From 341baa9db2415a0465eb18ee63ef76a6d9c3740b Mon Sep 17 00:00:00 2001
From: clemencegirard <clemgirard91@gmail.com>
Date: Fri, 29 Nov 2024 21:02:54 +0100
Subject: [PATCH] Adapt test accuracy to new trest dataset
modify the batchsize, data loaders, classes, ... compare dense and quantized models test accuracies. Function for repetitive operations (dense model accuracy, quantization accuracy)
---
TD2_Deep_Learning-2.ipynb | 1551 ++++++++++++++++++++++---------------
1 file changed, 919 insertions(+), 632 deletions(-)
diff --git a/TD2_Deep_Learning-2.ipynb b/TD2_Deep_Learning-2.ipynb
index 7fe9af3..728e772 100644
--- a/TD2_Deep_Learning-2.ipynb
+++ b/TD2_Deep_Learning-2.ipynb
@@ -62,7 +62,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": null,
"id": "b1950f0a",
"metadata": {
"id": "b1950f0a",
@@ -177,21 +177,21 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 27,
"id": "6e18f2fd",
"metadata": {
"id": "6e18f2fd",
"colab": {
"base_uri": "https://localhost:8080/"
},
- "outputId": "8345a666-e787-499a-8861-30cd663ba3c0"
+ "outputId": "4910a198-975a-4839-b0c9-7e3196cba297"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
- "CUDA is available! Training on GPU ...\n"
+ "CUDA is not available. Training on CPU ...\n"
]
}
],
@@ -219,14 +219,14 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 10,
"id": "462666a2",
"metadata": {
"id": "462666a2",
"colab": {
"base_uri": "https://localhost:8080/"
},
- "outputId": "a63aa205-2bce-46c1-97c3-8cd9a2977bbb"
+ "outputId": "d3520ce3-43a9-4037-8b88-17f171e51f82"
},
"outputs": [
{
@@ -240,7 +240,7 @@
"output_type": "stream",
"name": "stderr",
"text": [
- "100%|██████████| 170M/170M [00:14<00:00, 12.0MB/s]\n"
+ "100%|██████████| 170M/170M [00:05<00:00, 29.0MB/s]\n"
]
},
{
@@ -322,14 +322,14 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": null,
"id": "317bf070",
"metadata": {
"id": "317bf070",
"colab": {
"base_uri": "https://localhost:8080/"
},
- "outputId": "0c3a0a99-8c7c-49bf-a2b2-57212a45b6a1"
+ "outputId": "0ef89fb4-4845-4ce2-b2e7-ae7b982cd2d0"
},
"outputs": [
{
@@ -394,64 +394,63 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": null,
"id": "4b53f229",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "4b53f229",
- "outputId": "352fa2f2-4cf7-42b6-e51a-c68bf428e90c"
+ "outputId": "a9782921-b4b8-45e2-a81d-87203d361e09"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
- "Epoch: 0 \tTraining Loss: 35.027589 \tValidation Loss: 31.574753\n",
- "Validation loss decreased (inf --> 31.574753). Saving model ...\n",
- "Epoch: 1 \tTraining Loss: 30.186101 \tValidation Loss: 29.019872\n",
- "Validation loss decreased (31.574753 --> 29.019872). Saving model ...\n",
- "Epoch: 2 \tTraining Loss: 27.864272 \tValidation Loss: 27.127275\n",
- "Validation loss decreased (29.019872 --> 27.127275). Saving model ...\n",
- "Epoch: 3 \tTraining Loss: 26.258431 \tValidation Loss: 26.395611\n",
- "Validation loss decreased (27.127275 --> 26.395611). Saving model ...\n",
- "Epoch: 4 \tTraining Loss: 24.996900 \tValidation Loss: 25.757890\n",
- "Validation loss decreased (26.395611 --> 25.757890). Saving model ...\n",
- "Epoch: 5 \tTraining Loss: 23.965784 \tValidation Loss: 25.668354\n",
- "Validation loss decreased (25.757890 --> 25.668354). Saving model ...\n",
- "Epoch: 6 \tTraining Loss: 23.020208 \tValidation Loss: 24.161515\n",
- "Validation loss decreased (25.668354 --> 24.161515). Saving model ...\n",
- "Epoch: 7 \tTraining Loss: 22.237902 \tValidation Loss: 23.424892\n",
- "Validation loss decreased (24.161515 --> 23.424892). Saving model ...\n",
- "Epoch: 8 \tTraining Loss: 21.415062 \tValidation Loss: 23.035766\n",
- "Validation loss decreased (23.424892 --> 23.035766). Saving model ...\n",
- "Epoch: 9 \tTraining Loss: 20.656158 \tValidation Loss: 22.670803\n",
- "Validation loss decreased (23.035766 --> 22.670803). Saving model ...\n",
- "Epoch: 10 \tTraining Loss: 19.973196 \tValidation Loss: 23.388851\n",
- "Epoch: 11 \tTraining Loss: 19.298960 \tValidation Loss: 22.264116\n",
- "Validation loss decreased (22.670803 --> 22.264116). Saving model ...\n",
- "Epoch: 12 \tTraining Loss: 18.639930 \tValidation Loss: 22.163565\n",
- "Validation loss decreased (22.264116 --> 22.163565). Saving model ...\n",
- "Epoch: 13 \tTraining Loss: 18.014738 \tValidation Loss: 22.538633\n",
- "Epoch: 14 \tTraining Loss: 17.396607 \tValidation Loss: 23.147467\n",
- "Epoch: 15 \tTraining Loss: 16.883292 \tValidation Loss: 22.109338\n",
- "Validation loss decreased (22.163565 --> 22.109338). Saving model ...\n",
- "Epoch: 16 \tTraining Loss: 16.285820 \tValidation Loss: 21.778137\n",
- "Validation loss decreased (22.109338 --> 21.778137). Saving model ...\n",
- "Epoch: 17 \tTraining Loss: 15.763043 \tValidation Loss: 22.761066\n",
- "Epoch: 18 \tTraining Loss: 15.239761 \tValidation Loss: 23.015333\n",
- "Epoch: 19 \tTraining Loss: 14.686504 \tValidation Loss: 24.884198\n",
- "Epoch: 20 \tTraining Loss: 14.331884 \tValidation Loss: 23.479585\n",
- "Epoch: 21 \tTraining Loss: 13.738545 \tValidation Loss: 23.654407\n",
- "Epoch: 22 \tTraining Loss: 13.280116 \tValidation Loss: 24.813238\n",
- "Epoch: 23 \tTraining Loss: 12.799162 \tValidation Loss: 24.736943\n",
- "Epoch: 24 \tTraining Loss: 12.397551 \tValidation Loss: 25.655078\n",
- "Epoch: 25 \tTraining Loss: 11.895732 \tValidation Loss: 26.784175\n",
- "Epoch: 26 \tTraining Loss: 11.552095 \tValidation Loss: 26.599335\n",
- "Epoch: 27 \tTraining Loss: 11.088106 \tValidation Loss: 27.633717\n",
- "Epoch: 28 \tTraining Loss: 10.653680 \tValidation Loss: 27.694947\n",
- "Epoch: 29 \tTraining Loss: 10.327371 \tValidation Loss: 27.795613\n"
+ "Epoch: 0 \tTraining Loss: 43.564449 \tValidation Loss: 39.401691\n",
+ "Validation loss decreased (inf --> 39.401691). Saving model ...\n",
+ "Epoch: 1 \tTraining Loss: 36.429993 \tValidation Loss: 33.232551\n",
+ "Validation loss decreased (39.401691 --> 33.232551). Saving model ...\n",
+ "Epoch: 2 \tTraining Loss: 31.587478 \tValidation Loss: 29.693407\n",
+ "Validation loss decreased (33.232551 --> 29.693407). Saving model ...\n",
+ "Epoch: 3 \tTraining Loss: 28.931881 \tValidation Loss: 28.926743\n",
+ "Validation loss decreased (29.693407 --> 28.926743). Saving model ...\n",
+ "Epoch: 4 \tTraining Loss: 27.116493 \tValidation Loss: 26.553065\n",
+ "Validation loss decreased (28.926743 --> 26.553065). Saving model ...\n",
+ "Epoch: 5 \tTraining Loss: 25.753600 \tValidation Loss: 26.277814\n",
+ "Validation loss decreased (26.553065 --> 26.277814). Saving model ...\n",
+ "Epoch: 6 \tTraining Loss: 24.513149 \tValidation Loss: 24.396103\n",
+ "Validation loss decreased (26.277814 --> 24.396103). Saving model ...\n",
+ "Epoch: 7 \tTraining Loss: 23.497829 \tValidation Loss: 24.289655\n",
+ "Validation loss decreased (24.396103 --> 24.289655). Saving model ...\n",
+ "Epoch: 8 \tTraining Loss: 22.587542 \tValidation Loss: 23.743511\n",
+ "Validation loss decreased (24.289655 --> 23.743511). Saving model ...\n",
+ "Epoch: 9 \tTraining Loss: 21.761979 \tValidation Loss: 23.418204\n",
+ "Validation loss decreased (23.743511 --> 23.418204). Saving model ...\n",
+ "Epoch: 10 \tTraining Loss: 20.961232 \tValidation Loss: 22.665598\n",
+ "Validation loss decreased (23.418204 --> 22.665598). Saving model ...\n",
+ "Epoch: 11 \tTraining Loss: 20.214310 \tValidation Loss: 22.860846\n",
+ "Epoch: 12 \tTraining Loss: 19.528810 \tValidation Loss: 22.310116\n",
+ "Validation loss decreased (22.665598 --> 22.310116). Saving model ...\n",
+ "Epoch: 13 \tTraining Loss: 18.852388 \tValidation Loss: 21.724193\n",
+ "Validation loss decreased (22.310116 --> 21.724193). Saving model ...\n",
+ "Epoch: 14 \tTraining Loss: 18.205391 \tValidation Loss: 21.975597\n",
+ "Epoch: 15 \tTraining Loss: 17.702034 \tValidation Loss: 21.844710\n",
+ "Epoch: 16 \tTraining Loss: 17.032942 \tValidation Loss: 21.959283\n",
+ "Epoch: 17 \tTraining Loss: 16.515157 \tValidation Loss: 21.953856\n",
+ "Epoch: 18 \tTraining Loss: 15.985784 \tValidation Loss: 22.567592\n",
+ "Epoch: 19 \tTraining Loss: 15.489598 \tValidation Loss: 22.483113\n",
+ "Epoch: 20 \tTraining Loss: 15.024780 \tValidation Loss: 23.352970\n",
+ "Epoch: 21 \tTraining Loss: 14.468246 \tValidation Loss: 23.186520\n",
+ "Epoch: 22 \tTraining Loss: 13.962025 \tValidation Loss: 23.276291\n",
+ "Epoch: 23 \tTraining Loss: 13.585494 \tValidation Loss: 23.123791\n",
+ "Epoch: 24 \tTraining Loss: 13.055498 \tValidation Loss: 24.466580\n",
+ "Epoch: 25 \tTraining Loss: 12.674296 \tValidation Loss: 25.248112\n",
+ "Epoch: 26 \tTraining Loss: 12.264263 \tValidation Loss: 24.854419\n",
+ "Epoch: 27 \tTraining Loss: 11.821320 \tValidation Loss: 25.723901\n",
+ "Epoch: 28 \tTraining Loss: 11.409582 \tValidation Loss: 25.295074\n",
+ "Epoch: 29 \tTraining Loss: 11.061590 \tValidation Loss: 26.113249\n"
]
}
],
@@ -537,7 +536,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": null,
"id": "d39df818",
"metadata": {
"colab": {
@@ -545,7 +544,7 @@
"height": 472
},
"id": "d39df818",
- "outputId": "8deb6712-7903-4899-9709-9be0e022e600"
+ "outputId": "f1763d5e-9b89-45ef-fc74-6010d6a143c0"
},
"outputs": [
{
@@ -554,7 +553,7 @@
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
],
- "image/png": "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\n"
+ "image/png": "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\n"
},
"metadata": {}
}
@@ -581,21 +580,25 @@
},
{
"cell_type": "code",
- "execution_count": 21,
