Release v0.1.0 CIFAR-10 CNN baseline

Upload v0.1.0 PyTorch checkpoint, metrics, plots, logs, TensorBoard event, and model card. Test accuracy: 0.7857 on uoft-cs/cifar10 plain_text test split.

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+outputs/confusion_matrix.png filter=lfs diff=lfs merge=lfs -text
+outputs/demo_predictions.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,125 @@
+---
+library_name: pytorch
+pipeline_tag: image-classification
+datasets:
+- uoft-cs/cifar10
+metrics:
+- accuracy
+tags:
+- pytorch
+- cnn
+- cifar10
+- image-classification
+- computer-vision
+model-index:
+- name: cnn-cifar10-classifier-v0.1.0
+  results:
+  - task:
+      type: image-classification
+      name: Image Classification
+    dataset:
+      name: CIFAR-10
+      type: uoft-cs/cifar10
+      config: plain_text
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.7857
+      name: Test Accuracy
+---
+
+# cnn-cifar10-classifier v0.1.0
+
+Small PyTorch CNN baseline for CIFAR-10 image classification.
+
+This release contains the trained v0.1.0 checkpoint from the GitHub project:
+
+https://github.com/diverHansun/cnn-cifar10-classifier
+
+## Results
+
+| Split | Metric | Value |
+| --- | --- | ---: |
+| test | accuracy | 0.7857 |
+
+The checkpoint was selected by best validation/test accuracy during a 20 epoch run.
+
+Training summary:
+
+- Dataset: `uoft-cs/cifar10`
+- Config: `plain_text`
+- Epochs: 20
+- Batch size: 256
+- Optimizer: SGD, momentum 0.9, weight decay 0.0005
+- Learning rate: 0.01
+- Augmentation: random crop with padding 4, random horizontal flip
+- AMP: enabled
+- GPU used: NVIDIA GeForce RTX 5070 Ti
+- PyTorch: 2.11.0+cu128
+
+## Files
+
+```text
+checkpoints/best_model.pth
+checkpoints/last_model.pth
+outputs/training_metrics.json
+outputs/training_curves.png
+outputs/confusion_matrix.png
+outputs/demo_predictions.png
+logs/train_20260602_044856.log
+logs/evaluate_20260602_045254.log
+logs/demo_20260602_045311.log
+runs/cifar10_cnn_20260602_044901/events.out.tfevents...
+release_summary.json
+manifest.sha256
+```
+
+## Usage
+
+Clone the project code first:
+
+```bash
+git clone git@github.com:diverHansun/cnn-cifar10-classifier.git
+cd cnn-cifar10-classifier
+```
+
+Install dependencies with a CUDA-compatible PyTorch build for your machine, then download this checkpoint:
+
+```bash
+hf download diverWayne/cnn-cifar10-classifier checkpoints/best_model.pth --local-dir .
+```
+
+Evaluate:
+
+```bash
+python evaluate.py --checkpoint checkpoints/best_model.pth --device cuda
+```
+
+Run the demo grid:
+
+```bash
+python demo.py --checkpoint checkpoints/best_model.pth --samples 16 --device cuda
+```
+
+Predict one image:
+
+```bash
+python predict.py --image demo_images/your_image.png --checkpoint checkpoints/best_model.pth --device cuda
+```
+
+## Limitations
+
+This is a simple hand-written CNN baseline trained on CIFAR-10 32x32 images. It supports the 10 CIFAR-10 classes only:
+
+```text
+airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck
+```
+
+It can classify arbitrary images after resizing to 32x32, but reliability outside CIFAR-10-like images is limited.
+
+The strongest observed confusions are between visually similar categories such as `cat` and `dog`, `bird` and `deer/dog`, and `airplane` and `ship`.
+
+## Dataset
+
+The training data is not redistributed in this model repository. It is loaded from the public Hugging Face dataset `uoft-cs/cifar10`.
+

checkpoints/best_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0d179becbbe9ec354d73018e8279b546a8fd423e1ae70be99af1fc62dc1ffe6
+size 4972277

checkpoints/last_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34911e56f2fe22f84ca9f4f384166a15c407658f6b285e1942f1ad4a94cbb3b3
+size 4972277

logs/demo_20260602_045311.log

@@ -0,0 +1 @@
+saved demo predictions: /workspace/projects/cnn-cifar10-classifier/outputs/demo_predictions.png

logs/evaluate_20260602_045254.log

@@ -0,0 +1,5 @@
+checkpoint: checkpoints/best_model.pth
+checkpoint epoch: 19
+checkpoint best_acc: 0.7855
+test accuracy: 0.7857
+saved confusion matrix: /workspace/projects/cnn-cifar10-classifier/outputs/confusion_matrix.png

