napsternxg commited on
Commit
5917a24
1 Parent(s): 2c3961c

End of training

Browse files
README.md ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: jinaai/jina-embeddings-v2-small-en
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - nyt_ingredients
8
+ model-index:
9
+ - name: nyt-ingredient-tagger-jina-embeddings-v2-small-en
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # nyt-ingredient-tagger-jina-embeddings-v2-small-en
17
+
18
+ This model is a fine-tuned version of [jinaai/jina-embeddings-v2-small-en](https://huggingface.co/jinaai/jina-embeddings-v2-small-en) on the nyt_ingredients dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.9890
21
+ - Comment: {'precision': 0.4891238056515552, 'recall': 0.6700083542188805, 'f1': 0.5654524089306698, 'number': 7182}
22
+ - Name: {'precision': 0.7393011781290907, 'recall': 0.7889533634214485, 'f1': 0.7633206840983521, 'number': 9306}
23
+ - Qty: {'precision': 0.9253731343283582, 'recall': 0.9613688009624382, 'f1': 0.943027601127647, 'number': 7481}
24
+ - Range End: {'precision': 0.5454545454545454, 'recall': 0.5121951219512195, 'f1': 0.5283018867924528, 'number': 82}
25
+ - Unit: {'precision': 0.9031507061927674, 'recall': 0.9693486590038314, 'f1': 0.9350795436284751, 'number': 6003}
26
+ - Overall Precision: 0.7401
27
+ - Overall Recall: 0.8387
28
+ - Overall F1: 0.7863
29
+ - Overall Accuracy: 0.7817
30
+
31
+ ## Model description
32
+
33
+ More information needed
34
+
35
+ ## Intended uses & limitations
36
+
37
+ More information needed
38
+
39
+ ## Training and evaluation data
40
+
41
+ More information needed
42
+
43
+ ## Training procedure
44
+
45
+ ### Training hyperparameters
46
+
47
+ The following hyperparameters were used during training:
48
+ - learning_rate: 5e-05
49
+ - train_batch_size: 32
50
+ - eval_batch_size: 32
51
+ - seed: 42
52
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
+ - lr_scheduler_type: linear
54
+ - num_epochs: 5
55
+ - label_smoothing_factor: 0.1
56
+
57
+ ### Training results
58
+
59
+ | Training Loss | Epoch | Step | Validation Loss | Comment | Name | Qty | Range End | Unit | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
60
+ |:-------------:|:-----:|:-----:|:---------------:|:----------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:----------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
61
+ | 1.1585 | 0.2 | 1000 | 1.1247 | {'precision': 0.38455309241826097, 'recall': 0.5557343475716794, 'f1': 0.454561770864493, 'number': 6836} | {'precision': 0.6500338458563002, 'recall': 0.7587763855965685, 'f1': 0.7002083333333333, 'number': 8859} | {'precision': 0.8947789025039957, 'recall': 0.9468639887244539, 'f1': 0.9200849140587551, 'number': 7095} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 74} | {'precision': 0.8575376112987412, 'recall': 0.9760615062030403, 'f1': 0.912968864917872, 'number': 5723} | 0.6666 | 0.7984 | 0.7266 | 0.7220 |
62
+ | 1.1018 | 0.4 | 2000 | 1.0677 | {'precision': 0.3960273712222011, 'recall': 0.6095669982445875, 'f1': 0.48012443829934326, 'number': 6836} | {'precision': 0.7094423320659062, 'recall': 0.7582119878090078, 'f1': 0.733016860369946, 'number': 8859} | {'precision': 0.8986009327115256, 'recall': 0.9505285412262157, 'f1': 0.9238356164383562, 'number': 7095} | {'precision': 0.2, 'recall': 0.02702702702702703, 'f1': 0.047619047619047616, 'number': 74} | {'precision': 0.8794009877329935, 'recall': 0.9645290931329722, 'f1': 0.9199999999999999, 'number': 5723} | 0.6853 | 0.8098 | 0.7424 | 0.7415 |
