Upload folder using huggingface_hub

.gitattributes

@@ -1,35 +1,27 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
 *.ftz filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
 *.joblib filter=lfs diff=lfs merge=lfs -text
 *.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text
 *.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 *.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text 
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

README.md

@@ -0,0 +1,72 @@
+# 🌈 Mô hình phân loại hình ảnh
+## 📝 Mô tả
+Mô hình này là phiên bản fine-tuned của google/vit-base-patch16-224-in21k trên bộ dữ liệu CIFAR-10 (60.000 ảnh màu 32×32, 10 lớp).
+Trên tập đánh giá, mô hình đạt:
+Loss: 0.2564
+Accuracy: 97.88
+
+## 📌 Ứng dụng
+Phân loại ảnh CIFAR-10 (10 lớp như: airplane, automobile, bird, cat, …)
+Dùng để transfer learning cho các tập dữ liệu nhỏ hơn
+Làm nền tảng cho các thử nghiệm về fine-tuning ViT trên dữ liệu màu sắc nhỏ
+
+## 📥 Chuẩn bị đầu vào
+Ảnh màu RGB
+Thư viện sẽ xử lý bằng ViTImageProcessor:
+Resize hình về 224×224
+Chuẩn hóa theo mean/std của ImageNet
+
+## 📤 Đầu ra
+Logits tensor kích thước [batch_size, 10]
+Dùng argmax(-1) để nhận nhãn dự đoán
+
+## 🛠 Cài đặt
+```bash
+pip install torch torchvision transformers
+```
+
+## 🧪 Ví dụ sử dụng
+```python
+import torch
+from transformers import ViTForImageClassification, ViTImageProcessor
+from PIL import Image
+
+model_name = "zhaospei/Model_16"
+
+processor = ViTImageProcessor.from_pretrained(model_name)
+model = ViTForImageClassification.from_pretrained(model_name)
+
+img = Image.open("path/to/cifar10_image.png").convert("RGB")
+inputs = processor(img, return_tensors="pt")
+
+with torch.no_grad():
+    logits = model(**inputs).logits
+
+pred = logits.argmax(-1).item()
+label = model.config.id2label[pred]
+print(f"Predicted label: {label} (Index: {pred})")
+```
+
+## ⚙️ Thông số huấn luyện
+Thiết lập	Giá trị
+Learning rate	5e-5
+Batch size (train/eval)	32 / 32
+Gradient accumulation	4
+Total batch size	128
+Optimizer	Adam (β=(0.9,0.999), ε=1e-8)
+Scheduler	Learning rate linearly giảm (warmup 0.1)
+Epochs	1
+Seed	42
+Framework	Transformers 4.17.0, PyTorch 1.10.0+cu111 
+
+## 📊 Hiệu quả
+Báo cáo accuracy đạt 97.88 % trên tập đánh giá CIFAR-10
+
+Thích hợp để đánh giá khả năng áp dụng ViT ở dữ liệu nhỏ và cải tiến nhanh timemodels
+
+## ✅ Lưu ý & Mở rộng
+Đây là mô hình để demo hoặc baseline, không phù hợp cho trường hợp cần độ chính xác cao hơn (có thể fine-tune thêm)
+
+Dễ dàng kết hợp với các kỹ thuật augment, thêm epochs, hoặc sử dụng data robustness khi cần
+
+Có thể dùng làm backbone để ép nhỏ, kiểm thử OOD, adversarial training, v.v.

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.9788,
+    "eval_loss": 0.25641629099845886,
+    "eval_runtime": 263.6472,
+    "eval_samples_per_second": 37.929,
+    "eval_steps_per_second": 1.187,
+    "total_flos": 3.86867749153407e+18,
+    "train_loss": 0.7974265074118589,
+    "train_runtime": 3555.934,
+    "train_samples_per_second": 14.061,
+    "train_steps_per_second": 0.11
+}

config.json

@@ -0,0 +1,48 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "airplane",
+    "1": "automobile",
+    "2": "bird",
+    "3": "cat",
+    "4": "deer",
+    "5": "dog",
+    "6": "frog",
+    "7": "horse",
+    "8": "ship",
+    "9": "truck"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "airplane": "0",
+    "automobile": "1",
+    "bird": "2",
+    "cat": "3",
+    "deer": "4",
+    "dog": "5",
+    "frog": "6",
+    "horse": "7",
+    "ship": "8",
+    "truck": "9"
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0"
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.9788,
+    "eval_loss": 0.25641629099845886,
+    "eval_runtime": 263.6472,
+    "eval_samples_per_second": 37.929,
+    "eval_steps_per_second": 1.187
+}

preprocessor_config.json

@@ -0,0 +1,17 @@
+{
+  "do_normalize": true,
+  "do_resize": true,
+  "feature_extractor_type": "ViTFeatureExtractor",
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "size": 224
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e738962ed1175c80087a29447e899372301113b4995a6fbe1127ad2a53d7ac4
+size 343305329