- "id": "e93efdfc",
+ "source": [
+ "model.load_state_dict(torch.load(\"/content/save_data/model_cifar.pt\"))\n",
+ "dense_model_accuracy(model, test_loader, batch_size = 20)"
+ ],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
- "id": "e93efdfc",
- "outputId": "25166486-06b8-4855-fc5d-a70f4253ee31"
+ "id": "Iv1EUswaDBY1",
+ "outputId": "bdd78e9e-6b0c-49da-f7e8-1e938833af61"
},
+ "id": "Iv1EUswaDBY1",
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
- "<ipython-input-21-022d8c474780>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
+ "<ipython-input-18-79bff80cfa3e>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
" model.load_state_dict(torch.load(\"/content/save_data/model_cifar.pt\"))\n"
]
},
@@ -603,84 +606,22 @@
"output_type": "stream",
"name": "stdout",
"text": [
- "Test Loss: 22.087782\n",
+ "Test Loss: 21.741020\n",
"\n",
- "Test Accuracy of airplane: 64% (640/1000)\n",
- "Test Accuracy of automobile: 74% (741/1000)\n",
- "Test Accuracy of bird: 52% (528/1000)\n",
- "Test Accuracy of cat: 45% (455/1000)\n",
- "Test Accuracy of deer: 53% (539/1000)\n",
- "Test Accuracy of dog: 48% (482/1000)\n",
- "Test Accuracy of frog: 76% (766/1000)\n",
- "Test Accuracy of horse: 69% (693/1000)\n",
- "Test Accuracy of ship: 72% (729/1000)\n",
- "Test Accuracy of truck: 67% (679/1000)\n",
+ "Test Accuracy of airplane: 66% (667/1000)\n",
+ "Test Accuracy of automobile: 68% (688/1000)\n",
+ "Test Accuracy of bird: 49% (496/1000)\n",
+ "Test Accuracy of cat: 29% (296/1000)\n",
+ "Test Accuracy of deer: 45% (458/1000)\n",
+ "Test Accuracy of dog: 69% (694/1000)\n",
+ "Test Accuracy of frog: 67% (673/1000)\n",
+ "Test Accuracy of horse: 74% (748/1000)\n",
+ "Test Accuracy of ship: 79% (794/1000)\n",
+ "Test Accuracy of truck: 68% (684/1000)\n",
"\n",
- "Test Accuracy (Overall): 62% (6252/10000)\n"
+ "Test Accuracy (Overall): 61% (6198/10000)\n"
]
}
- ],
- "source": [
- "model.load_state_dict(torch.load(\"/content/save_data/model_cifar.pt\"))\n",
- "\n",
- "# track test loss\n",
- "test_loss = 0.0\n",
- "class_correct = list(0.0 for i in range(10))\n",
- "class_total = list(0.0 for i in range(10))\n",
- "\n",
- "model.eval()\n",
- "# iterate over test data\n",
- "for data, target in test_loader:\n",
- " # move tensors to GPU if CUDA is available\n",
- " if train_on_gpu:\n",
- " data, target = data.cuda(), target.cuda()\n",
- " # forward pass: compute predicted outputs by passing inputs to the model\n",
- " output = model(data)\n",
- " # calculate the batch loss\n",
- " loss = criterion(output, target)\n",
- " # update test loss\n",
- " test_loss += loss.item() * data.size(0)\n",
- " # convert output probabilities to predicted class\n",
- " _, pred = torch.max(output, 1)\n",
- " # compare predictions to true label\n",
- " correct_tensor = pred.eq(target.data.view_as(pred))\n",
- " correct = (\n",
- " np.squeeze(correct_tensor.numpy())\n",
- " if not train_on_gpu\n",
- " else np.squeeze(correct_tensor.cpu().numpy())\n",
- " )\n",
- " # calculate test accuracy for each object class\n",
- " for i in range(batch_size):\n",
- " label = target.data[i]\n",
- " class_correct[label] += correct[i].item()\n",
- " class_total[label] += 1\n",
- "\n",
- "# average test loss\n",
- "test_loss = test_loss / len(test_loader)\n",
- "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
- "\n",
- "for i in range(10):\n",
- " if class_total[i] > 0:\n",
- " print(\n",
- " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " classes[i],\n",
- " 100 * class_correct[i] / class_total[i],\n",
- " np.sum(class_correct[i]),\n",
- " np.sum(class_total[i]),\n",
- " )\n",
- " )\n",
- " else:\n",
- " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
- "\n",
- "print(\n",
- " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
- " np.sum(class_correct),\n",
- " np.sum(class_total),\n",
- " )\n",
- ")"
]
},
{
@@ -704,14 +645,14 @@
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": null,
"id": "2DvrdR_nsGqq",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "2DvrdR_nsGqq",
- "outputId": "1648c077-ba3f-4f7a-9a33-2d5777092e10"
+ "outputId": "40790089-efa7-4d25-dcdd-e831f3d6b2b0"
},
"outputs": [
{
@@ -772,69 +713,73 @@
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": null,
"id": "IJz2Q9T25Qc3",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "IJz2Q9T25Qc3",
- "outputId": "121b9032-8481-4be3-e542-2ece8d8dbe2a"
+ "outputId": "16c9fabc-39e4-4a4f-815c-c7eca8ea1146"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
- "Epoch: 0 \tTraining Loss: 45.904219 \tValidation Loss: 44.942664\n",
- "Validation loss decreased (inf --> 44.942664). Saving model ...\n",
- "Epoch: 1 \tTraining Loss: 42.813407 \tValidation Loss: 38.436903\n",
- "Validation loss decreased (44.942664 --> 38.436903). Saving model ...\n",
- "Epoch: 2 \tTraining Loss: 39.437505 \tValidation Loss: 34.759394\n",
- "Validation loss decreased (38.436903 --> 34.759394). Saving model ...\n",
- "Epoch: 3 \tTraining Loss: 37.662391 \tValidation Loss: 32.064202\n",
- "Validation loss decreased (34.759394 --> 32.064202). Saving model ...\n",
- "Epoch: 4 \tTraining Loss: 36.459317 \tValidation Loss: 31.174973\n",
- "Validation loss decreased (32.064202 --> 31.174973). Saving model ...\n",
- "Epoch: 5 \tTraining Loss: 35.285123 \tValidation Loss: 28.916658\n",
- "Validation loss decreased (31.174973 --> 28.916658). Saving model ...\n",
- "Epoch: 6 \tTraining Loss: 34.439550 \tValidation Loss: 27.563518\n",
- "Validation loss decreased (28.916658 --> 27.563518). Saving model ...\n",
- "Epoch: 7 \tTraining Loss: 33.610940 \tValidation Loss: 26.579835\n",
- "Validation loss decreased (27.563518 --> 26.579835). Saving model ...\n",
- "Epoch: 8 \tTraining Loss: 32.717580 \tValidation Loss: 25.121927\n",
- "Validation loss decreased (26.579835 --> 25.121927). Saving model ...\n",
- "Epoch: 9 \tTraining Loss: 31.905730 \tValidation Loss: 24.054429\n",
- "Validation loss decreased (25.121927 --> 24.054429). Saving model ...\n",
- "Epoch: 10 \tTraining Loss: 31.197301 \tValidation Loss: 22.744985\n",
- "Validation loss decreased (24.054429 --> 22.744985). Saving model ...\n",
- "Epoch: 11 \tTraining Loss: 30.428635 \tValidation Loss: 22.388091\n",
- "Validation loss decreased (22.744985 --> 22.388091). Saving model ...\n",
- "Epoch: 12 \tTraining Loss: 29.790207 \tValidation Loss: 21.463255\n",
- "Validation loss decreased (22.388091 --> 21.463255). Saving model ...\n",
- "Epoch: 13 \tTraining Loss: 29.057292 \tValidation Loss: 20.476804\n",
- "Validation loss decreased (21.463255 --> 20.476804). Saving model ...\n",
- "Epoch: 14 \tTraining Loss: 28.433569 \tValidation Loss: 20.292996\n",
- "Validation loss decreased (20.476804 --> 20.292996). Saving model ...\n",
- "Epoch: 15 \tTraining Loss: 27.846333 \tValidation Loss: 19.485786\n",
- "Validation loss decreased (20.292996 --> 19.485786). Saving model ...\n",
- "Epoch: 16 \tTraining Loss: 27.332442 \tValidation Loss: 19.583211\n",
- "Epoch: 17 \tTraining Loss: 26.768378 \tValidation Loss: 18.470492\n",
- "Validation loss decreased (19.485786 --> 18.470492). Saving model ...\n",
- "Epoch: 18 \tTraining Loss: 26.166088 \tValidation Loss: 18.522611\n",
- "Epoch: 19 \tTraining Loss: 25.595460 \tValidation Loss: 17.577921\n",
- "Validation loss decreased (18.470492 --> 17.577921). Saving model ...\n",
- "Epoch: 20 \tTraining Loss: 25.090311 \tValidation Loss: 17.772856\n",
- "Epoch: 21 \tTraining Loss: 24.457410 \tValidation Loss: 17.996507\n",
- "Epoch: 22 \tTraining Loss: 24.041111 \tValidation Loss: 17.588359\n",
- "Epoch: 23 \tTraining Loss: 23.373427 \tValidation Loss: 17.087331\n",
- "Validation loss decreased (17.577921 --> 17.087331). Saving model ...\n",
- "Epoch: 24 \tTraining Loss: 22.925357 \tValidation Loss: 17.181599\n",
- "Epoch: 25 \tTraining Loss: 22.313664 \tValidation Loss: 17.449139\n",
- "Epoch: 26 \tTraining Loss: 21.707366 \tValidation Loss: 17.998827\n",
- "Epoch: 27 \tTraining Loss: 21.505335 \tValidation Loss: 17.853451\n",
- "Epoch: 28 \tTraining Loss: 20.992254 \tValidation Loss: 17.821120\n",
- "Epoch: 29 \tTraining Loss: 20.592966 \tValidation Loss: 19.540560\n"
+ "Epoch: 0 \tTraining Loss: 45.979220 \tValidation Loss: 45.615242\n",
+ "Validation loss decreased (inf --> 45.615242). Saving model ...\n",
+ "Epoch: 1 \tTraining Loss: 43.284407 \tValidation Loss: 39.614000\n",
+ "Validation loss decreased (45.615242 --> 39.614000). Saving model ...\n",
+ "Epoch: 2 \tTraining Loss: 40.157336 \tValidation Loss: 35.868401\n",
+ "Validation loss decreased (39.614000 --> 35.868401). Saving model ...\n",
+ "Epoch: 3 \tTraining Loss: 38.141107 \tValidation Loss: 32.897560\n",
+ "Validation loss decreased (35.868401 --> 32.897560). Saving model ...\n",
+ "Epoch: 4 \tTraining Loss: 36.687158 \tValidation Loss: 30.790835\n",
+ "Validation loss decreased (32.897560 --> 30.790835). Saving model ...\n",
+ "Epoch: 5 \tTraining Loss: 35.596631 \tValidation Loss: 28.732090\n",
+ "Validation loss decreased (30.790835 --> 28.732090). Saving model ...\n",
+ "Epoch: 6 \tTraining Loss: 34.588117 \tValidation Loss: 28.290344\n",
+ "Validation loss decreased (28.732090 --> 28.290344). Saving model ...\n",
+ "Epoch: 7 \tTraining Loss: 33.695167 \tValidation Loss: 26.792482\n",
+ "Validation loss decreased (28.290344 --> 26.792482). Saving model ...\n",
+ "Epoch: 8 \tTraining Loss: 32.731687 \tValidation Loss: 25.169201\n",