63
+ | 1.0676 | 0.59 | 3000 | 1.0472 | {'precision': 0.41734173417341736, 'recall': 0.6104447045055588, 'f1': 0.4957528957528958, 'number': 6836} | {'precision': 0.7021688219122288, 'recall': 0.7784174286036799, 'f1': 0.7383297644539616, 'number': 8859} | {'precision': 0.8949375410913872, 'recall': 0.9592670894996477, 'f1': 0.9259863945578231, 'number': 7095} | {'precision': 0.5161290322580645, 'recall': 0.21621621621621623, 'f1': 0.3047619047619048, 'number': 74} | {'precision': 0.8788844621513944, 'recall': 0.9636554254761489, 'f1': 0.9193198866477745, 'number': 5723} | 0.6939 | 0.8188 | 0.7512 | 0.7541 |
64
+ | 1.0613 | 0.79 | 4000 | 1.0459 | {'precision': 0.4413024850042845, 'recall': 0.6026916325336454, 'f1': 0.5095226317091268, 'number': 6836} | {'precision': 0.7297499465697799, 'recall': 0.7708544982503669, 'f1': 0.7497392545424604, 'number': 8859} | {'precision': 0.9064651100013497, 'recall': 0.9465821000704722, 'f1': 0.9260893546607832, 'number': 7095} | {'precision': 0.34210526315789475, 'recall': 0.17567567567567569, 'f1': 0.23214285714285715, 'number': 74} | {'precision': 0.8965631196298744, 'recall': 0.9481041411846933, 'f1': 0.9216135881104034, 'number': 5723} | 0.7177 | 0.8082 | 0.7603 | 0.7502 |
65
+ | 1.045 | 0.99 | 5000 | 1.0292 | {'precision': 0.43983577218654596, 'recall': 0.6111761263897015, 'f1': 0.5115396388123661, 'number': 6836} | {'precision': 0.7188987787207618, 'recall': 0.7840614064792866, 'f1': 0.7500674909562118, 'number': 8859} | {'precision': 0.886005680351149, 'recall': 0.9673009161381254, 'f1': 0.9248702917593155, 'number': 7095} | {'precision': 0.3541666666666667, 'recall': 0.22972972972972974, 'f1': 0.27868852459016397, 'number': 74} | {'precision': 0.8777340676632572, 'recall': 0.974663637952123, 'f1': 0.9236628580890875, 'number': 5723} | 0.7080 | 0.8249 | 0.7620 | 0.7610 |
66
+ | 1.0334 | 1.19 | 6000 | 1.0344 | {'precision': 0.47399084477736164, 'recall': 0.6664716208308953, 'f1': 0.5539883268482491, 'number': 6836} | {'precision': 0.7198329853862213, 'recall': 0.7784174286036799, 'f1': 0.7479798253701395, 'number': 8859} | {'precision': 0.9296510806611104, 'recall': 0.927554615926709, 'f1': 0.92860166502046, 'number': 7095} | {'precision': 0.36666666666666664, 'recall': 0.2972972972972973, 'f1': 0.3283582089552239, 'number': 74} | {'precision': 0.8982691051600261, 'recall': 0.9612091560370435, 'f1': 0.9286739258884106, 'number': 5723} | 0.7258 | 0.8240 | 0.7718 | 0.7595 |
67
+ | 1.0187 | 1.39 | 7000 | 1.0210 | {'precision': 0.4423198816818086, 'recall': 0.6124926857811586, 'f1': 0.5136793031529874, 'number': 6836} | {'precision': 0.7155410238070911, 'recall': 0.7904955412574782, 'f1': 0.751153062318996, 'number': 8859} | {'precision': 0.8767850372804247, 'recall': 0.9778717406624383, 'f1': 0.9245735607675907, 'number': 7095} | {'precision': 0.37142857142857144, 'recall': 0.17567567567567569, 'f1': 0.23853211009174313, 'number': 74} | {'precision': 0.8888354957552459, 'recall': 0.9695963655425476, 'f1': 0.927461139896373, 'number': 5723} | 0.7083 | 0.8287 | 0.7638 | 0.7651 |
68
