End of training

README.md

@@ -0,0 +1,79 @@
+---
+base_model: microsoft/layoutlm-base-uncased
+tags:
+- generated_from_trainer
+datasets:
+- funsd
+model-index:
+- name: layoutlm-funsd
+  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. -->
+
+# layoutlm-funsd
+
+This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on the funsd dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7048
+- Answer: {'precision': 0.6989130434782609, 'recall': 0.7948084054388134, 'f1': 0.7437825332562175, 'number': 809}
+- Header: {'precision': 0.3049645390070922, 'recall': 0.36134453781512604, 'f1': 0.3307692307692308, 'number': 119}
+- Question: {'precision': 0.7769973661106233, 'recall': 0.8309859154929577, 'f1': 0.8030852994555354, 'number': 1065}
+- Overall Precision: 0.7141
+- Overall Recall: 0.7883
+- Overall F1: 0.7493
+- Overall Accuracy: 0.8041
+
+## 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: 3e-05
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 15
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Answer                                                                                                       | Header                                                                                                         | Question                                                                                                     | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:------------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| 1.8962        | 1.0   | 10   | 1.6314          | {'precision': 0.04096170970614425, 'recall': 0.05686032138442522, 'f1': 0.047619047619047616, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                    | {'precision': 0.21477162293488825, 'recall': 0.20751173708920187, 'f1': 0.21107927411652339, 'number': 1065} | 0.1241            | 0.1340         | 0.1288     | 0.3878           |
+| 1.4949        | 2.0   | 20   | 1.2748          | {'precision': 0.16384915474642392, 'recall': 0.1557478368355995, 'f1': 0.1596958174904943, 'number': 809}    | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                    | {'precision': 0.42642857142857143, 'recall': 0.5605633802816902, 'f1': 0.48438133874239353, 'number': 1065}  | 0.3333            | 0.3628         | 0.3474     | 0.5644           |
+| 1.1275        | 3.0   | 30   | 0.9637          | {'precision': 0.4481236203090508, 'recall': 0.5018541409147095, 'f1': 0.473469387755102, 'number': 809}      | {'precision': 0.030303030303030304, 'recall': 0.008403361344537815, 'f1': 0.013157894736842105, 'number': 119} | {'precision': 0.6073883161512027, 'recall': 0.6638497652582159, 'f1': 0.6343651861821444, 'number': 1065}    | 0.5297            | 0.5590         | 0.5439     | 0.6927           |
+| 0.8515        | 4.0   | 40   | 0.8074          | {'precision': 0.5718446601941748, 'recall': 0.7280593325092707, 'f1': 0.6405655247417075, 'number': 809}     | {'precision': 0.14492753623188406, 'recall': 0.08403361344537816, 'f1': 0.10638297872340426, 'number': 119}    | {'precision': 0.6351132686084142, 'recall': 0.7370892018779343, 'f1': 0.6823120382442416, 'number': 1065}    | 0.5927            | 0.6944         | 0.6396     | 0.7481           |
+| 0.6885        | 5.0   | 50   | 0.7250          | {'precision': 0.6376963350785341, 'recall': 0.7527812113720643, 'f1': 0.6904761904761906, 'number': 809}     | {'precision': 0.23529411764705882, 'recall': 0.13445378151260504, 'f1': 0.17112299465240638, 'number': 119}    | {'precision': 0.7158081705150977, 'recall': 0.7568075117370892, 'f1': 0.735737106344135, 'number': 1065}     | 0.6659            | 0.7180         | 0.6910     | 0.7705           |
+| 0.582         | 6.0   | 60   | 0.6796          | {'precision': 0.6464323748668797, 'recall': 0.7503090234857849, 'f1': 0.694508009153318, 'number': 809}      | {'precision': 0.2204724409448819, 'recall': 0.23529411764705882, 'f1': 0.22764227642276422, 'number': 119}     | {'precision': 0.6873015873015873, 'recall': 0.8131455399061033, 'f1': 0.7449462365591398, 'number': 1065}    | 0.6453            | 0.7531         | 0.6951     | 0.7920           |
+| 0.505         | 7.0   | 70   | 0.6522          | {'precision': 0.6307385229540918, 'recall': 0.7812113720642769, 'f1': 0.6979569298729985, 'number': 809}     | {'precision': 0.21367521367521367, 'recall': 0.21008403361344538, 'f1': 0.211864406779661, 'number': 119}      | {'precision': 0.7270450751252087, 'recall': 0.8178403755868544, 'f1': 0.7697746354396818, 'number': 1065}    | 0.6595            | 0.7667         | 0.7090     | 0.7973           |
