ 1.0383, 1.5610, -0.2773, 1.2003, -2.4670, -1.5935, 2.1415, 0.2746,\n", + " 0.8204, 0.6541],\n", + " [ 1.9339, 0.6013, 0.0238, 1.0870, -1.8386, -1.3108, 0.9036, 0.2563,\n", + " -1.0777, -1.3211],\n", + " [-0.2505, -1.1471, 0.1935, -0.6741, 1.6191, 0.0443, -1.7800, -0.5102,\n", + " -0.5699, -1.0624],\n", + " [-1.1685, 0.9333, 0.3156, -0.1072, -0.7786, -1.0011, 0.2067, -0.4636,\n", + " -0.6962, -0.0329],\n", + " [ 0.1517, -1.5085, -1.6767, -1.3724, -0.9289, -0.9293, -0.6079, -0.7638,\n", + " -0.6661, 0.1128],\n", + " [-0.6735, 1.3956, 2.0874, 0.6661, -0.8347, -0.9336, -0.4093, 1.7604,\n", + " 0.4021, 0.4577]])\n", "AlexNet(\n", " (features): Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n", @@ -182,7 +191,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 8, "id": "6e18f2fd", "metadata": {}, "outputs": [ @@ -216,7 +225,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 9, "id": "462666a2", "metadata": {}, "outputs": [ @@ -297,7 +306,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 10, "id": "317bf070", "metadata": {}, "outputs": [ @@ -361,7 +370,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 11, "id": "4b53f229", "metadata": {}, "outputs": [ @@ -369,36 +378,36 @@ "name": "stdout", "output_type": "stream", "text": [ - "Epoch: 0 \tTraining Loss: 43.332731 \tValidation Loss: 37.769932\n", - "Validation loss decreased (inf --> 37.769932). Saving model ...\n", - "Epoch: 1 \tTraining Loss: 34.436545 \tValidation Loss: 32.273618\n", - "Validation loss decreased (37.769932 --> 32.273618). Saving model ...\n", - "Epoch: 2 \tTraining Loss: 31.228934 \tValidation Loss: 30.754028\n", - "Validation loss decreased (32.273618 --> 30.754028). Saving model ...\n", - "Epoch: 3 \tTraining Loss: 28.964067 \tValidation Loss: 28.470241\n", - "Validation loss decreased (30.754028 --> 28.470241). Saving model ...\n", - "Epoch: 4 \tTraining Loss: 27.164884 \tValidation Loss: 27.049564\n", - "Validation loss decreased (28.470241 --> 27.049564). Saving model ...\n", - "Epoch: 5 \tTraining Loss: 25.689214 \tValidation Loss: 25.591283\n", - "Validation loss decreased (27.049564 --> 25.591283). Saving model ...\n", - "Epoch: 6 \tTraining Loss: 24.495015 \tValidation Loss: 25.218405\n", - "Validation loss decreased (25.591283 --> 25.218405). Saving model ...\n", - "Epoch: 7 \tTraining Loss: 23.439156 \tValidation Loss: 24.223355\n", - "Validation loss decreased (25.218405 --> 24.223355). Saving model ...\n", - "Epoch: 8 \tTraining Loss: 22.541391 \tValidation Loss: 23.207256\n", - "Validation loss decreased (24.223355 --> 23.207256). Saving model ...\n", - "Epoch: 9 \tTraining Loss: 21.637936 \tValidation Loss: 23.530791\n", - "Epoch: 10 \tTraining Loss: 20.842096 \tValidation Loss: 23.569502\n", - "Epoch: 11 \tTraining Loss: 20.134623 \tValidation Loss: 22.848159\n", - "Validation loss decreased (23.207256 --> 22.848159). Saving model ...\n", - "Epoch: 12 \tTraining Loss: 19.435100 \tValidation Loss: 22.734210\n", - "Validation loss decreased (22.848159 --> 22.734210). Saving model ...\n", - "Epoch: 13 \tTraining Loss: 18.774194 \tValidation Loss: 23.018766\n", - "Epoch: 14 \tTraining Loss: 18.167667 \tValidation Loss: 22.164276\n", - "Validation loss decreased (22.734210 --> 22.164276). Saving model ...\n", - "Epoch: 15 \tTraining Loss: 17.551007 \tValidation Loss: 22.408603\n", - "Epoch: 16 \tTraining Loss: 16.983553 \tValidation Loss: 22.816888\n", - "Epoch: 17 \tTraining Loss: 16.406738 \tValidation Loss: 23.193989\n", + "Epoch: 0 \tTraining Loss: 42.364139 \tValidation Loss: 37.301467\n", + "Validation loss decreased (inf --> 37.301467). Saving model ...\n", + "Epoch: 1 \tTraining Loss: 34.038406 \tValidation Loss: 31.281009\n", + "Validation loss decreased (37.301467 --> 31.281009). Saving model ...\n", + "Epoch: 2 \tTraining Loss: 30.468120 \tValidation Loss: 28.936136\n", + "Validation loss decreased (31.281009 --> 28.936136). Saving model ...\n", + "Epoch: 3 \tTraining Loss: 28.288131 \tValidation Loss: 27.592625\n", + "Validation loss decreased (28.936136 --> 27.592625). Saving model ...