+ "Validation loss decreased (26.792482 --> 25.169201). Saving model ...\n",
+ "Epoch: 9 \tTraining Loss: 31.809370 \tValidation Loss: 23.524243\n",
+ "Validation loss decreased (25.169201 --> 23.524243). Saving model ...\n",
+ "Epoch: 10 \tTraining Loss: 31.105776 \tValidation Loss: 22.690063\n",
+ "Validation loss decreased (23.524243 --> 22.690063). Saving model ...\n",
+ "Epoch: 11 \tTraining Loss: 30.263997 \tValidation Loss: 21.467096\n",
+ "Validation loss decreased (22.690063 --> 21.467096). Saving model ...\n",
+ "Epoch: 12 \tTraining Loss: 29.827772 \tValidation Loss: 21.264692\n",
+ "Validation loss decreased (21.467096 --> 21.264692). Saving model ...\n",
+ "Epoch: 13 \tTraining Loss: 28.822413 \tValidation Loss: 20.608588\n",
+ "Validation loss decreased (21.264692 --> 20.608588). Saving model ...\n",
+ "Epoch: 14 \tTraining Loss: 28.462034 \tValidation Loss: 19.394476\n",
+ "Validation loss decreased (20.608588 --> 19.394476). Saving model ...\n",
+ "Epoch: 15 \tTraining Loss: 27.697039 \tValidation Loss: 18.981446\n",
+ "Validation loss decreased (19.394476 --> 18.981446). Saving model ...\n",
+ "Epoch: 16 \tTraining Loss: 27.147811 \tValidation Loss: 18.569795\n",
+ "Validation loss decreased (18.981446 --> 18.569795). Saving model ...\n",
+ "Epoch: 17 \tTraining Loss: 26.746548 \tValidation Loss: 17.558674\n",
+ "Validation loss decreased (18.569795 --> 17.558674). Saving model ...\n",
+ "Epoch: 18 \tTraining Loss: 26.041856 \tValidation Loss: 16.911882\n",
+ "Validation loss decreased (17.558674 --> 16.911882). Saving model ...\n",
+ "Epoch: 19 \tTraining Loss: 25.422305 \tValidation Loss: 17.014422\n",
+ "Epoch: 20 \tTraining Loss: 25.027868 \tValidation Loss: 16.734118\n",
+ "Validation loss decreased (16.911882 --> 16.734118). Saving model ...\n",
+ "Epoch: 21 \tTraining Loss: 24.462839 \tValidation Loss: 16.592046\n",
+ "Validation loss decreased (16.734118 --> 16.592046). Saving model ...\n",
+ "Epoch: 22 \tTraining Loss: 23.893791 \tValidation Loss: 16.322902\n",
+ "Validation loss decreased (16.592046 --> 16.322902). Saving model ...\n",
+ "Epoch: 23 \tTraining Loss: 23.434180 \tValidation Loss: 15.947358\n",
+ "Validation loss decreased (16.322902 --> 15.947358). Saving model ...\n",
+ "Epoch: 24 \tTraining Loss: 22.923004 \tValidation Loss: 16.571980\n",
+ "Epoch: 25 \tTraining Loss: 22.375460 \tValidation Loss: 16.282680\n",
+ "Epoch: 26 \tTraining Loss: 21.815422 \tValidation Loss: 16.108936\n",
+ "Epoch: 27 \tTraining Loss: 21.441256 \tValidation Loss: 16.247329\n",
+ "Epoch: 28 \tTraining Loss: 20.928872 \tValidation Loss: 16.887838\n",
+ "Epoch: 29 \tTraining Loss: 20.562156 \tValidation Loss: 17.556099\n"
]
}
],
@@ -918,7 +863,7 @@
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": null,
"id": "aQVARMhv7y1b",
"metadata": {
"colab": {
@@ -952,21 +897,25 @@
},
{
"cell_type": "code",
- "execution_count": 25,
- "id": "06j_Dr6475Kb",
+ "source": [
+ "model2.load_state_dict(torch.load(\"/content/save_data/model2_cifar.pt\"))\n",
+ "dense_model_accuracy(model2, test_loader, batch_size = 20)"
+ ],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
- "id": "06j_Dr6475Kb",
- "outputId": "a359313d-70bb-43e0-918c-309684e43037"
+ "id": "NJeeTp4FIAq3",
+ "outputId": "7f55c7d0-49c1-40d5-cf21-aea2b9a15eb1"
},
+ "id": "NJeeTp4FIAq3",
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
- "<ipython-input-25-ee85c886ffd9>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
+ "<ipython-input-19-64160196089f>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
" model2.load_state_dict(torch.load(\"/content/save_data/model2_cifar.pt\"))\n"
]
},
@@ -974,84 +923,22 @@
"output_type": "stream",
"name": "stdout",
"text": [
- "Test Loss: 17.172099\n",
+ "Test Loss: 16.293813\n",
"\n",
- "Test Accuracy of airplane: 76% (765/1000)\n",
- "Test Accuracy of automobile: 77% (776/1000)\n",
- "Test Accuracy of bird: 58% (588/1000)\n",
- "Test Accuracy of cat: 57% (573/1000)\n",
- "Test Accuracy of deer: 65% (650/1000)\n",
- "Test Accuracy of dog: 53% (530/1000)\n",
- "Test Accuracy of frog: 78% (783/1000)\n",
- "Test Accuracy of horse: 78% (782/1000)\n",
- "Test Accuracy of ship: 82% (828/1000)\n",
- "Test Accuracy of truck: 79% (790/1000)\n",
+ "Test Accuracy of airplane: 78% (785/1000)\n",
+ "Test Accuracy of automobile: 86% (867/1000)\n",
+ "Test Accuracy of bird: 57% (577/1000)\n",
+ "Test Accuracy of cat: 48% (485/1000)\n",
+ "Test Accuracy of deer: 67% (670/1000)\n",
+ "Test Accuracy of dog: 73% (735/1000)\n",
+ "Test Accuracy of frog: 81% (815/1000)\n",
+ "Test Accuracy of horse: 76% (762/1000)\n",
+ "Test Accuracy of ship: 81% (812/1000)\n",
+ "Test Accuracy of truck: 77% (771/1000)\n",
"\n",
- "Test Accuracy (Overall): 70% (7065/10000)\n"
+ "Test Accuracy (Overall): 72% (7279/10000)\n"
]
}
- ],
- "source": [
- "model2.load_state_dict(torch.load(\"/content/save_data/model2_cifar.pt\"))\n",
- "\n",
- "# track test loss\n",
- "test_loss = 0.0\n",
- "class_correct = list(0.0 for i in range(10))\n",
- "class_total = list(0.0 for i in range(10))\n",
- "\n",
- "model2.eval()\n",
- "# iterate over test data\n",
- "for data, target in test_loader:\n",
- " # move tensors to GPU if CUDA is available\n",
- " if train_on_gpu:\n",
- " data, target = data.cuda(), target.cuda()\n",
- " # forward pass: compute predicted outputs by passing inputs to the model\n",
- " output = model2(data)\n",
- " # calculate the batch loss\n",
- " loss = criterion(output, target)\n",
- " # update test loss\n",
- " test_loss += loss.item() * data.size(0)\n",
- " # convert output probabilities to predicted class\n",
- " _, pred = torch.max(output, 1)\n",
- " # compare predictions to true label\n",
- " correct_tensor = pred.eq(target.data.view_as(pred))\n",
- " correct = (\n",
- " np.squeeze(correct_tensor.numpy())\n",
- " if not train_on_gpu\n",
- " else np.squeeze(correct_tensor.cpu().numpy())\n",
- " )\n",
- " # calculate test accuracy for each object class\n",
- " for i in range(batch_size):\n",
- " label = target.data[i]\n",
- " class_correct[label] += correct[i].item()\n",
- " class_total[label] += 1\n",
- "\n",
- "# average test loss\n",
- "test_loss = test_loss / len(test_loader)\n",
- "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
- "\n",
- "for i in range(10):\n",
- " if class_total[i] > 0:\n",
- " print(\n",
- " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " classes[i],\n",
- " 100 * class_correct[i] / class_total[i],\n",
- " np.sum(class_correct[i]),\n",
- " np.sum(class_total[i]),\n",
- " )\n",
- " )\n",
- " else:\n",
- " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
- "\n",
- "print(\n",
- " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
- " np.sum(class_correct),\n",
- " np.sum(class_total),\n",
- " )\n",
- ")"
]
},
{
@@ -1073,32 +960,32 @@
},
{
"cell_type": "code",
- "execution_count": 27,
+ "execution_count": 7,
"id": "ef623c26",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "ef623c26",
- "outputId": "b12fc3df-3534-4c11-bd78-01a56839bbb7"
+ "outputId": "cd098248-8e4c-4f82-f78e-354398131af4"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
- "model: fp32 \t Size (KB): 251.342\n"
+ "model: fp32 \t Size (KB): 45304.25\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
- "251342"
+ "45304250"
]
},
"metadata": {},
- "execution_count": 27
+ "execution_count": 7
}
],
"source": [
@@ -1141,10 +1028,10 @@
"base_uri": "https://localhost:8080/"
},
"id": "38_XJKcrX5Lq",
- "outputId": "0482c566-6374-46a1-ef5d-4988bd94002d"
+ "outputId": "5a025dbf-02be-457f-c07b-22616cd4dc0b"
},
"id": "38_XJKcrX5Lq",
- "execution_count": 28,
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
@@ -1162,7 +1049,7 @@
]
},
"metadata": {},
- "execution_count": 28
+ "execution_count": 21
}
]
},
@@ -1188,15 +1075,18 @@
},
{
"cell_type": "code",
- "execution_count": 29,
- "id": "ZXeLJC39QjOP",
+ "source": [
+ "quantization_accuracy(quantized_model, test_loader, batch_size = 20)"
+ ],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
- "id": "ZXeLJC39QjOP",
- "outputId": "8d7fdd63-6a5f-4289-931c-e3fa874288e2"
+ "id": "PbNQL__PJ70k",
+ "outputId": "68f7c7a2-3361-40b5-a948-2f16891c455c"
},
+ "id": "PbNQL__PJ70k",
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
@@ -1212,87 +1102,22 @@
" (fc3): DynamicQuantizedLinear(in_features=64, out_features=10, dtype=torch.qint8, qscheme=torch.per_tensor_affine)\n",
" (dropout): Dropout(p=0.5, inplace=False)\n",
")\n",
- "Test Loss: 17.187755\n",
+ "Test Loss: 16.297476\n",
"\n",
- "Test Accuracy of airplane: 76% (765/1000)\n",
- "Test Accuracy of automobile: 77% (775/1000)\n",
- "Test Accuracy of bird: 58% (585/1000)\n",
- "Test Accuracy of cat: 57% (572/1000)\n",
- "Test Accuracy of deer: 65% (653/1000)\n",
- "Test Accuracy of dog: 52% (527/1000)\n",
- "Test Accuracy of frog: 78% (785/1000)\n",
- "Test Accuracy of horse: 78% (782/1000)\n",
- "Test Accuracy of ship: 82% (829/1000)\n",
- "Test Accuracy of truck: 79% (790/1000)\n",
+ "Test Accuracy of airplane: 78% (787/1000)\n",
+ "Test Accuracy of automobile: 86% (868/1000)\n",
+ "Test Accuracy of bird: 57% (577/1000)\n",
+ "Test Accuracy of cat: 48% (482/1000)\n",
+ "Test Accuracy of deer: 67% (670/1000)\n",
+ "Test Accuracy of dog: 73% (735/1000)\n",
+ "Test Accuracy of frog: 81% (813/1000)\n",
+ "Test Accuracy of horse: 76% (764/1000)\n",
+ "Test Accuracy of ship: 81% (813/1000)\n",
+ "Test Accuracy of truck: 77% (775/1000)\n",
"\n",