+ | 1.0319 | 1.58 | 8000 | 1.0136 | {'precision': 0.46955690149824675, 'recall': 0.6464306612053833, 'f1': 0.5439773496645535, 'number': 6836} | {'precision': 0.7399957428693061, 'recall': 0.7848515633818716, 'f1': 0.7617639003012875, 'number': 8859} | {'precision': 0.8963893249607535, 'recall': 0.9657505285412262, 'f1': 0.9297781396295542, 'number': 7095} | {'precision': 0.45098039215686275, 'recall': 0.3108108108108108, 'f1': 0.368, 'number': 74} | {'precision': 0.8991981672394044, 'recall': 0.9601607548488555, 'f1': 0.9286800743620078, 'number': 5723} | 0.7280 | 0.8305 | 0.7759 | 0.7700 |
69
+ | 1.0154 | 1.78 | 9000 | 1.0071 | {'precision': 0.47295907875796833, 'recall': 0.6729081334113517, 'f1': 0.5554884675763797, 'number': 6836} | {'precision': 0.7443054218800128, 'recall': 0.7856417202844564, 'f1': 0.7644151565074134, 'number': 8859} | {'precision': 0.9104236718224613, 'recall': 0.9540521494009866, 'f1': 0.9317274604267033, 'number': 7095} | {'precision': 0.4126984126984127, 'recall': 0.35135135135135137, 'f1': 0.3795620437956204, 'number': 74} | {'precision': 0.8885522959183674, 'recall': 0.9737899702952997, 'f1': 0.9292205085452273, 'number': 5723} | 0.7285 | 0.8370 | 0.7790 | 0.7732 |
70
+ | 1.011 | 1.98 | 10000 | 1.0127 | {'precision': 0.4703654417033473, 'recall': 0.6721767115272089, 'f1': 0.5534477566997892, 'number': 6836} | {'precision': 0.742723104808615, 'recall': 0.7863189976295293, 'f1': 0.7638995503892971, 'number': 8859} | {'precision': 0.915743991358358, 'recall': 0.9558844256518675, 'f1': 0.9353837666367836, 'number': 7095} | {'precision': 0.39285714285714285, 'recall': 0.2972972972972973, 'f1': 0.3384615384615385, 'number': 74} | {'precision': 0.901291060630822, 'recall': 0.9636554254761489, 'f1': 0.9314305016044586, 'number': 5723} | 0.7296 | 0.8353 | 0.7789 | 0.7712 |
71
+ | 0.9958 | 2.18 | 11000 | 1.0024 | {'precision': 0.4715127701375246, 'recall': 0.6670567583382094, 'f1': 0.5524928818077179, 'number': 6836} | {'precision': 0.7483029845921776, 'recall': 0.7839485269217744, 'f1': 0.7657111356119073, 'number': 8859} | {'precision': 0.9206457791763579, 'recall': 0.9484143763213531, 'f1': 0.9343237989447376, 'number': 7095} | {'precision': 0.35, 'recall': 0.3783783783783784, 'f1': 0.36363636363636365, 'number': 74} | {'precision': 0.8947876447876448, 'recall': 0.9718679014502883, 'f1': 0.9317363263254879, 'number': 5723} | 0.7318 | 0.8334 | 0.7793 | 0.7763 |
72
+ | 1.0042 | 2.38 | 12000 | 1.0007 | {'precision': 0.4789602641951635, 'recall': 0.657694558221182, 'f1': 0.5542747950440732, 'number': 6836} | {'precision': 0.7298981923955953, 'recall': 0.7930917710802574, 'f1': 0.7601839329185826, 'number': 8859} | {'precision': 0.9267211525141986, 'recall': 0.9429175475687104, 'f1': 0.9347491965907503, 'number': 7095} | {'precision': 0.35555555555555557, 'recall': 0.43243243243243246, 'f1': 0.3902439024390244, 'number': 74} | {'precision': 0.9089250165892502, 'recall': 0.9573650183470208, 'f1': 0.932516381584546, 'number': 5723} | 0.7333 | 0.8299 | 0.7786 | 0.7769 |
73
+ | 1.0048 | 2.57 | 13000 | 0.9943 | {'precision': 0.47168994262206343, 'recall': 0.6373610298420129, 'f1': 0.542151434082001, 'number': 6836} | {'precision': 0.7342649994746243, 'recall': 0.7888023478947963, 'f1': 0.7605572485851111, 'number': 8859} | {'precision': 0.8968563263185243, 'recall': 0.9730796335447498, 'f1': 0.9334144527817211, 'number': 7095} | {'precision': 0.4307692307692308, 'recall': 0.3783783783783784, 'f1': 0.4028776978417266, 'number': 74} | {'precision': 0.8915373540233562, 'recall': 0.9737899702952997, 'f1': 0.9308501753799899, 'number': 5723} | 0.7278 | 0.8343 | 0.7774 | 0.7787 |
