End of training

README.md

@@ -16,15 +16,6 @@ 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: 1.0370
-- Answer: {'precision': 0.39444995044598613, 'recall': 0.4919653893695921, 'f1': 0.43784378437843785, 'number': 809}
-- Header: {'precision': 0.26548672566371684, 'recall': 0.25210084033613445, 'f1': 0.25862068965517243, 'number': 119}
-- Question: {'precision': 0.5051546391752577, 'recall': 0.644131455399061, 'f1': 0.566240198101527, 'number': 1065}
-- Overall Precision: 0.4492
-- Overall Recall: 0.5590
-- Overall F1: 0.4981
-- Overall Accuracy: 0.6347
 
 ## Model description
 
@@ -50,32 +41,10 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 15
-- mixed_precision_training: Native AMP
-
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss | Answer                                                                                                        | Header                                                                                                      | Question                                                                                                     | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:-------------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
-| 1.7638        | 1.0   | 10   | 1.5767          | {'precision': 0.02685284640171858, 'recall': 0.030902348578491966, 'f1': 0.028735632183908046, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                 | {'precision': 0.23186119873817035, 'recall': 0.13802816901408452, 'f1': 0.17304296645085343, 'number': 1065} | 0.1099            | 0.0863         | 0.0967     | 0.3463           |
-| 1.4838        | 2.0   | 20   | 1.3817          | {'precision': 0.20173267326732675, 'recall': 0.40296662546353523, 'f1': 0.2688659793814433, 'number': 809}    | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}                                                 | {'precision': 0.27599486521181, 'recall': 0.40375586854460094, 'f1': 0.3278688524590164, 'number': 1065}     | 0.2365            | 0.3793         | 0.2913     | 0.4223           |
-| 1.2933        | 3.0   | 30   | 1.2178          | {'precision': 0.22845691382765532, 'recall': 0.4227441285537701, 'f1': 0.29661751951431053, 'number': 809}    | {'precision': 0.1076923076923077, 'recall': 0.058823529411764705, 'f1': 0.07608695652173912, 'number': 119} | {'precision': 0.3385542168674699, 'recall': 0.5276995305164319, 'f1': 0.4124770642201835, 'number': 1065}    | 0.2827            | 0.4571         | 0.3494     | 0.4903           |
-| 1.1591        | 4.0   | 40   | 1.1134          | {'precision': 0.2775800711743772, 'recall': 0.4820766378244747, 'f1': 0.3523035230352303, 'number': 809}      | {'precision': 0.27848101265822783, 'recall': 0.18487394957983194, 'f1': 0.2222222222222222, 'number': 119}  | {'precision': 0.37922077922077924, 'recall': 0.5483568075117371, 'f1': 0.4483685220729367, 'number': 1065}   | 0.3294            | 0.4997         | 0.3971     | 0.5555           |
-| 1.0602        | 5.0   | 50   | 1.0998          | {'precision': 0.2914764079147641, 'recall': 0.47342398022249693, 'f1': 0.3608101742816769, 'number': 809}     | {'precision': 0.3146067415730337, 'recall': 0.23529411764705882, 'f1': 0.2692307692307692, 'number': 119}   | {'precision': 0.4076227390180879, 'recall': 0.5924882629107981, 'f1': 0.4829697665518562, 'number': 1065}    | 0.3531            | 0.5228         | 0.4215     | 0.5676           |
-| 0.9666        | 6.0   | 60   | 1.2008          | {'precision': 0.3184044786564031, 'recall': 0.5624227441285538, 'f1': 0.4066130473637176, 'number': 809}      | {'precision': 0.36666666666666664, 'recall': 0.18487394957983194, 'f1': 0.24581005586592175, 'number': 119} | {'precision': 0.46556233653007845, 'recall': 0.5014084507042254, 'f1': 0.4828209764918625, 'number': 1065}   | 0.3835            | 0.5073         | 0.4368     | 0.5518           |
-| 0.9177        | 7.0   | 70   | 1.0682          | {'precision': 0.33756345177664976, 'recall': 0.4932014833127318, 'f1': 0.4008036162732296, 'number': 809}     | {'precision': 0.2978723404255319, 'recall': 0.23529411764705882, 'f1': 0.2629107981220657, 'number': 119}   | {'precision': 0.4583333333333333, 'recall': 0.5784037558685446, 'f1': 0.5114155251141553, 'number': 1065}    | 0.3981            | 0.5233         | 0.4522     | 0.5949           |
-| 0.8429        | 8.0   | 80   | 1.0318          | {'precision': 0.35264227642276424, 'recall': 0.4289245982694685, 'f1': 0.3870607919687674, 'number': 809}     | {'precision': 0.2909090909090909, 'recall': 0.2689075630252101, 'f1': 0.2794759825327511, 'number': 119}    | {'precision': 0.4475703324808184, 'recall': 0.6572769953051644, 'f1': 0.5325218714340054, 'number': 1065}    | 0.4059            | 0.5414         | 0.4640     | 0.6117           |
-| 0.8062        | 9.0   | 90   | 1.0339          | {'precision': 0.37058823529411766, 'recall': 0.4672435105067985, 'f1': 0.41334062329141613, 'number': 809}    | {'precision': 0.275, 'recall': 0.2773109243697479, 'f1': 0.27615062761506276, 'number': 119}                | {'precision': 0.4913728432108027, 'recall': 0.6150234741784038, 'f1': 0.5462885738115096, 'number': 1065}    | 0.4311            | 0.5349         | 0.4774     | 0.6213           |
-| 0.7862        | 10.0  | 100  | 1.0501          | {'precision': 0.35284552845528455, 'recall': 0.5364647713226205, 'f1': 0.42569887199607653, 'number': 809}    | {'precision': 0.3048780487804878, 'recall': 0.21008403361344538, 'f1': 0.24875621890547264, 'number': 119}  | {'precision': 0.4933852140077821, 'recall': 0.5953051643192488, 'f1': 0.5395744680851065, 'number': 1065}    | 0.4209            | 0.5484         | 0.4763     | 0.6103           |
-| 0.7206        | 11.0  | 110  | 1.0571          | {'precision': 0.3714020427112349, 'recall': 0.49443757725587145, 'f1': 0.4241781548250265, 'number': 809}     | {'precision': 0.2962962962962963, 'recall': 0.2689075630252101, 'f1': 0.28193832599118945, 'number': 119}   | {'precision': 0.5184893784421715, 'recall': 0.6187793427230047, 'f1': 0.5642123287671232, 'number': 1065}    | 0.4442            | 0.5474         | 0.4904     | 0.6234           |
-| 0.7242        | 12.0  | 120  | 1.0352          | {'precision': 0.3682771194165907, 'recall': 0.49938195302843014, 'f1': 0.4239244491080797, 'number': 809}     | {'precision': 0.3333333333333333, 'recall': 0.2689075630252101, 'f1': 0.29767441860465116, 'number': 119}   | {'precision': 0.4966984592809978, 'recall': 0.6356807511737089, 'f1': 0.557660626029654, 'number': 1065}     | 0.4354            | 0.5585         | 0.4893     | 0.6251           |
-| 0.6858        | 13.0  | 130  | 1.0447          | {'precision': 0.37962962962962965, 'recall': 0.5067985166872683, 'f1': 0.4340921122286925, 'number': 809}     | {'precision': 0.2672413793103448, 'recall': 0.2605042016806723, 'f1': 0.26382978723404255, 'number': 119}   | {'precision': 0.5022058823529412, 'recall': 0.6413145539906103, 'f1': 0.5632989690721649, 'number': 1065}    | 0.4397            | 0.5640         | 0.4942     | 0.6298           |
-| 0.6549        | 14.0  | 140  | 1.0332          | {'precision': 0.3821892393320965, 'recall': 0.5092707045735476, 'f1': 0.43667196608373077, 'number': 809}     | {'precision': 0.3263157894736842, 'recall': 0.2605042016806723, 'f1': 0.2897196261682243, 'number': 119}    | {'precision': 0.5044843049327354, 'recall': 0.6338028169014085, 'f1': 0.5617977528089888, 'number': 1065}    | 0.4452            | 0.5610         | 0.4964     | 0.6354           |
-| 0.6462        | 15.0  | 150  | 1.0370          | {'precision': 0.39444995044598613, 'recall': 0.4919653893695921, 'f1': 0.43784378437843785, 'number': 809}    | {'precision': 0.26548672566371684, 'recall': 0.25210084033613445, 'f1': 0.25862068965517243, 'number': 119} | {'precision': 0.5051546391752577, 'recall': 0.644131455399061, 'f1': 0.566240198101527, 'number': 1065}      | 0.4492            | 0.5590         | 0.4981     | 0.6347           |
-
 