- "Test Accuracy (Overall): 70% (7063/10000)\n"
+ "Test Accuracy (Overall): 72% (7284/10000)\n"
]
}
- ],
- "source": [
- "#try with CPU dynamic quantization --> need to convert GPU to CPU device\n",
- "\n",
- "# track test loss\n",
- "test_loss = 0.0\n",
- "class_correct = list(0.0 for i in range(10))\n",
- "class_total = list(0.0 for i in range(10))\n",
- "\n",
- "quantized_model.eval()\n",
- "print(quantized_model)\n",
- "\n",
- "# iterate over test data\n",
- "for data, target in test_loader:\n",
- " # move tensors to GPU if CUDA is available\n",
- " data, target = data.cpu(), target.cpu()\n",
- "\n",
- " #print(data.device, target.device, next(quantized_model.parameters()).device)\n",
- " # forward pass: compute predicted outputs by passing inputs to the model\n",
- " with torch.no_grad() :\n",
- " output = quantized_model(data)\n",
- " # calculate the batch loss\n",
- " loss = criterion(output, target)\n",
- " # update test loss\n",
- " test_loss += loss.item() * data.size(0)\n",
- " # convert output probabilities to predicted class\n",
- " _, pred = torch.max(output, 1)\n",
- " # compare predictions to true label\n",
- " correct_tensor = pred.eq(target.data.view_as(pred))\n",
- " correct = (\n",
- " np.squeeze(correct_tensor.cpu().numpy())\n",
- "\n",
- " )\n",
- " # calculate test accuracy for each object class\n",
- " for i in range(batch_size):\n",
- " label = target.data[i]\n",
- " class_correct[label] += correct[i].item()\n",
- " class_total[label] += 1\n",
- "\n",
- "# average test loss\n",
- "test_loss = test_loss / len(test_loader)\n",
- "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
- "\n",
- "for i in range(10):\n",
- " if class_total[i] > 0:\n",
- " print(\n",
- " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " classes[i],\n",
- " 100 * class_correct[i] / class_total[i],\n",
- " np.sum(class_correct[i]),\n",
- " np.sum(class_total[i]),\n",
- " )\n",
- " )\n",
- " else:\n",
- " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
- "\n",
- "print(\n",
- " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
- " np.sum(class_correct),\n",
- " np.sum(class_total),\n",
- " )\n",
- ")"
]
},
{
@@ -1372,7 +1197,7 @@
"outputId": "75edab52-1360-46f4-eb1f-f50d4e928044"
},
"id": "ylrc7tZaKWd4",
- "execution_count": 30,
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
@@ -1558,7 +1383,7 @@
"outputId": "1cf5a022-aeff-485b-b7c9-47cf52879034"
},
"id": "uG_1vRQJRvkd",
- "execution_count": 31,
+ "execution_count": null,
"outputs": [
{
"output_type": "stream",
@@ -1639,7 +1464,7 @@
"outputId": "3bc958ec-20c5-462b-ed56-284a5f6b1ce0"
},
"id": "tYqC1SIM7ap1",
- "execution_count": 32,
+ "execution_count": null,
"outputs": [
{
"output_type": "display_data",
@@ -1731,113 +1556,28 @@
{
"cell_type": "code",
"source": [
- "#try with CPU dynamic quantization --> need to convert GPU to CPU device\n",
- "\n",
- "# track test loss\n",
- "test_loss = 0.0\n",
- "class_correct = list(0.0 for i in range(10))\n",
- "class_total = list(0.0 for i in range(10))\n",
- "\n",
- "aware_quantized_model.to(\"cpu\")\n",
- "aware_quantized_model.eval()\n",
- "print(aware_quantized_model)\n",
- "\n",
- "# iterate over test data\n",
- "for data, target in test_loader:\n",
- " # move tensors to CPU\n",
- " data, target = data.cpu(), target.cpu()\n",
- " # forward pass: compute predicted outputs by passing inputs to the model\n",
- " with torch.no_grad() :\n",
- " output = aware_quantized_model(data)\n",
- " # calculate the batch loss\n",
- " loss = criterion(output, target)\n",
- " # update test loss\n",
- " test_loss += loss.item() * data.size(0)\n",
- " # convert output probabilities to predicted class\n",
- " _, pred = torch.max(output, 1)\n",
- " # compare predictions to true label\n",
- " correct_tensor = pred.eq(target.data.view_as(pred))\n",
- " correct = (\n",
- " np.squeeze(correct_tensor.cpu().numpy())\n",
- "\n",
- " )\n",
- " # calculate test accuracy for each object class\n",
- " for i in range(batch_size):\n",
- " label = target.data[i]\n",
- " class_correct[label] += correct[i].item()\n",
- " class_total[label] += 1\n",
- "\n",
- "# average test loss\n",
- "test_loss = test_loss / len(test_loader)\n",
- "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
- "\n",
- "for i in range(10):\n",
- " if class_total[i] > 0:\n",
- " print(\n",
- " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " classes[i],\n",
- " 100 * class_correct[i] / class_total[i],\n",
- " np.sum(class_correct[i]),\n",
- " np.sum(class_total[i]),\n",
- " )\n",
- " )\n",
- " else:\n",
- " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
- "\n",
- "print(\n",
- " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
- " % (\n",
- " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
- " np.sum(class_correct),\n",
- " np.sum(class_total),\n",
- " )\n",
- ")"
+ "quantization_accuracy(aware_quantized_model, test_loader, batch_size = 20)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
- "height": 1000
+ "height": 139
},
- "id": "aYDwEmQzUPL5",
- "outputId": "7b477a0b-9d86-4e0d-dd12-7b92938c10ff"
+ "id": "Evnxp9N5Kdwg",
+ "outputId": "f5a7189d-cfe3-4520-cab4-e5d1efa80dd0"
},
- "id": "aYDwEmQzUPL5",
+ "id": "Evnxp9N5Kdwg",
"execution_count": null,
"outputs": [
- {
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "QatNet(\n",
- " (quant): Quantize(scale=tensor([1.]), zero_point=tensor([0]), dtype=torch.quint8)\n",
- " (conv1): QuantizedConv2d(3, 16, kernel_size=(3, 3), stride=(1, 1), scale=0.0678861141204834, zero_point=62, padding=(1, 1))\n",
- " (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
- " (conv2): QuantizedConv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), scale=0.0762503445148468, zero_point=47, padding=(1, 1))\n",
- " (conv3): QuantizedConv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), scale=0.10137047618627548, zero_point=69, padding=(1, 1))\n",
- " (fc1): QuantizedLinear(in_features=1024, out_features=512, scale=0.06663341820240021, zero_point=56, qscheme=torch.per_channel_affine)\n",
- " (fc2): QuantizedLinear(in_features=512, out_features=64, scale=0.05890081077814102, zero_point=51, qscheme=torch.per_channel_affine)\n",
- " (fc3): QuantizedLinear(in_features=64, out_features=10, scale=0.06607885658740997, zero_point=76, qscheme=torch.per_channel_affine)\n",
- " (dropout): QuantizedDropout(p=0.5, inplace=False)\n",
- ")\n"
- ]
- },
{
"output_type": "error",
- "ename": "NotImplementedError",
- "evalue": "Could not run 'quantized::conv2d.new' with arguments from the 'CPU' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'quantized::conv2d.new' is only available for these backends: [Meta, QuantizedCPU, QuantizedCUDA, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, AutogradMeta, Tracer, AutocastCPU, AutocastXPU, AutocastMPS, AutocastCUDA, FuncTorchBatched, BatchedNestedTensor, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PreDispatch, PythonDispatcher].\n\nMeta: registered at ../aten/src/ATen/core/MetaFallbackKernel.cpp:23 [backend fallback]\nQuantizedCPU: registered at ../aten/src/ATen/native/quantized/cpu/qconv.cpp:1972 [kernel]\nQuantizedCUDA: registered at ../aten/src/ATen/native/quantized/cudnn/Conv.cpp:391 [kernel]\nBackendSelect: fallthrough registered at ../aten/src/ATen/core/BackendSelectFallbackKernel.cpp:3 [backend fallback]\nPython: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:153 [backend fallback]\nFuncTorchDynamicLayerBackMode: registered at ../aten/src/ATen/functorch/DynamicLayer.cpp:497 [backend fallback]\nFunctionalize: registered at ../aten/src/ATen/FunctionalizeFallbackKernel.cpp:349 [backend fallback]\nNamed: registered at ../aten/src/ATen/core/NamedRegistrations.cpp:7 [backend fallback]\nConjugate: registered at ../aten/src/ATen/ConjugateFallback.cpp:17 [backend fallback]\nNegative: registered at ../aten/src/ATen/native/NegateFallback.cpp:18 [backend fallback]\nZeroTensor: registered at ../aten/src/ATen/ZeroTensorFallback.cpp:86 [backend fallback]\nADInplaceOrView: fallthrough registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:96 [backend fallback]\nAutogradOther: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:63 [backend fallback]\nAutogradCPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:67 [backend fallback]\nAutogradCUDA: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:75 [backend fallback]\nAutogradXLA: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:79 [backend fallback]\nAutogradMPS: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:87 [backend fallback]\nAutogradXPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:71 [backend fallback]\nAutogradHPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:100 [backend fallback]\nAutogradLazy: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:83 [backend fallback]\nAutogradMeta: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:91 [backend fallback]\nTracer: registered at ../torch/csrc/autograd/TraceTypeManual.cpp:294 [backend fallback]\nAutocastCPU: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:321 [backend fallback]\nAutocastXPU: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:463 [backend fallback]\nAutocastMPS: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:209 [backend fallback]\nAutocastCUDA: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:165 [backend fallback]\nFuncTorchBatched: registered at ../aten/src/ATen/functorch/LegacyBatchingRegistrations.cpp:731 [backend fallback]\nBatchedNestedTensor: registered at ../aten/src/ATen/functorch/LegacyBatchingRegistrations.cpp:758 [backend fallback]\nFuncTorchVmapMode: fallthrough registered at ../aten/src/ATen/functorch/VmapModeRegistrations.cpp:27 [backend fallback]\nBatched: registered at ../aten/src/ATen/LegacyBatchingRegistrations.cpp:1075 [backend fallback]\nVmapMode: fallthrough registered at ../aten/src/ATen/VmapModeRegistrations.cpp:33 [backend fallback]\nFuncTorchGradWrapper: registered at ../aten/src/ATen/functorch/TensorWrapper.cpp:207 [backend fallback]\nPythonTLSSnapshot: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:161 [backend fallback]\nFuncTorchDynamicLayerFrontMode: registered at ../aten/src/ATen/functorch/DynamicLayer.cpp:493 [backend fallback]\nPreDispatch: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:165 [backend fallback]\nPythonDispatcher: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:157 [backend fallback]\n",