74
+ | 0.9911 | 2.77 | 14000 | 0.9951 | {'precision': 0.4768817204301075, 'recall': 0.6487712112346401, 'f1': 0.5497025285076846, 'number': 6836} | {'precision': 0.729257190151045, 'recall': 0.7956880009030365, 'f1': 0.7610256410256411, 'number': 8859} | {'precision': 0.9042483230303827, 'recall': 0.9689922480620154, 'f1': 0.9355014287658184, 'number': 7095} | {'precision': 0.5319148936170213, 'recall': 0.33783783783783783, 'f1': 0.4132231404958678, 'number': 74} | {'precision': 0.8965628529933839, 'recall': 0.9708195002621003, 'f1': 0.9322147651006711, 'number': 5723} | 0.7296 | 0.8374 | 0.7798 | 0.7786 |
75
+ | 0.9991 | 2.97 | 15000 | 0.9921 | {'precision': 0.4791033832617576, 'recall': 0.6691047396138092, 'f1': 0.5583836904107916, 'number': 6836} | {'precision': 0.7481054541573273, 'recall': 0.7911728186025511, 'f1': 0.7690366469168313, 'number': 8859} | {'precision': 0.9137861466039005, 'recall': 0.9575757575757575, 'f1': 0.935168616655196, 'number': 7095} | {'precision': 0.4166666666666667, 'recall': 0.40540540540540543, 'f1': 0.4109589041095891, 'number': 74} | {'precision': 0.8894720101781171, 'recall': 0.977284640922593, 'f1': 0.9313129631171426, 'number': 5723} | 0.7337 | 0.8395 | 0.7831 | 0.7807 |
76
+ | 0.9805 | 3.17 | 16000 | 0.9880 | {'precision': 0.4859154929577465, 'recall': 0.6560854300760679, 'f1': 0.5583219220714553, 'number': 6836} | {'precision': 0.7423922231614539, 'recall': 0.7930917710802574, 'f1': 0.7669049828084921, 'number': 8859} | {'precision': 0.9187102018696653, 'recall': 0.9557434813248766, 'f1': 0.9368610113290964, 'number': 7095} | {'precision': 0.4444444444444444, 'recall': 0.3783783783783784, 'f1': 0.4087591240875913, 'number': 74} | {'precision': 0.9000486775920817, 'recall': 0.9692468984798183, 'f1': 0.9333669863705198, 'number': 5723} | 0.7389 | 0.8349 | 0.7840 | 0.7822 |
77
+ | 0.9848 | 3.37 | 17000 | 0.9842 | {'precision': 0.48933174482833314, 'recall': 0.6609128145114102, 'f1': 0.5623249735515589, 'number': 6836} | {'precision': 0.7466623945316672, 'recall': 0.7891409865673327, 'f1': 0.7673142355394577, 'number': 8859} | {'precision': 0.9149737656397148, 'recall': 0.9585623678646934, 'f1': 0.936261013215859, 'number': 7095} | {'precision': 0.4126984126984127, 'recall': 0.35135135135135137, 'f1': 0.3795620437956204, 'number': 74} | {'precision': 0.899171943497321, 'recall': 0.9676742966975362, 'f1': 0.9321663019693655, 'number': 5723} | 0.7403 | 0.8351 | 0.7848 | 0.7824 |
78
+ | 0.9771 | 3.56 | 18000 | 0.9834 | {'precision': 0.4883396023643203, 'recall': 0.6647162083089526, 'f1': 0.5630382256365777, 'number': 6836} | {'precision': 0.7373874816830647, 'recall': 0.795236482672988, 'f1': 0.7652202248411449, 'number': 8859} | {'precision': 0.9162388543636855, 'recall': 0.9558844256518675, 'f1': 0.9356418569359177, 'number': 7095} | {'precision': 0.4583333333333333, 'recall': 0.44594594594594594, 'f1': 0.4520547945205479, 'number': 74} | {'precision': 0.8992864093415505, 'recall': 0.968897431417089, 'f1': 0.9327950206072841, 'number': 5723} | 0.7369 | 0.8378 | 0.7841 | 0.7837 |
79