\n", + "Epoch: 4 \tTraining Loss: 26.558347 \tValidation Loss: 25.626183\n", + "Validation loss decreased (27.592625 --> 25.626183). Saving model ...\n", + "Epoch: 5 \tTraining Loss: 25.071003 \tValidation Loss: 24.514871\n", + "Validation loss decreased (25.626183 --> 24.514871). Saving model ...\n", + "Epoch: 6 \tTraining Loss: 23.819706 \tValidation Loss: 24.086769\n", + "Validation loss decreased (24.514871 --> 24.086769). Saving model ...\n", + "Epoch: 7 \tTraining Loss: 22.775944 \tValidation Loss: 23.118577\n", + "Validation loss decreased (24.086769 --> 23.118577). Saving model ...\n", + "Epoch: 8 \tTraining Loss: 21.865358 \tValidation Loss: 22.737836\n", + "Validation loss decreased (23.118577 --> 22.737836). Saving model ...\n", + "Epoch: 9 \tTraining Loss: 20.979377 \tValidation Loss: 22.304552\n", + "Validation loss decreased (22.737836 --> 22.304552). Saving model ...\n", + "Epoch: 10 \tTraining Loss: 20.214349 \tValidation Loss: 22.393152\n", + "Epoch: 11 \tTraining Loss: 19.487387 \tValidation Loss: 21.463038\n", + "Validation loss decreased (22.304552 --> 21.463038). Saving model ...\n", + "Epoch: 12 \tTraining Loss: 18.779783 \tValidation Loss: 21.601195\n", + "Epoch: 13 \tTraining Loss: 18.118136 \tValidation Loss: 21.570440\n", + "Epoch: 14 \tTraining Loss: 17.533442 \tValidation Loss: 21.150484\n", + "Validation loss decreased (21.463038 --> 21.150484). Saving model ...\n", + "Epoch: 15 \tTraining Loss: 16.911449 \tValidation Loss: 21.705741\n", + "Epoch: 16 \tTraining Loss: 16.392025 \tValidation Loss: 21.431641\n", + "Epoch: 17 \tTraining Loss: 15.851257 \tValidation Loss: 21.271693\n", "Early stopping after 17 epochs.\n" ] } @@ -496,13 +505,13 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 12, "id": "d39df818", "metadata": {}, "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "<Figure size 640x480 with 1 Axes>" ] @@ -532,7 +541,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 13, "id": "e93efdfc", "metadata": {}, "outputs": [ @@ -540,20 +549,20 @@ "name": "stdout", "output_type": "stream", "text": [ - "Test Loss: 21.819643\n", + "Test Loss: 21.458072\n", "\n", - "Test Accuracy of airplane: 59% (596/1000)\n", - "Test Accuracy of automobile: 74% (742/1000)\n", - "Test Accuracy of bird: 42% (421/1000)\n", - "Test Accuracy of cat: 45% (455/1000)\n", - "Test Accuracy of deer: 63% (639/1000)\n", - "Test Accuracy of dog: 55% (558/1000)\n", - "Test Accuracy of frog: 70% (701/1000)\n", - "Test Accuracy of horse: 63% (632/1000)\n", - "Test Accuracy of ship: 81% (811/1000)\n", - "Test Accuracy of truck: 65% (653/1000)\n", + "Test Accuracy of airplane: 61% (611/1000)\n", + "Test Accuracy of automobile: 84% (847/1000)\n", + "Test Accuracy of bird: 44% (441/1000)\n", + "Test Accuracy of cat: 45% (451/1000)\n", + "Test Accuracy of deer: 59% (598/1000)\n", + "Test Accuracy of dog: 53% (530/1000)\n", + "Test Accuracy of frog: 66% (662/1000)\n", + "Test Accuracy of horse: 69% (697/1000)\n", + "Test Accuracy of ship: 79% (795/1000)\n", + "Test Accuracy of truck: 63% (638/1000)\n", "\n", - "Test Accuracy (Overall): 62% (6208/10000)\n" + "Test Accuracy (Overall): 62% (6270/10000)\n" ] } ], @@ -639,7 +648,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 14, "metadata": {}, "outputs": [ { @@ -694,55 +703,58 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch: 0 \tTraining Loss: 45.642221 \tValidation Loss: 43.091903\n", - "Validation loss decreased (inf --> 43.091903). Saving model ...\n", - "Epoch: 1 \tTraining Loss: 39.640042 \tValidation Loss: 36.765509\n", - "Validation loss decreased (43.091903 --> 36.765509). Saving model ...\n", - "Epoch: 2 \tTraining Loss: 34.094025 \tValidation Loss: 32.752938\n", - "Validation loss decreased (36.765509 --> 32.752938). Saving model ...\n", - "Epoch: 3 \tTraining Loss: 31.503831 \tValidation Loss: 30.753025\n", - "Validation loss decreased (32.752938 --> 30.753025). Saving model ...\n", - "Epoch: 4 \tTraining Loss: 29.382825 \tValidation Loss: 28.668912\n", - "Validation loss decreased (30.753025 --> 28.668912). Saving model ...