+ "ename": "NameError",
+ "evalue": "name 'aware_quantized_model' is not defined",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
- "\u001b[0;31mNotImplementedError\u001b[0m Traceback (most recent call last)",
- "\u001b[0;32m<ipython-input-39-edca34e920f8>\u001b[0m in \u001b[0;36m<cell line: 13>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0;31m# forward pass: compute predicted outputs by passing inputs to the model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 17\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mno_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 18\u001b[0;31m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0maware_quantized_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 19\u001b[0m \u001b[0;31m# calculate the batch loss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 20\u001b[0m \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcriterion\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_wrapped_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1734\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_compiled_call_impl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# type: ignore[misc]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1735\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1736\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call_impl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1737\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1738\u001b[0m \u001b[0;31m# torchrec tests the code consistency with the following code\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1745\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1746\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1747\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1748\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1749\u001b[0m \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m<ipython-input-31-736a5f5ba3fe>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m 22\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 23\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 24\u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpool\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 25\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpool\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 26\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpool\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv3\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_wrapped_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1734\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_compiled_call_impl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# type: ignore[misc]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1735\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1736\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call_impl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1737\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1738\u001b[0m \u001b[0;31m# torchrec tests the code consistency with the following code\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1745\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_pre_hooks\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_global_backward_hooks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1746\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1747\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1748\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1749\u001b[0m \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/ao/nn/quantized/modules/conv.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m 593\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_reversed_padding_repeated_twice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpadding_mode\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 594\u001b[0m )\n\u001b[0;32m--> 595\u001b[0;31m return ops.quantized.conv2d(\n\u001b[0m\u001b[1;32m 596\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_packed_params\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscale\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_point\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 597\u001b[0m )\n",
- "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/torch/_ops.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1114\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_has_torchbind_op_overload\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0m_must_dispatch_in_python\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1115\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0m_call_overload_packet_from_python\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1116\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_op\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwargs\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1117\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1118\u001b[0m \u001b[0;31m# TODO: use this to make a __dir__\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;31mNotImplementedError\u001b[0m: Could not run 'quantized::conv2d.new' with arguments from the 'CPU' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'quantized::conv2d.new' is only available for these backends: [Meta, QuantizedCPU, QuantizedCUDA, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, AutogradMeta, Tracer, AutocastCPU, AutocastXPU, AutocastMPS, AutocastCUDA, FuncTorchBatched, BatchedNestedTensor, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PreDispatch, PythonDispatcher].\n\nMeta: registered at ../aten/src/ATen/core/MetaFallbackKernel.cpp:23 [backend fallback]\nQuantizedCPU: registered at ../aten/src/ATen/native/quantized/cpu/qconv.cpp:1972 [kernel]\nQuantizedCUDA: registered at ../aten/src/ATen/native/quantized/cudnn/Conv.cpp:391 [kernel]\nBackendSelect: fallthrough registered at ../aten/src/ATen/core/BackendSelectFallbackKernel.cpp:3 [backend fallback]\nPython: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:153 [backend fallback]\nFuncTorchDynamicLayerBackMode: registered at ../aten/src/ATen/functorch/DynamicLayer.cpp:497 [backend fallback]\nFunctionalize: registered at ../aten/src/ATen/FunctionalizeFallbackKernel.cpp:349 [backend fallback]\nNamed: registered at ../aten/src/ATen/core/NamedRegistrations.cpp:7 [backend fallback]\nConjugate: registered at ../aten/src/ATen/ConjugateFallback.cpp:17 [backend fallback]\nNegative: registered at ../aten/src/ATen/native/NegateFallback.cpp:18 [backend fallback]\nZeroTensor: registered at ../aten/src/ATen/ZeroTensorFallback.cpp:86 [backend fallback]\nADInplaceOrView: fallthrough registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:96 [backend fallback]\nAutogradOther: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:63 [backend fallback]\nAutogradCPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:67 [backend fallback]\nAutogradCUDA: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:75 [backend fallback]\nAutogradXLA: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:79 [backend fallback]\nAutogradMPS: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:87 [backend fallback]\nAutogradXPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:71 [backend fallback]\nAutogradHPU: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:100 [backend fallback]\nAutogradLazy: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:83 [backend fallback]\nAutogradMeta: registered at ../aten/src/ATen/core/VariableFallbackKernel.cpp:91 [backend fallback]\nTracer: registered at ../torch/csrc/autograd/TraceTypeManual.cpp:294 [backend fallback]\nAutocastCPU: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:321 [backend fallback]\nAutocastXPU: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:463 [backend fallback]\nAutocastMPS: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:209 [backend fallback]\nAutocastCUDA: fallthrough registered at ../aten/src/ATen/autocast_mode.cpp:165 [backend fallback]\nFuncTorchBatched: registered at ../aten/src/ATen/functorch/LegacyBatchingRegistrations.cpp:731 [backend fallback]\nBatchedNestedTensor: registered at ../aten/src/ATen/functorch/LegacyBatchingRegistrations.cpp:758 [backend fallback]\nFuncTorchVmapMode: fallthrough registered at ../aten/src/ATen/functorch/VmapModeRegistrations.cpp:27 [backend fallback]\nBatched: registered at ../aten/src/ATen/LegacyBatchingRegistrations.cpp:1075 [backend fallback]\nVmapMode: fallthrough registered at ../aten/src/ATen/VmapModeRegistrations.cpp:33 [backend fallback]\nFuncTorchGradWrapper: registered at ../aten/src/ATen/functorch/TensorWrapper.cpp:207 [backend fallback]\nPythonTLSSnapshot: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:161 [backend fallback]\nFuncTorchDynamicLayerFrontMode: registered at ../aten/src/ATen/functorch/DynamicLayer.cpp:493 [backend fallback]\nPreDispatch: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:165 [backend fallback]\nPythonDispatcher: registered at ../aten/src/ATen/core/PythonFallbackKernel.cpp:157 [backend fallback]\n"