 ### Framework versions
 
-- Transformers 4.38.2
-- Pytorch 2.2.1+cu121
+- Transformers 4.39.0
+- Pytorch 2.2.1
 - Datasets 2.18.0
 - Tokenizers 0.15.2

logs/events.out.tfevents.1711077848.ethanmbp.lan.714.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfcce9d911c5b6c7ca0da8de977eab24edd6b0a649f7562b69856707a60cb560
-size 10464
+oid sha256:6d54371f32839295b4d4758a11e1fb02697742fe95557d1997bcdfaf7b0468e5
+size 15771

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:207a7b88b767ee79cc9f2f18009901834929cc2ed04fefd119abd719498b0dcf
+oid sha256:44d47fb4d81bb81cf165b696b83ee9db1fd8a00d4206d25d0efabb39198763a2
 size 450558212

preprocessor_config.json

@@ -1,4 +1,16 @@
 {
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "resample",
+    "apply_ocr",
+    "ocr_lang",
+    "tesseract_config",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
   "apply_ocr": true,
   "do_resize": true,
   "image_processor_type": "LayoutLMv2ImageProcessor",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55a66d0eec0191ab73e6094e16baaf0c1297c4a463a889d8969d8ad76ef93641
+oid sha256:9ddc249426746aa87fc67e93ccb4cb9c1e346b41f5e664bf12e540282111d766
 size 4920