\n", - "Epoch: 5 \tTraining Loss: 27.670345 \tValidation Loss: 27.498096\n", - "Validation loss decreased (28.668912 --> 27.498096). Saving model ...\n", - "Epoch: 6 \tTraining Loss: 26.104155 \tValidation Loss: 25.743836\n", - "Validation loss decreased (27.498096 --> 25.743836). Saving model ...\n", - "Epoch: 7 \tTraining Loss: 24.752746 \tValidation Loss: 24.593233\n", - "Validation loss decreased (25.743836 --> 24.593233). Saving model ...\n", - "Epoch: 8 \tTraining Loss: 23.427782 \tValidation Loss: 24.164169\n", - "Validation loss decreased (24.593233 --> 24.164169). Saving model ...\n", - "Epoch: 9 \tTraining Loss: 22.255248 \tValidation Loss: 22.511518\n", - "Validation loss decreased (24.164169 --> 22.511518). Saving model ...\n", - "Epoch: 10 \tTraining Loss: 21.019939 \tValidation Loss: 21.948025\n", - "Validation loss decreased (22.511518 --> 21.948025). Saving model ...\n", - "Epoch: 11 \tTraining Loss: 19.894535 \tValidation Loss: 21.125246\n", - "Validation loss decreased (21.948025 --> 21.125246). Saving model ...\n", - "Epoch: 12 \tTraining Loss: 18.844355 \tValidation Loss: 19.916274\n", - "Validation loss decreased (21.125246 --> 19.916274). Saving model ...\n", - "Epoch: 13 \tTraining Loss: 17.964014 \tValidation Loss: 19.722094\n", - "Validation loss decreased (19.916274 --> 19.722094). Saving model ...\n", - "Epoch: 14 \tTraining Loss: 17.103886 \tValidation Loss: 19.501696\n", - "Validation loss decreased (19.722094 --> 19.501696). Saving model ...\n", - "Epoch: 15 \tTraining Loss: 16.276358 \tValidation Loss: 18.857704\n", - "Validation loss decreased (19.501696 --> 18.857704). Saving model ...\n", - "Epoch: 16 \tTraining Loss: 15.643052 \tValidation Loss: 18.830612\n", - "Validation loss decreased (18.857704 --> 18.830612). Saving model ...\n", - "Epoch: 17 \tTraining Loss: 14.828651 \tValidation Loss: 18.274699\n", - "Validation loss decreased (18.830612 --> 18.274699). Saving model ...\n", - "Epoch: 18 \tTraining Loss: 14.199379 \tValidation Loss: 18.204317\n", - "Validation loss decreased (18.274699 --> 18.204317). Saving model ...\n", - "Epoch: 19 \tTraining Loss: 13.636911 \tValidation Loss: 18.570810\n", - "Epoch: 20 \tTraining Loss: 12.955069 \tValidation Loss: 18.817341\n", - "Epoch: 21 \tTraining Loss: 12.413357 \tValidation Loss: 18.511569\n", - "Early stopping after 21 epochs.\n" + "Epoch: 0 \tTraining Loss: 45.369728 \tValidation Loss: 42.691277\n", + "Validation loss decreased (inf --> 42.691277). Saving model ...\n", + "Epoch: 1 \tTraining Loss: 38.824839 \tValidation Loss: 36.059104\n", + "Validation loss decreased (42.691277 --> 36.059104). Saving model ...\n", + "Epoch: 2 \tTraining Loss: 33.579168 \tValidation Loss: 32.372725\n", + "Validation loss decreased (36.059104 --> 32.372725). Saving model ...\n", + "Epoch: 3 \tTraining Loss: 30.625164 \tValidation Loss: 29.699576\n", + "Validation loss decreased (32.372725 --> 29.699576). Saving model ...\n", + "Epoch: 4 \tTraining Loss: 28.611405 \tValidation Loss: 27.687103\n", + "Validation loss decreased (29.699576 --> 27.687103). Saving model ...\n", + "Epoch: 5 \tTraining Loss: 26.902317 \tValidation Loss: 26.093671\n", + "Validation loss decreased (27.687103 --> 26.093671). Saving model ...\n", + "Epoch: 6 \tTraining Loss: 25.427897 \tValidation Loss: 25.468752\n", + "Validation loss decreased (26.093671 --> 25.468752). Saving model ...\n", + "Epoch: 7 \tTraining Loss: 24.026698 \tValidation Loss: 23.640172\n", + "Validation loss decreased (25.468752 --> 23.640172). Saving model ...\n", + "Epoch: 8 \tTraining Loss: 22.714501 \tValidation Loss: 22.726492\n", + "Validation loss decreased (23.640172 --> 22.726492). Saving model ...\n", + "Epoch: 9 \tTraining Loss: 21.534951 \tValidation Loss: 22.016264\n", + "Validation loss decreased (22.726492 --> 22.016264). Saving model ...\n", + "Epoch: 10 \tTraining Loss: 20.424777 \tValidation Loss: 21.124198\n", + "Validation loss decreased (22.016264 --> 21.124198). Saving model ...\n", + "Epoch: 11 \tTraining Loss: 19.326325 \tValidation Loss: 19.992738\n", + "Validation loss decreased (21.124198 --> 19.992738). Saving model ...