+ "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
+ "\u001b[0;32m<ipython-input-24-f13bdc5ef89b>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mquantization_accuracy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maware_quantized_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m20\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+ "\u001b[0;31mNameError\u001b[0m: name 'aware_quantized_model' is not defined"
]
}
]
@@ -2285,29 +2025,24 @@
],
"metadata": {
"colab": {
- "base_uri": "https://localhost:8080/"
+ "base_uri": "https://localhost:8080/",
+ "height": 207
},
"id": "Y5hc5nkONkYP",
- "outputId": "02dcd741-8dda-46e6-f137-74c3413cfe35"
+ "outputId": "f9e665ab-70b0-4b13-b632-0eb0394e1304"
},
"id": "Y5hc5nkONkYP",
"execution_count": null,
"outputs": [
{
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "Predicted class with MobileNet50 is: Golden Retriever\n"
- ]
- },
- {
- "output_type": "stream",
- "name": "stderr",
- "text": [
- "/usr/local/lib/python3.10/dist-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",
- "/usr/local/lib/python3.10/dist-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=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.\n",
- " warnings.warn(msg)\n"
+ "output_type": "error",
+ "ename": "FileNotFoundError",
+ "evalue": "[Errno 2] No such file or directory: 'imagenet-simple-labels.json'",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mFileNotFoundError\u001b[0m Traceback (most recent call last)",
+ "\u001b[0;32m<ipython-input-25-b14d37f3485c>\u001b[0m in \u001b[0;36m<cell line: 14>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 12\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[0;31m# Prepare the labels\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 14\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"imagenet-simple-labels.json\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 15\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'imagenet-simple-labels.json'"
]
}
]
@@ -2409,7 +2144,7 @@
"colab": {
"base_uri": "https://localhost:8080/"
},
- "outputId": "5c990e86-d34a-45c2-985c-ddbf4c5158ff"
+ "outputId": "bebc8c94-5d03-40e5-e8f1-006d7661ff5b"
},
"outputs": [
{
@@ -2427,7 +2162,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 45,
"id": "be2d31f5",
"metadata": {
"id": "be2d31f5",
@@ -2435,7 +2170,7 @@
"base_uri": "https://localhost:8080/",
"height": 207
},
- "outputId": "8418f722-3bd4-4f85-a08c-3e71d1d16402"
+ "outputId": "c3754549-20a5-4a71-b2a2-82a794075d6a"
},
"outputs": [
{
@@ -2444,7 +2179,7 @@
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
],
- "image/png": "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\n"
+ "image/png": "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\n"
},
"metadata": {}
}
@@ -2539,103 +2274,12 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 46,
"id": "572d824c",
"metadata": {
- "id": "572d824c",
- "colab": {
- "base_uri": "https://localhost:8080/"
- },
- "outputId": "7ab4cd64-f395-4f25-f574-eef36b1d68ce"
+ "id": "572d824c"
},
- "outputs": [
- {
- "output_type": "stream",
- "name": "stderr",
- "text": [
- "/usr/local/lib/python3.10/dist-packages/torch/utils/data/dataloader.py:617: UserWarning: This DataLoader will create 4 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.\n",
- " warnings.warn(\n",
- "/usr/local/lib/python3.10/dist-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",
- "/usr/local/lib/python3.10/dist-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",
- "Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n",
- "100%|██████████| 44.7M/44.7M [00:00<00:00, 220MB/s]\n"
- ]
- },
- {
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "Epoch 1/10\n",
- "----------\n"
- ]
- },
- {
- "output_type": "stream",
- "name": "stderr",
- "text": [
- "/usr/local/lib/python3.10/dist-packages/torch/optim/lr_scheduler.py:224: 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(\n"
- ]
- },
- {
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "train Loss: 0.7472 Acc: 0.5738\n",
- "val Loss: 0.3010 Acc: 0.8562\n",
- "\n",
- "Epoch 2/10\n",
- "----------\n",
- "train Loss: 0.4474 Acc: 0.7910\n",
- "val Loss: 0.1926 Acc: 0.9542\n",
- "\n",
- "Epoch 3/10\n",
- "----------\n",
- "train Loss: 0.4452 Acc: 0.8033\n",
- "val Loss: 0.2513 Acc: 0.9020\n",
- "\n",
- "Epoch 4/10\n",
- "----------\n",
- "train Loss: 0.3650 Acc: 0.8279\n",
- "val Loss: 0.1967 Acc: 0.9542\n",
- "\n",
- "Epoch 5/10\n",
- "----------\n",
- "train Loss: 0.3299 Acc: 0.8607\n",
- "val Loss: 0.1995 Acc: 0.9412\n",
- "\n",
- "Epoch 6/10\n",
- "----------\n",
- "train Loss: 0.4849 Acc: 0.7910\n",
- "val Loss: 0.2128 Acc: 0.9346\n",
- "\n",
- "Epoch 7/10\n",
- "----------\n",
- "train Loss: 0.3027 Acc: 0.8689\n",
- "val Loss: 0.2118 Acc: 0.9281\n",
- "\n",
- "Epoch 8/10\n",
- "----------\n",
- "train Loss: 0.3632 Acc: 0.8361\n",
- "val Loss: 0.2298 Acc: 0.9281\n",
- "\n",
- "Epoch 9/10\n",
- "----------\n",
- "train Loss: 0.4297 Acc: 0.7992\n",
- "val Loss: 0.2540 Acc: 0.9150\n",
- "\n",
- "Epoch 10/10\n",
- "----------\n",
- "train Loss: 0.3748 Acc: 0.8607\n",
- "val Loss: 0.2135 Acc: 0.9346\n",
- "\n",
- "Training complete in 2m 14s\n",
- "Best val Acc: 0.954248\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"import copy\n",
"import os\n",
@@ -2689,36 +2333,16 @@
"}\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",
+ "classes = [\n",
+ " \"ants\",\n",
+ " \"bees\"\n",
+ "]\n",
+ "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
"def train_model(model, criterion, optimizer, scheduler, num_epochs=25):\n",
" since = time.time()\n",
"\n",
@@ -2792,9 +2416,18 @@
"\n",
" # Load best model weights\n",
" model.load_state_dict(best_model_wts)\n",
- " return model, epoch_time\n",
- "\n",
- "\n",
+ " return model, epoch_time"
+ ],
+ "metadata": {
+ "id": "n9txVkiGI6MH"
+ },
+ "id": "n9txVkiGI6MH",
+ "execution_count": 47,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
"# Download a pre-trained ResNet18 model and freeze its weights\n",
"model = torchvision.models.resnet18(pretrained=True)\n",
"for param in model.parameters():\n",
@@ -2814,7 +2447,100 @@
"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"
+ ")"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "CxQwLzk2I_4G",
+ "outputId": "b5eb3398-24e2-4e0d-b53b-afb87bc39452"
+ },
+ "id": "CxQwLzk2I_4G",
+ "execution_count": 48,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "/usr/local/lib/python3.10/dist-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",
+ "/usr/local/lib/python3.10/dist-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"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Epoch 1/10\n",
+ "----------\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "/usr/local/lib/python3.10/dist-packages/torch/optim/lr_scheduler.py:224: 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(\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "train Loss: 0.6502 Acc: 0.6680\n",
+ "val Loss: 0.2079 Acc: 0.9216\n",
+ "\n",
+ "Epoch 2/10\n",
+ "----------\n",
+ "train Loss: 0.3901 Acc: 0.8115\n",
+ "val Loss: 0.1938 Acc: 0.9216\n",
+ "\n",
+ "Epoch 3/10\n",
+ "----------\n",
+ "train Loss: 0.4604 Acc: 0.7992\n",
+ "val Loss: 0.2011 Acc: 0.9477\n",
+ "\n",
+ "Epoch 4/10\n",
+ "----------\n",
+ "train Loss: 0.4188 Acc: 0.7992\n",
+ "val Loss: 0.2322 Acc: 0.9085\n",
+ "\n",
+ "Epoch 5/10\n",
+ "----------\n",
+ "train Loss: 0.6148 Acc: 0.7582\n",
+ "val Loss: 0.2142 Acc: 0.9085\n",
+ "\n",
+ "Epoch 6/10\n",
+ "----------\n",
+ "train Loss: 0.5279 Acc: 0.7787\n",
+ "val Loss: 0.1803 Acc: 0.9281\n",
+ "\n",
+ "Epoch 7/10\n",
+ "----------\n",
+ "train Loss: 0.3512 Acc: 0.8566\n",
+ "val Loss: 0.1602 Acc: 0.9542\n",
+ "\n",
+ "Epoch 8/10\n",
+ "----------\n",
+ "train Loss: 0.3805 Acc: 0.8402\n",
+ "val Loss: 0.1627 Acc: 0.9542\n",
+ "\n",
+ "Epoch 9/10\n",
+ "----------\n",
+ "train Loss: 0.4193 Acc: 0.8033\n",
+ "val Loss: 0.1577 Acc: 0.9542\n",
+ "\n",
+ "Epoch 10/10\n",
+ "----------\n",
+ "train Loss: 0.3446 Acc: 0.8525\n",
+ "val Loss: 0.1639 Acc: 0.9542\n",
+ "\n",
+ "Training complete in 7m 25s\n",
+ "Best val Acc: 0.954248\n"
+ ]
+ }
]
},
{
@@ -2828,17 +2554,13 @@
"Study the code and the results obtained.\n",
"\n",
"Modify the code and add an \"eval_model\" function to allow\n",
- "the evaluation of the model on a test set (different from the learning and validation sets used during the learning phase). Study the results obtained.\n",
- "\n",
- "Now modify the code to replace the current classification layer with a set of two layers using a \"relu\" activation function for the middle layer, and the \"dropout\" mechanism for both layers. Renew the experiments and study the results obtained.\n",
- "\n",
- "Apply ther quantization (post and quantization aware) and evaluate impact on model size and accuracy."
