Saving best model to hub

README.md

@@ -0,0 +1,78 @@
+---
+license: apache-2.0
+base_model: jordyvl/vit-base_rvl-cdip
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: vit-base_rvl_cdip_aurc
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# vit-base_rvl_cdip_aurc
+
+This model is a fine-tuned version of [jordyvl/vit-base_rvl-cdip](https://huggingface.co/jordyvl/vit-base_rvl-cdip) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2759
+- Accuracy: 0.893
+- Brier Loss: 0.1798
+- Nll: 0.8614
+- F1 Micro: 0.893
+- F1 Macro: 0.8928
+- Ece: 0.0750
+- Aurc: 0.0215
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Brier Loss | Nll    | F1 Micro | F1 Macro | Ece    | Aurc   |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------:|:------:|:--------:|:--------:|:------:|:------:|
+| 0.0303        | 1.0   | 500  | 0.1865          | 0.8795   | 0.1840     | 1.2087 | 0.8795   | 0.8791   | 0.0495 | 0.0241 |
+| 0.0262        | 2.0   | 1000 | 0.2146          | 0.8788   | 0.1909     | 1.1956 | 0.8788   | 0.8789   | 0.0603 | 0.0257 |
+| 0.0121        | 3.0   | 1500 | 0.2117          | 0.886    | 0.1799     | 1.0878 | 0.886    | 0.8865   | 0.0611 | 0.0230 |
+| 0.0057        | 4.0   | 2000 | 0.2279          | 0.8878   | 0.1803     | 1.0108 | 0.8878   | 0.8879   | 0.0678 | 0.0228 |
+| 0.0038        | 5.0   | 2500 | 0.2491          | 0.8872   | 0.1827     | 0.9661 | 0.8872   | 0.8877   | 0.0725 | 0.0234 |
+| 0.0028        | 6.0   | 3000 | 0.2398          | 0.89     | 0.1806     | 0.9378 | 0.89     | 0.8901   | 0.0725 | 0.0215 |
+| 0.0016        | 7.0   | 3500 | 0.2736          | 0.891    | 0.1792     | 0.8975 | 0.891    | 0.8914   | 0.0744 | 0.0221 |
+| 0.0014        | 8.0   | 4000 | 0.2357          | 0.8905   | 0.1811     | 0.8993 | 0.8905   | 0.8910   | 0.0764 | 0.0210 |
+| 0.001         | 9.0   | 4500 | 0.2714          | 0.8898   | 0.1807     | 0.8650 | 0.8898   | 0.8897   | 0.0783 | 0.0213 |
+| 0.0009        | 10.0  | 5000 | 0.2759          | 0.893    | 0.1798     | 0.8614 | 0.893    | 0.8928   | 0.0750 | 0.0215 |
+
+
+### Framework versions
+
+- Transformers 4.33.3
+- Pytorch 2.2.0.dev20231002
+- Datasets 2.7.1
+- Tokenizers 0.13.3

config.json

@@ -0,0 +1,60 @@
+{
+  "_name_or_path": "jordyvl/vit-base_rvl-cdip",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "letter",
+    "1": "form",
+    "2": "email",
+    "3": "handwritten",
+    "4": "advertisement",
+    "5": "scientific_report",
+    "6": "scientific_publication",
+    "7": "specification",
+    "8": "file_folder",
+    "9": "news_article",
+    "10": "budget",
+    "11": "invoice",
+    "12": "presentation",
+    "13": "questionnaire",
+    "14": "resume",
+    "15": "memo"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "advertisement": 4,
+    "budget": 10,
+    "email": 2,
+    "file_folder": 8,
+    "form": 1,
+    "handwritten": 3,
+    "invoice": 11,
+    "letter": 0,
+    "memo": 15,
+    "news_article": 9,
+    "presentation": 12,
+    "questionnaire": 13,
+    "resume": 14,
+    "scientific_publication": 6,
+    "scientific_report": 5,
+    "specification": 7
+  },
+  "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": "4.33.3"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea866a5ceacf37051663d8828bc2b38f54e65c0e689f6032d9d409365e3c886
+size 343312234

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4992def92b48a094b2babbbbc5c2225b67c4dea89b30dbee94c238986a4df32e
+size 4856