\n", + "Epoch: 12 \tTraining Loss: 18.396354 \tValidation Loss: 19.536515\n", + "Validation loss decreased (19.992738 --> 19.536515). Saving model ...\n", + "Epoch: 13 \tTraining Loss: 17.590070 \tValidation Loss: 19.022870\n", + "Validation loss decreased (19.536515 --> 19.022870). Saving model ...\n", + "Epoch: 14 \tTraining Loss: 16.734188 \tValidation Loss: 18.731410\n", + "Validation loss decreased (19.022870 --> 18.731410). Saving model ...\n", + "Epoch: 15 \tTraining Loss: 15.948698 \tValidation Loss: 18.202837\n", + "Validation loss decreased (18.731410 --> 18.202837). Saving model ...\n", + "Epoch: 16 \tTraining Loss: 15.233767 \tValidation Loss: 18.607395\n", + "Epoch: 17 \tTraining Loss: 14.603055 \tValidation Loss: 18.331243\n", + "Epoch: 18 \tTraining Loss: 13.910214 \tValidation Loss: 18.034153\n", + "Validation loss decreased (18.202837 --> 18.034153). Saving model ...\n", + "Epoch: 19 \tTraining Loss: 13.272603 \tValidation Loss: 17.963906\n", + "Validation loss decreased (18.034153 --> 17.963906). Saving model ...\n", + "Epoch: 20 \tTraining Loss: 12.770533 \tValidation Loss: 18.075906\n", + "Epoch: 21 \tTraining Loss: 12.149368 \tValidation Loss: 17.510612\n", + "Validation loss decreased (17.963906 --> 17.510612). Saving model ...\n", + "Epoch: 22 \tTraining Loss: 11.595121 \tValidation Loss: 17.848336\n", + "Epoch: 23 \tTraining Loss: 11.122050 \tValidation Loss: 17.816858\n", + "Epoch: 24 \tTraining Loss: 10.528249 \tValidation Loss: 18.406580\n", + "Early stopping after 24 epochs.\n" ] } ], @@ -830,12 +842,12 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 16, "metadata": {}, "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "<Figure size 640x480 with 1 Axes>" ] @@ -855,27 +867,27 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Test Loss: 18.610740\n", + "Test Loss: 17.570148\n", "\n", - "Test Accuracy of airplane: 78% (788/1000)\n", - "Test Accuracy of automobile: 82% (828/1000)\n", - "Test Accuracy of bird: 49% (495/1000)\n", - "Test Accuracy of cat: 50% (504/1000)\n", - "Test Accuracy of deer: 62% (625/1000)\n", - "Test Accuracy of dog: 62% (624/1000)\n", - "Test Accuracy of frog: 65% (651/1000)\n", - "Test Accuracy of horse: 73% (739/1000)\n", - "Test Accuracy of ship: 80% (803/1000)\n", - "Test Accuracy of truck: 75% (758/1000)\n", + "Test Accuracy of airplane: 72% (720/1000)\n", + "Test Accuracy of automobile: 83% (836/1000)\n", + "Test Accuracy of bird: 63% (633/1000)\n", + "Test Accuracy of cat: 55% (551/1000)\n", + "Test Accuracy of deer: 69% (690/1000)\n", + "Test Accuracy of dog: 57% (570/1000)\n", + "Test Accuracy of frog: 80% (809/1000)\n", + "Test Accuracy of horse: 67% (675/1000)\n", + "Test Accuracy of ship: 85% (853/1000)\n", + "Test Accuracy of truck: 77% (770/1000)\n", "\n", - "Test Accuracy (Overall): 68% (6815/10000)\n" + "Test Accuracy (Overall): 71% (7107/10000)\n" ] } ], @@ -966,7 +978,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 18, "id": "ef623c26", "metadata": {}, "outputs": [ @@ -983,7 +995,7 @@ "2330946" ] }, - "execution_count": 26, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -1013,7 +1025,7 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 19, "id": "c4c65d4b", "metadata": {}, "outputs": [ @@ -1030,7 +1042,7 @@ "659678" ] }, - "execution_count": 27, + "execution_count": 19, "metadata": {}, "output_type": "execute_result" } @@ -1046,7 +1058,7 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 20, "metadata": {}, "outputs": [ { @@ -1060,20 +1072,20 @@ "name": "stdout", "output_type": "stream", "text": [ - "Test Loss: 18.704158\n", + "Test Loss: 17.560515\n", "\n", - "Test Accuracy of airplane: 77% (778/1000)\n", - "Test Accuracy of automobile: 84% (841/1000)\n", - "Test Accuracy of bird: 50% (500/1000)\n", - "Test Accuracy of cat: 49% (497/1000)\n", - "Test Accuracy of deer: 62% (626/1000)\n", - "Test Accuracy of dog: 62% (624/1000)\n", - "Test Accuracy of frog: 64% (647/1000)\n", - "Test Accuracy of horse: 75% (751/1000)\n", - "Test Accuracy of ship: 80% (805/1000)\n", + "Test Accuracy of airplane: 