runs/Mar30_00-25-17_63aa4d2b9b7b/1648599984.214082/events.out.tfevents.1648599984.63aa4d2b9b7b.72.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03bb9e01d56d15e4a392190d0e8dc945e96742d2c7fbe32703c534179815443d
+size 4848

runs/Mar30_00-25-17_63aa4d2b9b7b/events.out.tfevents.1648599984.63aa4d2b9b7b.72.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9100f335986bd21d71d25d72f24b71981fa6d520383a8b1cdbb7cadb66df34a
+size 10428

runs/Mar30_00-25-17_63aa4d2b9b7b/events.out.tfevents.1648604459.63aa4d2b9b7b.72.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:515ff0376876bc4288a268c5e13124885a28f859f74ec720ab2be86d92968184
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 3.86867749153407e+18,
+    "train_loss": 0.7974265074118589,
+    "train_runtime": 3555.934,
+    "train_samples_per_second": 14.061,
+    "train_steps_per_second": 0.11
+}

trainer_state.json

@@ -0,0 +1,268 @@
+{
+  "best_metric": 0.9788,
+  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-cifar10/checkpoint-390",
+  "epoch": 0.9980806142034548,
+  "global_step": 390,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.282051282051282e-05,
+      "loss": 2.3249,
+      "step": 10
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.564102564102564e-05,
+      "loss": 2.2694,
+      "step": 20
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 2.1649,
+      "step": 30
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.985754985754986e-05,
+      "loss": 1.9757,
+      "step": 40
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.8433048433048433e-05,
+      "loss": 1.6938,
+      "step": 50
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.700854700854701e-05,
+      "loss": 1.4069,
+      "step": 60
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.558404558404559e-05,
+      "loss": 1.1938,
+      "step": 70
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.415954415954416e-05,
+      "loss": 1.045,
+      "step": 80
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.2735042735042735e-05,
+      "loss": 0.9435,
+      "step": 90
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.131054131054131e-05,
+      "loss": 0.8542,
+      "step": 100
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.988603988603989e-05,
+      "loss": 0.7705,
+      "step": 110
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.7295,
+      "step": 120
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.6826,
+      "step": 130
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.561253561253561e-05,
+      "loss": 0.6886,
+      "step": 140
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.418803418803419e-05,
+      "loss": 0.604,
+      "step": 150
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.2763532763532764e-05,
+      "loss": 0.5829,
+      "step": 160
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.133903133903134e-05,
+      "loss": 0.5991,
+      "step": 170
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.9914529914529915e-05,
+      "loss": 0.5614,
+      "step": 180
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 2.8490028490028492e-05,
+      "loss": 0.5863,
+      "step": 190
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 2.706552706552707e-05,
+      "loss": 0.5146,
+      "step": 200
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.564102564102564e-05,
+      "loss": 0.5278,
+      "step": 210
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.4216524216524217e-05,
+      "loss": 0.4996,
+      "step": 220
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 2.2792022792022794e-05,
+      "loss": 0.5387,
+      "step": 230
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.1367521367521368e-05,
+      "loss": 0.4954,
+      "step": 240
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9943019943019945e-05,
+      "loss": 0.5052,
+      "step": 250
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.4617,
+      "step": 260
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7094017094017095e-05,
+      "loss": 0.4812,
+      "step": 270
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.566951566951567e-05,
+      "loss": 0.4475,
+      "step": 280
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4245014245014246e-05,
+      "loss": 0.4786,
+      "step": 290
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.282051282051282e-05,
+      "loss": 0.4722,
+      "step": 300
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1396011396011397e-05,
+      "loss": 0.4365,
+      "step": 310
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.971509971509972e-06,
+      "loss": 0.4803,
+      "step": 320
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 8.547008547008548e-06,
+      "loss": 0.4504,
+      "step": 330
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 7.122507122507123e-06,
+      "loss": 0.4467,
+      "step": 340
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.6980056980056985e-06,
+      "loss": 0.4477,
+      "step": 350
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.273504273504274e-06,
+      "loss": 0.4257,
+      "step": 360
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.8490028490028492e-06,
+      "loss": 0.4344,
+      "step": 370
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.4245014245014246e-06,
+      "loss": 0.4493,
+      "step": 380
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0,
+      "loss": 0.4291,
+      "step": 390
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9788,
+      "eval_loss": 0.25641629099845886,
+      "eval_runtime": 262.0666,
+      "eval_samples_per_second": 38.158,
+      "eval_steps_per_second": 1.194,
+      "step": 390
+    },
+    {
+      "epoch": 1.0,
+      "step": 390,
+      "total_flos": 3.86867749153407e+18,
+      "train_loss": 0.7974265074118589,
+      "train_runtime": 3555.934,
+      "train_samples_per_second": 14.061,
+      "train_steps_per_second": 0.11
+    }
+  ],
+  "max_steps": 390,
+  "num_train_epochs": 1,
+  "total_flos": 3.86867749153407e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cdc7435c7264a2bb289df316efe90d25aca40256ed73e6e61296869d73ff057
+size 3055