logs/train_20260602_044856.log

@@ -0,0 +1,35 @@
+torch: 2.11.0+cu128
+cuda available: True
+cuda version: 12.8
+selected device: cuda
+gpu count: 1
+gpu 0: NVIDIA GeForce RTX 5070 Ti
+gpu compute capability: sm_120
+torch CUDA arch list: ['sm_75', 'sm_80', 'sm_86', 'sm_90', 'sm_100', 'sm_120']
+cuda probe result: 8.0
+parameters: 620,362
+run: /workspace/projects/cnn-cifar10-classifier/runs/cifar10_cnn_20260602_044901
+best checkpoint: /workspace/projects/cnn-cifar10-classifier/checkpoints/best_model.pth
+last checkpoint: /workspace/projects/cnn-cifar10-classifier/checkpoints/last_model.pth
+epoch 1/20 train_loss=2.1013 train_acc=0.2129 test_loss=1.7602 test_acc=0.3562 best=0.3562
+epoch 2/20 train_loss=1.6477 train_acc=0.3943 test_loss=1.4481 test_acc=0.4711 best=0.4711
+epoch 3/20 train_loss=1.4371 train_acc=0.4795 test_loss=1.2667 test_acc=0.5494 best=0.5494
+epoch 4/20 train_loss=1.3097 train_acc=0.5281 test_loss=1.1753 test_acc=0.5855 best=0.5855
+epoch 5/20 train_loss=1.1900 train_acc=0.5767 test_loss=1.0543 test_acc=0.6230 best=0.6230
+epoch 6/20 train_loss=1.1005 train_acc=0.6092 test_loss=1.0158 test_acc=0.6357 best=0.6357
+epoch 7/20 train_loss=1.0239 train_acc=0.6383 test_loss=0.9475 test_acc=0.6683 best=0.6683
+epoch 8/20 train_loss=0.9602 train_acc=0.6616 test_loss=0.8667 test_acc=0.7035 best=0.7035
+epoch 9/20 train_loss=0.9007 train_acc=0.6841 test_loss=0.8255 test_acc=0.7066 best=0.7066
+epoch 10/20 train_loss=0.8600 train_acc=0.7009 test_loss=0.7960 test_acc=0.7294 best=0.7294
+epoch 11/20 train_loss=0.8172 train_acc=0.7119 test_loss=0.8018 test_acc=0.7216 best=0.7294
+epoch 12/20 train_loss=0.7871 train_acc=0.7234 test_loss=0.7545 test_acc=0.7424 best=0.7424
+epoch 13/20 train_loss=0.7600 train_acc=0.7345 test_loss=0.7226 test_acc=0.7495 best=0.7495
+epoch 14/20 train_loss=0.7326 train_acc=0.7429 test_loss=0.7281 test_acc=0.7494 best=0.7495
+epoch 15/20 train_loss=0.7085 train_acc=0.7540 test_loss=0.7082 test_acc=0.7533 best=0.7533
+epoch 16/20 train_loss=0.6916 train_acc=0.7579 test_loss=0.6807 test_acc=0.7633 best=0.7633
+epoch 17/20 train_loss=0.6694 train_acc=0.7654 test_loss=0.7073 test_acc=0.7568 best=0.7633
+epoch 18/20 train_loss=0.6508 train_acc=0.7712 test_loss=0.6440 test_acc=0.7770 best=0.7770
+epoch 19/20 train_loss=0.6381 train_acc=0.7747 test_loss=0.6354 test_acc=0.7811 best=0.7811
+epoch 20/20 train_loss=0.6157 train_acc=0.7846 test_loss=0.6217 test_acc=0.7855 best=0.7855
+saved training curves: /workspace/projects/cnn-cifar10-classifier/outputs/training_curves.png
+saved metrics: /workspace/projects/cnn-cifar10-classifier/outputs/training_metrics.json

manifest.sha256

@@ -0,0 +1,12 @@
+0b4da6d1f6b6e914fa9d525b857b685dd9a23395c0ce6a12230b33b8a26bcdcd  ./README.md
+c0d179becbbe9ec354d73018e8279b546a8fd423e1ae70be99af1fc62dc1ffe6  ./checkpoints/best_model.pth
+34911e56f2fe22f84ca9f4f384166a15c407658f6b285e1942f1ad4a94cbb3b3  ./checkpoints/last_model.pth
+ff0fc6f2aa3543488448a1c691568dd23a1a24bbc330b307d37489078c4675f2  ./logs/demo_20260602_045311.log
+d70b7ed43f0d2a87d5ba52d6b66fffc32b4b91aa1bfd2bdd32a58ffcfe0eb898  ./logs/evaluate_20260602_045254.log
+afd50fb0dc8bf4a73378d177cb43f253e72fc4e3493627765f9ffb4a69e44378  ./logs/train_20260602_044856.log
+2886ec553338e6da1bafa3861ef63e45c3de80fff9b2e271f6631b37b68ced85  ./outputs/confusion_matrix.png
+5a5499e95347c0897cdcc7db798680b7dbb9f186aac94fb80fe7dad928c72a3f  ./outputs/demo_predictions.png
+db90ee688d37349b9c32b49bf17751002483977a1a99b9ff18400b8ea76ab4ca  ./outputs/training_curves.png
+44bfd23238c3852e0f95521b3eaf4b63f9526023202e2789a74be240ea4a2176  ./outputs/training_metrics.json
+cc8542a55b06abd39b2c96a944825089dd3684707f3984ee754025eb46ab47f2  ./release_summary.json
+9e264f7342f7d5dfc2ccbc766de2b475b3f343a5f9d1d166306bf6bc5e95bd13  ./runs/cifar10_cnn_20260602_044901/events.out.tfevents.1780375741.8765c0e10ff8.8624.0