72% (728/1000)\n", + "Test Accuracy of automobile: 83% (834/1000)\n", + "Test Accuracy of bird: 62% (623/1000)\n", + "Test Accuracy of cat: 53% (535/1000)\n", + "Test Accuracy of deer: 68% (680/1000)\n", + "Test Accuracy of dog: 56% (564/1000)\n", + "Test Accuracy of frog: 78% (788/1000)\n", + "Test Accuracy of horse: 68% (687/1000)\n", + "Test Accuracy of ship: 84% (843/1000)\n", "Test Accuracy of truck: 77% (774/1000)\n", "\n", - "Test Accuracy (Overall): 68% (6843/10000)\n" + "Test Accuracy (Overall): 70% (7056/10000)\n" ] } ], @@ -1178,7 +1190,7 @@ }, { "cell_type": "code", - "execution_count": 48, + "execution_count": 21, "id": "b4d13080", "metadata": {}, "outputs": [ @@ -1186,8 +1198,6 @@ "name": "stderr", "output_type": "stream", "text": [ - "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", - " warnings.warn(\n", "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n", " warnings.warn(msg)\n" ] @@ -1277,19 +1287,9 @@ }, { "cell_type": "code", - "execution_count": 50, + "execution_count": 22, "metadata": {}, "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", - " warnings.warn(\n", - "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n", - " warnings.warn(msg)\n" - ] - }, { "name": "stdout", "output_type": "stream", @@ -1319,7 +1319,7 @@ }, { "cell_type": "code", - "execution_count": 53, + "execution_count": 23, "metadata": {}, "outputs": [ { @@ -1354,7 +1354,7 @@ }, { "cell_type": "code", - "execution_count": 54, + "execution_count": 24, "metadata": {}, "outputs": [ { @@ -1373,7 +1373,7 @@ }, { "cell_type": "code", - "execution_count": 55, + "execution_count": 25, "metadata": {}, "outputs": [ { @@ -1406,7 +1406,7 @@ }, { "cell_type": "code", - "execution_count": 56, + "execution_count": 26, "metadata": {}, "outputs": [ { @@ -1441,10 +1441,21 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 2, "id": "be2d31f5", "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "<Figure size 640x480 with 1 Axes>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "import os\n", "\n", @@ -1533,10 +1544,87 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 3, "id": "572d824c", "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Epoch 1/10\n", + "----------\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", + " warnings.warn(\n", + "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.\n", + " warnings.warn(msg)\n", + "/opt/homebrew/lib/python3.11/site-packages/torch/optim/lr_scheduler.py:136: UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`. In PyTorch 1.1.0 and later, you should call them in the opposite order: `optimizer.step()` before `lr_scheduler.step()`. Failure to do this will result in PyTorch skipping the first value of the learning rate schedule. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate\n", + " warnings.warn(\"Detected call of `lr_scheduler.step()` before `optimizer.step()`. \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "train Loss: 0.6145 Acc: 0.6885\n", + "val Loss: 0.2200 Acc: 0.9150\n", + "\n", + "Epoch 2/10\n", + "----------\n", + "train Loss: 0.4196 Acc: 0.8115\n", + "val Loss: 0.1769 Acc: 0.9412\n", + "\n", + "Epoch 3/10\n", + "----------\n", + "train Loss: 0.4180 Acc: 0.8074\n", + "val Loss: 0.2414 Acc: 0.9216\n", + "\n", + "Epoch 4/10\n", + "----------\n", + "train Loss: 0.4360 Acc: 0.7910\n", + "val Loss: 0.3016 Acc: 0.8758\n", + "\n", + "Epoch 5/10\n", + "----------\n", + "train Loss: 0.4552 Acc: 0.8033\n", + "val Loss: 0.1784 Acc: 0.9346\n", + "\n", + "Epoch 6/10\n", + "----------\n", + "train Loss: 0.5639 Acc: 0.7910\n", + "val Loss: 0.2022 Acc: 0.9216\n", + "\n", + "Epoch 7/10\n", + "----------\n", + "train Loss: 0.4463 Acc: 0.7992\n", + "val Loss: 0.1593 Acc: 0.9477\n", + "\n", + "Epoch 8/10\n", + "----------\n", + "train Loss: 0.4029 Acc: 0.8279\n", + "val Loss: 0.1541 Acc: 0.9412\n", + "\n", + "Epoch 9/10\n", + "----------\n", + "train Loss: 0.3753 Acc: 0.8361\n", + "val Loss: 0.1542 Acc: 0.9412\n", + "\n", + "Epoch 10/10\n", + "----------\n", + "train Loss: 0.3245 Acc: 0.8525\n", + "val Loss: 0.1474 Acc: 0.9477\n", + "\n", + "Training