+ "the evaluation of the model on a test set (different from the learning and validation sets used during the learning phase). Study the results obtained."
]
},
{
"cell_type": "markdown",
"source": [
- "We first need to load a new dataset and split it into a train test and validation set. We dataset found on Kaggle containing a train folder and a validation folder.\n",
+ "We first need to load a new dataset for testing, that we found on Kaggle.\n",
"\n"
],
"metadata": {
@@ -2896,53 +2618,40 @@
"device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
],
"metadata": {
- "colab": {
- "base_uri": "https://localhost:8080/"
- },
- "id": "tA0TM1zno5cY",
- "outputId": "320fe225-269e-4d3f-ff11-01d531cca3ad"
+ "id": "tA0TM1zno5cY"
},
"id": "tA0TM1zno5cY",
- "execution_count": 11,
- "outputs": [
- {
- "output_type": "stream",
- "name": "stderr",
- "text": [
- "/usr/local/lib/python3.10/dist-packages/torch/utils/data/dataloader.py:617: UserWarning: This DataLoader will create 4 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.\n",
- " warnings.warn(\n"
- ]
- }
- ]
+ "execution_count": 49,
+ "outputs": []
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 50,
"id": "PnI7tSTqXRZ8",
"metadata": {
"id": "PnI7tSTqXRZ8",
"colab": {
"base_uri": "https://localhost:8080/"
},
- "outputId": "bd43177b-dd27-4982-9bc3-4e0555093d99"
+ "outputId": "b47a46e8-f261-4955-dc16-c43f13aae6fb"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
- "Validation Loss: 0.1706 Accuracy: 0.9303\n"
+ "Validation Loss: 0.1131 Accuracy: 0.9508\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
- "(0.17056238605472884, tensor(0.9303, device='cuda:0', dtype=torch.float64))"
+ "(0.11311063478074845, tensor(0.9508, dtype=torch.float64))"
]
},
"metadata": {},
- "execution_count": 13
+ "execution_count": 50
}
],
"source": [
@@ -2975,6 +2684,584 @@
"eval_model(model, dataloaders, criterion)"
]
},
+ {
+ "cell_type": "markdown",
+ "source": [
+ "Now modify the code to replace the current classification layer with a set of two layers using a \"relu\" activation function for the middle layer, and the \"dropout\" mechanism for both layers. Renew the experiments and study the results obtained."
+ ],
+ "metadata": {
+ "id": "5XTT9N6GiVps"
+ },
+ "id": "5XTT9N6GiVps"
+ },
+ {
+ "cell_type": "code",
+ "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 = \"/content/drive/MyDrive/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",
+ "# 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",
+ "\n",
+ "# Replace the final fully connected layer with two layers + ReLU and Dropout\n",
+ "num_ftrs = model.fc.in_features\n",
+ "model.fc = nn.Sequential(\n",
+ " nn.Linear(num_ftrs, 256), # First hidden layer\n",
+ " nn.ReLU(), # Activation function\n",
+ " nn.Dropout(0.5), # Dropout for regularization\n",
+ " nn.Linear(256, 10), # Final output layer\n",
+ " nn.Dropout(0.5), # Additional Dropout\n",
+ ")\n",
+ "\n",
+ "# Send the model to the GPU\n",
+ "model = model.to(device)\n",
+ "\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_relu_resnet, epoch_time = train_model(\n",
+ " model, criterion, optimizer_conv, exp_lr_scheduler, num_epochs=10\n",
+ ")"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "lDbUv0iHhRno",
+ "outputId": "7ee2f465-245a-4cba-9c48-03f67c0e2b8b"
+ },
+ "id": "lDbUv0iHhRno",
+ "execution_count": 18,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Epoch 1/10\n",
+ "----------\n",
+ "train Loss: 1.7383 Acc: 0.4713\n",
+ "val Loss: 0.9543 Acc: 0.8824\n",
+ "\n",
+ "Epoch 2/10\n",
+ "----------\n",
+ "train Loss: 1.4051 Acc: 0.5410\n",
+ "val Loss: 0.5446 Acc: 0.9020\n",
+ "\n",
+ "Epoch 3/10\n",
+ "----------\n",
+ "train Loss: 1.3050 Acc: 0.6352\n",
+ "val Loss: 0.4305 Acc: 0.9346\n",
+ "\n",
+ "Epoch 4/10\n",
+ "----------\n",
+ "train Loss: 1.2025 Acc: 0.6475\n",
+ "val Loss: 0.4116 Acc: 0.8889\n",
+ "\n",
+ "Epoch 5/10\n",
+ "----------\n",
+ "train Loss: 1.3266 Acc: 0.6270\n",
+ "val Loss: 0.2871 Acc: 0.9673\n",
+ "\n",
+ "Epoch 6/10\n",
+ "----------\n",
+ "train Loss: 1.3812 Acc: 0.6311\n",
+ "val Loss: 0.2952 Acc: 0.9346\n",
+ "\n",
+ "Epoch 7/10\n",
+ "----------\n",
+ "train Loss: 1.2949 Acc: 0.6516\n",
+ "val Loss: 0.2783 Acc: 0.9542\n",
+ "\n",
+ "Epoch 8/10\n",
+ "----------\n",
+ "train Loss: 1.2060 Acc: 0.6680\n",
+ "val Loss: 0.2904 Acc: 0.9542\n",
+ "\n",
+ "Epoch 9/10\n",
+ "----------\n",
+ "train Loss: 1.2004 Acc: 0.6721\n",
+ "val Loss: 0.3089 Acc: 0.9477\n",
+ "\n",
+ "Epoch 10/10\n",
+ "----------\n",
+ "train Loss: 1.3334 Acc: 0.5861\n",
+ "val Loss: 0.3138 Acc: 0.9412\n",
+ "\n",
+ "Training complete in 7m 18s\n",
+ "Best val Acc: 0.967320\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "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",
+ " \"ant_bees\": 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",
+ "}\n",
+ "\n",
+ "data_dir = \"/content/drive/MyDrive\"\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 [\"ant_bees\"]\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 [\"ant_bees\"]\n",
+ "}\n",
+ "test_loader = dataloaders['ant_bees']\n",
+ "dataset_sizes = {x: len(image_datasets[x]) for x in [\"ant_bees\"]}\n",
+ "class_names = image_datasets[\"ant_bees\"].classes\n",
+ "classes = [\n",
+ " \"ants\",\n",
+ " \"bees\"\n",
+ "]\n",
+ "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
+ ],
+ "metadata": {
+ "id": "PJER_Ta5BDFI"
+ },
+ "id": "PJER_Ta5BDFI",
+ "execution_count": 52,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "eval_model(model_relu_resnet, dataloaders, criterion)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "zscbvAR3Wdy-",
+ "outputId": "96c59f95-7547-483b-ed59-a19e579d3f91"
+ },
+ "id": "zscbvAR3Wdy-",
+ "execution_count": 53,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Validation Loss: 0.2994 Accuracy: 0.9344\n"
+ ]
+ },
+ {
+ "output_type": "execute_result",
+ "data": {
+ "text/plain": [
+ "(0.29941917992517597, tensor(0.9344, dtype=torch.float64))"
+ ]
+ },
+ "metadata": {},
+ "execution_count": 53
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "Apply the quantization (post and quantization aware) and evaluate impact on model size and accuracy."