+ | 0.9787 | 3.76 | 19000 | 0.9832 | {'precision': 0.4892808110676946, 'recall': 0.677735517846694, 'f1': 0.5682919349892671, 'number': 6836} | {'precision': 0.7466029723991507, 'recall': 0.7938819279828423, 'f1': 0.7695169319984682, 'number': 8859} | {'precision': 0.9206090266449157, 'recall': 0.9544749823819592, 'f1': 0.9372361774271676, 'number': 7095} | {'precision': 0.4189189189189189, 'recall': 0.4189189189189189, 'f1': 0.4189189189189189, 'number': 74} | {'precision': 0.9048244174597965, 'recall': 0.9634806919447843, 'f1': 0.9332317847169331, 'number': 5723} | 0.7399 | 0.8389 | 0.7863 | 0.7844 |
80
+ | 0.9746 | 3.96 | 20000 | 0.9827 | {'precision': 0.4950890447922288, 'recall': 0.6710064365125804, 'f1': 0.5697782746413266, 'number': 6836} | {'precision': 0.7460368124268539, 'recall': 0.7915114572750874, 'f1': 0.768101654069449, 'number': 8859} | {'precision': 0.9120629837203096, 'recall': 0.9633544749823819, 'f1': 0.9370073342929603, 'number': 7095} | {'precision': 0.40963855421686746, 'recall': 0.4594594594594595, 'f1': 0.43312101910828027, 'number': 74} | {'precision': 0.9003893575600259, 'recall': 0.9697710990739122, 'f1': 0.9337932194834694, 'number': 5723} | 0.7412 | 0.8402 | 0.7876 | 0.7846 |
81
+ | 0.976 | 4.16 | 21000 | 0.9836 | {'precision': 0.4884607241160279, 'recall': 0.6749561146869514, 'f1': 0.5667608401916225, 'number': 6836} | {'precision': 0.7483774869666986, 'recall': 0.7939948075403545, 'f1': 0.7705115565779385, 'number': 8859} | {'precision': 0.9147193123824873, 'recall': 0.9599718111346018, 'f1': 0.936799394814662, 'number': 7095} | {'precision': 0.4125, 'recall': 0.44594594594594594, 'f1': 0.42857142857142855, 'number': 74} | {'precision': 0.9035317200784827, 'recall': 0.9655774943211602, 'f1': 0.9335247909451813, 'number': 5723} | 0.7393 | 0.8402 | 0.7865 | 0.7854 |
82
+ | 0.9635 | 4.36 | 22000 | 0.9832 | {'precision': 0.49533612093920876, 'recall': 0.6758338209479228, 'f1': 0.5716760502381983, 'number': 6836} | {'precision': 0.7475583864118895, 'recall': 0.7948978440004515, 'f1': 0.7705016685814322, 'number': 8859} | {'precision': 0.9186408555570597, 'recall': 0.9564482029598309, 'f1': 0.9371633752244165, 'number': 7095} | {'precision': 0.391304347826087, 'recall': 0.4864864864864865, 'f1': 0.43373493975903615, 'number': 74} | {'precision': 0.9026418786692759, 'recall': 0.9671500961034423, 'f1': 0.9337832138338253, 'number': 5723} | 0.7423 | 0.8402 | 0.7882 | 0.7851 |
83
+ | 0.9688 | 4.55 | 23000 | 0.9836 | {'precision': 0.4930739135032251, 'recall': 0.6821240491515506, 'f1': 0.5723930522310194, 'number': 6836} | {'precision': 0.7467897697124058, 'recall': 0.7943334462128908, 'f1': 0.7698282463625424, 'number': 8859} | {'precision': 0.9208593962469405, 'recall': 0.9544749823819592, 'f1': 0.9373659076752716, 'number': 7095} | {'precision': 0.35714285714285715, 'recall': 0.5405405405405406, 'f1': 0.4301075268817204, 'number': 74} | {'precision': 0.9021013194331324, 'recall': 0.9676742966975362, 'f1': 0.9337379868487607, 'number': 5723} | 0.7403 | 0.8413 | 0.7876 | 0.7860 |
84
+ | 0.9669 | 4.75 | 24000 | 0.9803 | {'precision': 0.4968897468897469, 'recall': 0.677735517846694, 'f1': 0.5733910891089109, 'number': 6836} | {'precision': 0.7478168264110756, 'recall': 0.7926402528502088, 'f1': 0.7695764151460354, 'number': 8859} | {'precision': 0.9173631706659477, 'recall': 0.9591261451726568, 'f1': 0.9377799214497348, 'number': 7095} | {'precision': 0.4, 'recall': 0.4864864864864865, 'f1': 0.43902439024390244, 'number': 74} | {'precision': 0.8998864189518092, 'recall': 0.9690721649484536, 'f1': 0.9331987211845869, 'number': 5723} | 0.7424 | 0.8410 | 0.7886 | 0.7873 |
85