+| 0.4537        | 8.0   | 80   | 0.6537          | {'precision': 0.6717391304347826, 'recall': 0.7639060568603214, 'f1': 0.714864083285136, 'number': 809}      | {'precision': 0.272, 'recall': 0.2857142857142857, 'f1': 0.27868852459016397, 'number': 119}                   | {'precision': 0.7433110367892977, 'recall': 0.8347417840375587, 'f1': 0.7863777089783283, 'number': 1065}    | 0.6876            | 0.7732         | 0.7279     | 0.8010           |
+| 0.3975        | 9.0   | 90   | 0.6624          | {'precision': 0.6649269311064718, 'recall': 0.7873918417799752, 'f1': 0.7209960384833051, 'number': 809}     | {'precision': 0.2676056338028169, 'recall': 0.31932773109243695, 'f1': 0.2911877394636015, 'number': 119}      | {'precision': 0.7576285963382737, 'recall': 0.815962441314554, 'f1': 0.7857142857142857, 'number': 1065}     | 0.6871            | 0.7747         | 0.7283     | 0.8007           |
+| 0.3619        | 10.0  | 100  | 0.6649          | {'precision': 0.6825053995680346, 'recall': 0.7812113720642769, 'f1': 0.7285302593659942, 'number': 809}     | {'precision': 0.3178294573643411, 'recall': 0.3445378151260504, 'f1': 0.33064516129032256, 'number': 119}      | {'precision': 0.7808098591549296, 'recall': 0.8328638497652582, 'f1': 0.8059972739663789, 'number': 1065}    | 0.7120            | 0.7827         | 0.7457     | 0.8012           |
+| 0.3266        | 11.0  | 110  | 0.6796          | {'precision': 0.6776246023329798, 'recall': 0.7898640296662547, 'f1': 0.7294520547945205, 'number': 809}     | {'precision': 0.2740740740740741, 'recall': 0.31092436974789917, 'f1': 0.29133858267716534, 'number': 119}     | {'precision': 0.7656116338751069, 'recall': 0.8403755868544601, 'f1': 0.801253357206804, 'number': 1065}     | 0.6992            | 0.7883         | 0.7410     | 0.8007           |
+| 0.3091        | 12.0  | 120  | 0.6863          | {'precision': 0.6908893709327549, 'recall': 0.7873918417799752, 'f1': 0.7359907567879839, 'number': 809}     | {'precision': 0.3157894736842105, 'recall': 0.35294117647058826, 'f1': 0.33333333333333337, 'number': 119}     | {'precision': 0.7680278019113814, 'recall': 0.8300469483568075, 'f1': 0.7978339350180504, 'number': 1065}    | 0.7085            | 0.7842         | 0.7445     | 0.8036           |
+| 0.2903        | 13.0  | 130  | 0.7025          | {'precision': 0.6961748633879782, 'recall': 0.7873918417799752, 'f1': 0.7389791183294663, 'number': 809}     | {'precision': 0.29931972789115646, 'recall': 0.3697478991596639, 'f1': 0.33082706766917297, 'number': 119}     | {'precision': 0.7734855136084284, 'recall': 0.8272300469483568, 'f1': 0.7994555353901996, 'number': 1065}    | 0.7097            | 0.7837         | 0.7449     | 0.8029           |
+| 0.2685        | 14.0  | 140  | 0.7073          | {'precision': 0.6958424507658644, 'recall': 0.7861557478368356, 'f1': 0.7382472431804992, 'number': 809}     | {'precision': 0.2896551724137931, 'recall': 0.35294117647058826, 'f1': 0.31818181818181823, 'number': 119}     | {'precision': 0.7727666955767563, 'recall': 0.8366197183098592, 'f1': 0.8034265103697025, 'number': 1065}    | 0.7093            | 0.7873         | 0.7463     | 0.8034           |
+| 0.2678        | 15.0  | 150  | 0.7048          | {'precision': 0.6989130434782609, 'recall': 0.7948084054388134, 'f1': 0.7437825332562175, 'number': 809}     | {'precision': 0.3049645390070922, 'recall': 0.36134453781512604, 'f1': 0.3307692307692308, 'number': 119}      | {'precision': 0.7769973661106233, 'recall': 0.8309859154929577, 'f1': 0.8030852994555354, 'number': 1065}    | 0.7141            | 0.7883         | 0.7493     | 0.8041           |
+
+
+### Framework versions
+
+- Transformers 4.34.1
+- Pytorch 2.1.0+cu118
+- Datasets 2.14.6
+- Tokenizers 0.14.1

preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "apply_ocr": true,
+  "do_resize": true,
+  "feature_extractor_type": "LayoutLMv2FeatureExtractor",
+  "image_processor_type": "LayoutLMv2ImageProcessor",
+  "ocr_lang": null,
+  "processor_class": "LayoutLMv2Processor",
+  "resample": 2,
+  "size": {
+    "height": 224,
+    "width": 224
+  },
+  "tesseract_config": ""
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,80 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "apply_ocr": false,
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "cls_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "only_label_first_subword": true,
+  "pad_token": "[PAD]",
+  "pad_token_box": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "pad_token_label": -100,
+  "processor_class": "LayoutLMv2Processor",
+  "sep_token": "[SEP]",
+  "sep_token_box": [
+    1000,
+    1000,
+    1000,
+    1000
+  ],
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "LayoutLMv2Tokenizer",
+  "unk_token": "[UNK]"
+}