+ ],
+ "metadata": {
+ "id": "Q9y1xoZ6l19m"
+ },
+ "id": "Q9y1xoZ6l19m"
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "model_relu_resnet.to(\"cpu\")\n",
+ "quantized_model = torch.quantization.quantize_dynamic(model_relu_resnet, dtype=torch.qint8)\n",
+ "print_size_of_model(model_relu_resnet, \"fp32\")\n",
+ "print_size_of_model(quantized_model, \"int8\")"
+ ],
+ "metadata": {
+ "id": "WUC_7Ocyl0H9",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "01a19cc4-8bb0-4bc1-f218-b6c2dc490e25"
+ },
+ "id": "WUC_7Ocyl0H9",
+ "execution_count": 22,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "model: fp32 \t Size (KB): 45312.442\n",
+ "model: int8 \t Size (KB): 44913.062\n"
+ ]
+ },
+ {
+ "output_type": "execute_result",
+ "data": {
+ "text/plain": [
+ "44913062"
+ ]
+ },
+ "metadata": {},
+ "execution_count": 22
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "eval_model(quantized_model, dataloaders, criterion)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "i02R-4KWf6Dg",
+ "outputId": "8a1fe8f3-4219-43ad-e093-25b0688d36d4"
+ },
+ "id": "i02R-4KWf6Dg",
+ "execution_count": 55,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Validation Loss: 0.3189 Accuracy: 0.9385\n"
+ ]
+ },
+ {
+ "output_type": "execute_result",
+ "data": {
+ "text/plain": [
+ "(0.31886099401067514, tensor(0.9385, dtype=torch.float64))"
+ ]
+ },
+ "metadata": {},
+ "execution_count": 55
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "#Compare dense model and quantized model acuracy for the two classes ants and bees\n",
+ "batch_size = 2\n",
+ "# track test loss\n",
+ "test_loss = 0.0\n",
+ "class_correct = [0.0, 0.0]\n",
+ "quantized_class_correct = [0, 0]\n",
+ "class_total = [0.0, 0.0]\n",
+ "\n",
+ "model.eval(); quantized_model.eval()\n",
+ "# iterate over test data\n",
+ "for data, target in test_loader:\n",
+ " # move tensors to GPU if CUDA is available\n",
+ " if train_on_gpu:\n",
+ " data, target = data.cuda(), target.cuda()\n",
+ " # forward pass: compute predicted outputs by passing inputs to the model\n",
+ " output = model(data)\n",
+ " quantized_output = quantized_model(data)\n",
+ "\n",
+ " # convert output probabilities to predicted class\n",
+ " _, pred = torch.max(output, 1)\n",
+ " _, quantized_pred = torch.max(quantized_output, 1)\n",
+ "\n",
+ " # compare predictions to true label\n",
+ " correct_tensor = pred.eq(target.data.view_as(pred))\n",
+ " quantized_correct_tensor = quantized_pred.eq(target.data.view_as(quantized_pred))\n",
+ "\n",
+ " correct = (\n",
+ " np.squeeze(correct_tensor.numpy())\n",
+ " if not train_on_gpu\n",
+ " else np.squeeze(correct_tensor.cpu().numpy())\n",
+ " )\n",
+ " quantized_correct = (\n",
+ " np.squeeze(quantized_correct_tensor.numpy())\n",
+ " if not train_on_gpu\n",
+ " else np.squeeze(quantized_correct_tensor.cpu().numpy())\n",
+ " )\n",
+ " # calculate test accuracy for each object class\n",
+ " for i in range(batch_size):\n",
+ " label = target.data[i]\n",
+ " class_correct[label] += correct[i].item()\n",
+ " quantized_class_correct[label] += quantized_correct[i].item()\n",
+ " class_total[label] += 1\n",
+ "\n",
+ "for i in range(2): #two classes : ants and bees\n",
+ " if class_total[i] > 0:\n",
+ " print(\n",
+ " \"Test Accuracy difference of %5s between model and its quantized version: %2d instance(s) (%.2f%%)\"\n",
+ " % (\n",
+ " classes[i],\n",
+ " class_correct[i] - quantized_class_correct[i],\n",
+ " 100 * (class_correct[i] - quantized_class_correct[i]) / class_correct[i],\n",
+ " )\n",
+ " )\n",
+ " else:\n",
+ " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
+ "\n",
+ "print(\n",
+ " \"\\nTest Accuracy (Overall) difference between model and its quantized version: %2d instance(s) (%.2f%%)\"\n",
+ " % (\n",
+ " np.sum(class_correct) - np.sum(quantized_class_correct),\n",
+ " 100.0 * (np.sum(class_correct) - np.sum(quantized_class_correct)) / np.sum(class_correct)\n",
+ " )\n",
+ " )"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "JlFmpPMKS8VW",
+ "outputId": "2fc5d0aa-db27-4d6f-a9f8-7f462aa045f4"
+ },
+ "id": "JlFmpPMKS8VW",
+ "execution_count": 54,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Test Accuracy difference of ants between model and its quantized version: 0 instance(s) (0.00%)\n",
+ "Test Accuracy difference of bees between model and its quantized version: 1 instance(s) (1.67%)\n",
+ "\n",
+ "Test Accuracy (Overall) difference between model and its quantized version: 1 instance(s) (0.84%)\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "def quantization_accuracy(quantized_model, test_loader, batch_size = 20) :\n",
+ " #try with CPU dynamic quantization --> need to convert GPU to CPU device\n",
+ "\n",
+ " # track test loss\n",
+ " test_loss = 0.0\n",
+ " class_correct = list(0.0 for i in range(10))\n",
+ " class_total = list(0.0 for i in range(10))\n",
+ "\n",
+ " quantized_model.eval()\n",
+ " print(quantized_model)\n",
+ "\n",
+ " # iterate over test data\n",
+ " for data, target in test_loader:\n",
+ " # move tensors to CPU\n",
+ " data, target = data.cpu(), target.cpu()\n",
+ "\n",
+ " #print(data.device, target.device, next(quantized_model.parameters()).device)\n",
+ " # forward pass: compute predicted outputs by passing inputs to the model\n",
+ " with torch.no_grad() :\n",
+ " output = quantized_model(data)\n",
+ " # calculate the batch loss\n",
+ " loss = criterion(output, target)\n",
+ " # update test loss\n",
+ " test_loss += loss.item() * data.size(0)\n",
+ " # convert output probabilities to predicted class\n",
+ " _, pred = torch.max(output, 1)\n",
+ " # compare predictions to true label\n",
+ " correct_tensor = pred.eq(target.data.view_as(pred))\n",
+ " correct = (\n",
+ " np.squeeze(correct_tensor.cpu().numpy())\n",
+ "\n",
+ " )\n",
+ " # calculate test accuracy for each object class\n",
+ " for i in range(batch_size):\n",
+ " label = target.data[i]\n",
+ " class_correct[label] += correct[i].item()\n",
+ " class_total[label] += 1\n",
+ "\n",
+ " # average test loss\n",
+ " test_loss = test_loss / len(test_loader)\n",
+ " print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
+ "\n",
+ " for i in range(10):\n",
+ " if class_total[i] > 0:\n",
+ " print(\n",
+ " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
+ " % (\n",
+ " classes[i],\n",
+ " 100 * class_correct[i] / class_total[i],\n",
+ " np.sum(class_correct[i]),\n",
+ " np.sum(class_total[i]),\n",
+ " )\n",
+ " )\n",
+ " else:\n",
+ " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
+ "\n",
+ " print(\n",
+ " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
+ " % (\n",
+ " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
+ " np.sum(class_correct),\n",
+ " np.sum(class_total),\n",
+ " )\n",
+ " )"
+ ],
+ "metadata": {
+ "id": "eFvkgzp8B6vV"
+ },
+ "id": "eFvkgzp8B6vV",
+ "execution_count": 12,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "def dense_model_accuracy(model, test_loader, batch_size = 20) :\n",
+ "\n",
+ " # track test loss\n",
+ " test_loss = 0.0\n",
+ " class_correct = list(0.0 for i in range(10))\n",
+ " class_total = list(0.0 for i in range(10))\n",
+ "\n",
+ " model.eval()\n",
+ " # iterate over test data\n",
+ " for data, target in test_loader:\n",
+ " # move tensors to GPU if CUDA is available\n",
+ " if train_on_gpu:\n",
+ " data, target = data.cuda(), target.cuda()\n",
+ " # forward pass: compute predicted outputs by passing inputs to the model\n",
+ " output = model(data)\n",
+ " # calculate the batch loss\n",
+ " loss = criterion(output, target)\n",
+ " # update test loss\n",
+ " test_loss += loss.item() * data.size(0)\n",
+ " # convert output probabilities to predicted class\n",
+ " _, pred = torch.max(output, 1)\n",
+ " # compare predictions to true label\n",
+ " correct_tensor = pred.eq(target.data.view_as(pred))\n",
+ " correct = (\n",
+ " np.squeeze(correct_tensor.numpy())\n",
+ " if not train_on_gpu\n",
+ " else np.squeeze(correct_tensor.cpu().numpy())\n",
+ " )\n",
+ " # calculate test accuracy for each object class\n",
+ " for i in range(batch_size):\n",
+ " label = target.data[i]\n",
+ " class_correct[label] += correct[i].item()\n",
+ " class_total[label] += 1\n",
+ "\n",
+ " # average test loss\n",
+ " test_loss = test_loss / len(test_loader)\n",
+ " print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
+ "\n",
+ " for i in range(10):\n",
+ " if class_total[i] > 0:\n",
+ " print(\n",
+ " \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
+ " % (\n",
+ " classes[i],\n",
+ " 100 * class_correct[i] / class_total[i],\n",
+ " np.sum(class_correct[i]),\n",
+ " np.sum(class_total[i]),\n",
+ " )\n",
+ " )\n",
+ " else:\n",
+ " print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
+ "\n",
+ " print(\n",
+ " \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
+ " % (\n",
+ " 100.0 * np.sum(class_correct) / np.sum(class_total),\n",
+ " np.sum(class_correct),\n",
+ " np.sum(class_total),\n",
+ " )\n",
+ " )"
+ ],
+ "metadata": {
+ "id": "ARf4ffFIGbHT"
+ },
+ "id": "ARf4ffFIGbHT",
+ "execution_count": null,
+ "outputs": []
+ },
{
"cell_type": "markdown",
"id": "04a263f0",
--
GitLab