complete in 7m 52s\n", + "Best val Acc: 0.947712\n" + ] + } + ], "source": [ "import copy\n", "import os\n", @@ -1715,7 +1803,434 @@ "exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)\n", "model, epoch_time = train_model(\n", " model, criterion, optimizer_conv, exp_lr_scheduler, num_epochs=10\n", - ")\n" + ")\n", + "\n", + "torch.save(model.state_dict(), \"model_exo_4.pt\")" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/PIL/Image.py:992: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Validation Loss: 0.6253 Accuracy: 0.7649\n" + ] + }, + { + "data": { + "text/plain": [ + "(0.6252933401546583, tensor(0.7649, dtype=torch.float64))" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_transforms = {\n", + " \"test\": transforms.Compose(\n", + " [\n", + " transforms.Resize(256),\n", + " transforms.CenterCrop(224),\n", + " transforms.ToTensor(),\n", + " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),\n", + " ]\n", + " ),\n", + "}\n", + "\n", + "image_datasets = {\n", + " x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x])\n", + " for x in [\"test\"]\n", + "}\n", + "\n", + "dataloader = {\n", + " x: torch.utils.data.DataLoader(\n", + " image_datasets[x], batch_size=4, shuffle=True, num_workers=4\n", + " )\n", + " for x in [\"test\"]\n", + "}\n", + "\n", + "dataset_sizes = {x: len(image_datasets[x]) for x in [\"test\"]}\n", + "class_names = image_datasets[\"test\"].classes\n", + "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n", + "\n", + "def eval_model(model, dataloader, criterion):\n", + " model.eval() # Set the model to evaluation mode\n", + " \n", + " running_loss = 0.0\n", + " running_corrects = 0\n", + "\n", + " with torch.no_grad(): # No need to calculate gradients during evaluation\n", + " for inputs, labels in dataloader['test']:\n", + " inputs = inputs.to(device)\n", + " labels = labels.to(device)\n", + "\n", + " outputs = model(inputs)\n", + " _, preds = torch.max(outputs, 1)\n", + " loss = criterion(outputs, labels)\n", + "\n", + " running_loss += loss.item() * inputs.size(0)\n", + " running_corrects += torch.sum(preds == labels.data)\n", + "\n", + " # Calculate the loss and accuracy for the entire dataset\n", + " total_loss = running_loss / dataset_sizes['test']\n", + " accuracy = running_corrects.double() / dataset_sizes['test']\n", + "\n", + " print(\"Validation Loss: {:.4f} Accuracy: {:.4f}\".format(total_loss, accuracy))\n", + " return total_loss, accuracy\n", + " \n", + "\n", + "eval_model(model, dataloader, criterion)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Epoch 1/10\n", + "----------\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", + " warnings.warn(\n", + "/opt/homebrew/lib/python3.11/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.\n", + " warnings.warn(msg)\n", + "/opt/homebrew/lib/python3.11/site-packages/torch/optim/lr_scheduler.py:136: UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`. In PyTorch 1.1.0 and later, you should call them in the opposite order: `optimizer.step()` before `lr_scheduler.step()`. Failure to do this will result in PyTorch skipping the first value of the learning rate schedule. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate\n", + " warnings.warn(\"Detected call of `lr_scheduler.step()` before `optimizer.step()`. \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "train Loss: 0.6543 Acc: 0.5820\n", + "val Loss: 0.4400 Acc: 0.8562\n", + "\n", + "Epoch 2/10\n", + "----------\n", + "train Loss: 0.5217 Acc: 0.7705\n", + "val Loss: 0.3166 Acc: 0.8889\n", + "\n", + "Epoch 3/10\n", + "----------\n", + "train Loss: 0.4697 Acc: 0.8156\n", + "val Loss: 0.2591 Acc: 0.9150\n", + "\n", + "Epoch 4/10\n", + "----------\n", + "train Loss: 0.4282 Acc: 0.8115\n", + "val Loss: 0.2265 Acc: 0.9216\n", + "\n", + "Epoch 5/10\n", + "----------\n", + "train Loss: 0.4332 Acc: 0.8115\n", + "val Loss: 0.2711 Acc: 0.8889\n", + "\n", + "Epoch 6/10\n", + "----------\n", + "train Loss: 0.4745 Acc: 0.7705\n", + "val Loss: 0.2211 Acc: 0.9477\n", + "\n", + "Epoch 7/10\n", + "----------\n", + "train Loss: 0.3897 