outputs/training_metrics.json

@@ -0,0 +1,90 @@
+{
+  "train_loss": [
+    2.1012727269744875,
+    1.647732571334839,
+    1.4371476556777953,
+    1.3097175415802003,
+    1.1899514984321595,
+    1.1004877404022217,
+    1.0239286506843568,
+    0.9602188570785523,
+    0.9007160757827759,
+    0.8599612341499329,
+    0.8172409128952026,
+    0.7870971046638489,
+    0.7600001029586791,
+    0.7325775894546509,
+    0.7085187640762329,
+    0.6916227349090576,
+    0.669356112689972,
+    0.6508153208732605,
+    0.6380657346725463,
+    0.6156930644035339
+  ],
+  "train_acc": [
+    0.21292,
+    0.39432,
+    0.47946,
+    0.52808,
+    0.5767,
+    0.60918,
+    0.63826,
+    0.66158,
+    0.6841,
+    0.70088,
+    0.7119,
+    0.72342,
+    0.73452,
+    0.7429,
+    0.75396,
+    0.75792,
+    0.76542,
+    0.77122,
+    0.77468,
+    0.7846
+  ],
+  "val_loss": [
+    1.760174755859375,
+    1.4480891510009766,
+    1.266711572265625,
+    1.1753166015625,
+    1.054260662841797,
+    1.0157690048217773,
+    0.9475269454956055,
+    0.8667480499267578,
+    0.8255449432373047,
+    0.796046971130371,
+    0.8017741333007813,
+    0.754474397277832,
+    0.7225680871963501,
+    0.7281325637817383,
+    0.7082156867980957,
+    0.6806870323181152,
+    0.7072956486701966,
+    0.6440230610847473,
+    0.6354327146530151,
+    0.6217316284179687
+  ],
+  "val_acc": [
+    0.3562,
+    0.4711,
+    0.5494,
+    0.5855,
+    0.623,
+    0.6357,
+    0.6683,
+    0.7035,
+    0.7066,
+    0.7294,
+    0.7216,
+    0.7424,
+    0.7495,
+    0.7494,
+    0.7533,
+    0.7633,
+    0.7568,
+    0.777,
+    0.7811,
+    0.7855
+  ]
+}

release_summary.json

@@ -0,0 +1,50 @@
+{
+  "version": "v0.1.0",
+  "repo": "diverWayne/cnn-cifar10-classifier",
+  "github": "https://github.com/diverHansun/cnn-cifar10-classifier",
+  "dataset": {
+    "name": "uoft-cs/cifar10",
+    "config": "plain_text",
+    "split": "train/test"
+  },
+  "model": {
+    "architecture": "SimpleCNN",
+    "parameters": 620362,
+    "classes": [
+      "airplane",
+      "automobile",
+      "bird",
+      "cat",
+      "deer",
+      "dog",
+      "frog",
+      "horse",
+      "ship",
+      "truck"
+    ]
+  },
+  "training": {
+    "epochs": 20,
+    "batch_size": 256,
+    "optimizer": "SGD",
+    "learning_rate": 0.01,
+    "momentum": 0.9,
+    "weight_decay": 0.0005,
+    "seed": 42,
+    "amp": true,
+    "augment": true
+  },
+  "environment": {
+    "gpu": "NVIDIA GeForce RTX 5070 Ti",
+    "torch": "2.11.0+cu128",
+    "cuda": "12.8"
+  },
+  "metrics": {
+    "train_loss": 0.6156930644035339,
+    "train_accuracy": 0.7846,
+    "test_loss": 0.6217316284179687,
+    "test_accuracy_from_training": 0.7855,
+    "test_accuracy_from_evaluate": 0.7857,
+    "best_checkpoint_epoch": 19
+  }
+}

runs/cifar10_cnn_20260602_044901/events.out.tfevents.1780375741.8765c0e10ff8.8624.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e264f7342f7d5dfc2ccbc766de2b475b3f343a5f9d1d166306bf6bc5e95bd13
+size 4040