Acc: 0.8320\n", + "val Loss: 0.2169 Acc: 0.9412\n", + "\n", + "Epoch 8/10\n", + "----------\n", + "train Loss: 0.4254 Acc: 0.8074\n", + "val Loss: 0.2073 Acc: 0.9477\n", + "\n", + "Epoch 9/10\n", + "----------\n", + "train Loss: 0.3462 Acc: 0.8525\n", + "val Loss: 0.2176 Acc: 0.9412\n", + "\n", + "Epoch 10/10\n", + "----------\n", + "train Loss: 0.3830 Acc: 0.8156\n", + "val Loss: 0.1969 Acc: 0.9542\n", + "\n", + "Training complete in 7m 52s\n", + "Best val Acc: 0.954248\n" + ] + } + ], + "source": [ + "import copy\n", + "import os\n", + "import time\n", + "\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim\n", + "import torchvision\n", + "from torch.optim import lr_scheduler\n", + "from torchvision import datasets, transforms\n", + "\n", + "# Data augmentation and normalization for training\n", + "# Just normalization for validation\n", + "data_transforms = {\n", + " \"train\": transforms.Compose(\n", + " [\n", + " transforms.RandomResizedCrop(\n", + " 224\n", + " ), # ImageNet models were trained on 224x224 images\n", + " transforms.RandomHorizontalFlip(), # flip horizontally 50% of the time - increases train set variability\n", + " transforms.ToTensor(), # convert it to a PyTorch tensor\n", + " transforms.Normalize(\n", + " [0.485, 0.456, 0.406], [0.229, 0.224, 0.225]\n", + " ), # ImageNet models expect this norm\n", + " ]\n", + " ),\n", + " \"val\": transforms.Compose(\n", + " [\n", + " transforms.Resize(256),\n", + " transforms.CenterCrop(224),\n", + " transforms.ToTensor(),\n", + " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),\n", + " ]\n", + " ),\n", + "}\n", + "\n", + "data_dir = \"hymenoptera_data\"\n", + "# Create train and validation datasets and loaders\n", + "image_datasets = {\n", + " x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x])\n", + " for x in [\"train\", \"val\"]\n", + "}\n", + "dataloaders = {\n", + " x: torch.utils.data.DataLoader(\n", + " image_datasets[x], batch_size=4, shuffle=True, num_workers=4\n", + " )\n", + " for x in [\"train\", \"val\"]\n", + "}\n", + "dataset_sizes = {x: len(image_datasets[x]) for x in [\"train\", \"val\"]}\n", + "class_names = image_datasets[\"train\"].classes\n", + "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n", + "\n", + "# Helper function for displaying images\n", + "def imshow(inp, title=None):\n", + " \"\"\"Imshow for Tensor.\"\"\"\n", + " inp = inp.numpy().transpose((1, 2, 0))\n", + " mean = np.array([0.485, 0.456, 0.406])\n", + " std = np.array([0.229, 0.224, 0.225])\n", + "\n", + " # Un-normalize the images\n", + " inp = std * inp + mean\n", + " # Clip just in case\n", + " inp = np.clip(inp, 0, 1)\n", + " plt.imshow(inp)\n", + " if title is not None:\n", + " plt.title(title)\n", + " plt.pause(0.001) # pause a bit so that plots are updated\n", + " plt.show()\n", + "\n", + "\n", + "# Get a batch of training data\n", + "# inputs, classes = next(iter(dataloaders['train']))\n", + "\n", + "# Make a grid from batch\n", + "# out = torchvision.utils.make_grid(inputs)\n", + "\n", + "# imshow(out, title=[class_names[x] for x in classes])\n", + "# training\n", + "\n", + "\n", + "def train_model(model, criterion, optimizer, scheduler, num_epochs=25):\n", + " since = time.time()\n", + "\n", + " best_model_wts = copy.deepcopy(model.state_dict())\n", + " best_acc = 0.0\n", + "\n", + " epoch_time = [] # we'll keep track of the time needed for each epoch\n", + "\n", + " for epoch in range(num_epochs):\n", + " epoch_start = time.time()\n", + " print(\"Epoch {}/{}\".format(epoch + 1, num_epochs))\n", + " print(\"-\" * 10)\n", + "\n", + " # Each epoch has a training and validation phase\n", + " for phase in [\"train\", \"val\"]:\n", + " if phase == \"train\":\n", + " scheduler.step()\n", + " model.train() # Set model to training mode\n", + " else:\n", + " model.eval() # Set model to evaluate mode\n", + "\n", + " running_loss = 0.0\n", + " running_corrects = 0\n", + "\n", + " # Iterate over data.\n", + " for inputs, labels in dataloaders[phase]:\n", + " inputs = inputs.to(device)\n", + " labels = labels.to(device)\n", + "\n", + " # zero the parameter gradients\n", + " optimizer.zero_grad()\n", + "\n", + " # Forward\n", + " # Track history if only in training phase\n", + " with torch.set_grad_enabled(phase == \"train\"):\n", + " outputs = model(inputs)\n", + " _, preds = torch.max(outputs, 1)\n", + " loss = criterion(outputs, labels)\n", + "\n", + " # backward + optimize only if in training phase\n", + " if phase == \"train\":\n", + " loss.backward()\n", + " optimizer.step()\n", + "\n", + " # Statistics\n", + " running_loss += loss.item() * inputs.size(0)\n", + " running_corrects += torch.sum(preds == labels.data)\n", + "\n", + " epoch_loss = running_loss / dataset_sizes[phase]\n", + " epoch_acc = running_corrects.double() / dataset_sizes[phase]\n", + "\n", + " print(\"{} Loss: {:.4f} Acc: {:.4f}\".format(phase, epoch_loss, epoch_acc))\n", + "\n", + " # Deep copy the model\n", + " if phase == \"val\" and epoch_acc > best_acc:\n", + " best_acc = epoch_acc\n", + " best_model_wts = copy.deepcopy(model.state_dict())\n", + "\n", + " # Add the epoch time\n", + " t_epoch = time.time() - epoch_start\n", + " epoch_time.append(t_epoch)\n", + " print()\n", + "\n", + " time_elapsed = time.time() - since\n", + " print(\n", + " \"Training complete in {:.0f}m {:.0f}s\".format(\n", + " time_elapsed // 60, time_elapsed % 60\n", + " )\n", + " )\n", + " print(\"Best val Acc: {:4f}\".format(best_acc))\n", + "\n", + " # Load best model weights\n", + " model.load_state_dict(best_model_wts)\n", + " return model, epoch_time\n", + "\n", + "\n", + "# Download a pre-trained ResNet18 model and freeze its weights\n", + "model = torchvision.models.resnet18(pretrained=True)\n", + "for param in model.parameters():\n", + " param.requires_grad = False\n", + "\n", + "# Replace the final fully connected layer\n", + "# Parameters of newly constructed modules have requires_grad=True by default\n", + "num_ftrs = model.fc.in_features\n", + "\n", + "\n", + "model.fc = nn.Sequential(\n", + " nn.Linear(num_ftrs, 256), # Add a hidden layer with 256 units\n", + " nn.ReLU(), # Apply ReLU activation\n", + " nn.Dropout(0.5), # Apply dropout with a probability of 0.5\n", + " nn.Linear(256, 2), # Add the final output layer with 2 units\n", + ")\n", + "# Send the model to the GPU\n", + "model = model.to(device)\n", + "# Set the loss function\n", + "criterion = nn.CrossEntropyLoss()\n", + "\n", + "# Observe that only the parameters of the final layer are being optimized\n", + "optimizer_conv = optim.SGD(model.fc.parameters(), lr=0.001, momentum=0.9)\n", + "exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)\n", + "model, epoch_time = train_model(\n", + " model, criterion, optimizer_conv, exp_lr_scheduler, num_epochs=10\n", + ")\n", + "\n", + "torch.save(model.state_dict(), \"model_exo_4_relu.pt\")" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/PIL/Image.py:992: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n", + " warnings.warn(\n", + "/opt/homebrew/lib/python3.11/site-packages/PIL/Image.py:992: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Validation Loss: 0.3294 Accuracy: 0.8344\n" + ] + }, + { + "data": { + "text/plain": [ + "(0.32943220670945594, tensor(0.8344, dtype=torch.float64))" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_transforms = {\n", + " \"test\": transforms.Compose(\n", + " [\n", + " transforms.Resize(256),\n", + " transforms.CenterCrop(224),\n", + " transforms.ToTensor(),\n", + " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),\n", + " ]\n", + " ),\n", + "}\n", + "\n", + "image_datasets = {\n", + " x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x])\n", + " for x in [\"test\"]\n", + "}\n", + "\n", + "dataloader = {\n", + " x: torch.utils.data.DataLoader(\n", + " image_datasets[x], batch_size=4, shuffle=True, num_workers=4\n", + " )\n", + " for x in [\"test\"]\n", + "}\n", + "\n", + "dataset_sizes = {x: len(image_datasets[x]) for x in [\"test\"]}\n", + "class_names = image_datasets[\"test\"].classes\n", + "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n", + " \n", + "\n", + "eval_model(model, dataloader, criterion)" ] }, {