+ | 0.9691 | 4.95 | 25000 | 0.9796 | {'precision': 0.4962978860392746, 'recall': 0.6765652428320655, 'f1': 0.5725781491798204, 'number': 6836} | {'precision': 0.7474457215836526, 'recall': 0.7927531324077209, 'f1': 0.7694330320460149, 'number': 8859} | {'precision': 0.9183783783783783, 'recall': 0.9578576462297392, 'f1': 0.9377026560883062, 'number': 7095} | {'precision': 0.4065934065934066, 'recall': 0.5, 'f1': 0.4484848484848485, 'number': 74} | {'precision': 0.9002599090318388, 'recall': 0.968373230822995, 'f1': 0.933075174678003, 'number': 5723} | 0.7423 | 0.8403 | 0.7883 | 0.7871 |
86
+
87
+
88
+ ### Framework versions
89
+
90
+ - Transformers 4.34.1
91
+ - Pytorch 2.1.0+cu118
92
+ - Datasets 2.14.6
93
+ - Tokenizers 0.14.1
all_results.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_COMMENT": {
4
+ "f1": 0.5654524089306698,
5
+ "number": 7182,
6
+ "precision": 0.4891238056515552,
7
+ "recall": 0.6700083542188805
8
+ },
9
+ "eval_NAME": {
10
+ "f1": 0.7633206840983521,
11
+ "number": 9306,
12
+ "precision": 0.7393011781290907,
13
+ "recall": 0.7889533634214485
14
+ },
15
+ "eval_QTY": {
16
+ "f1": 0.943027601127647,
17
+ "number": 7481,
18
+ "precision": 0.9253731343283582,
19
+ "recall": 0.9613688009624382
20
+ },
21
+ "eval_RANGE_END": {
22
+ "f1": 0.5283018867924528,
23
+ "number": 82,
24
+ "precision": 0.5454545454545454,
25
+ "recall": 0.5121951219512195
26
+ },
27
+ "eval_UNIT": {
28
+ "f1": 0.9350795436284751,
29
+ "number": 6003,
30
+ "precision": 0.9031507061927674,
31
+ "recall": 0.9693486590038314
32
+ },
33
+ "eval_loss": 0.989044725894928,
34
+ "eval_overall_accuracy": 0.7816934549666956,
35
+ "eval_overall_f1": 0.7863058566638071,
36
+ "eval_overall_precision": 0.7400546079093391,
37
+ "eval_overall_recall": 0.8387236307978971,
38
+ "eval_runtime": 11.1131,
39
+ "eval_samples_per_second": 805.715,
40
+ "eval_steps_per_second": 25.195
41
+ }
config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "jinaai/jina-embeddings-v2-small-en",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "auto_map": {
8
+ "AutoConfig": "jinaai/jina-bert-implementation--configuration_bert.JinaBertConfig",
9
+ "AutoModel": "jinaai/jina-bert-implementation--modeling_bert.JinaBertModel",
10
+ "AutoModelForMaskedLM": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForMaskedLM",
11
+ "AutoModelForSequenceClassification": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForSequenceClassification"
12
+ },
13
+ "classifier_dropout": null,
14
+ "emb_pooler": "mean",
15
+ "feed_forward_type": "geglu",
16
+ "gradient_checkpointing": false,
17
+ "hidden_act": "gelu",
18
+ "hidden_dropout_prob": 0.1,
19
+ "hidden_size": 512,
20
+ "id2label": {
21
+ "0": "O",
22
+ "1": "B-COMMENT",
23
+ "2": "I-COMMENT",
24
+ "3": "B-NAME",
25
+ "4": "I-NAME",
26
+ "5": "B-RANGE_END",
27
+ "6": "I-RANGE_END",
28
+ "7": "B-QTY",
29
+ "8": "I-QTY",
30
+ "9": "B-UNIT",
31
+ "10": "I-UNIT"
32
+ },
33
+ "initializer_range": 0.02,
34
+ "intermediate_size": 2048,
35
+ "label2id": {
36
+ "B-COMMENT": 1,
37
+ "B-NAME": 3,
38
+ "B-QTY": 7,
39
+ "B-RANGE_END": 5,
40
+ "B-UNIT": 9,
41
+ "I-COMMENT": 2,
42
+ "I-NAME": 4,
43
+ "I-QTY": 8,
44
+ "I-RANGE_END": 6,
45
+ "I-UNIT": 10,
46
+ "O": 0
47
+ },
48
+ "layer_norm_eps": 1e-12,
49
+ "max_position_embeddings": 8192,
50
+ "model_max_length": 8192,
51
+ "model_type": "bert",
52
+ "num_attention_heads": 8,
53
+ "num_hidden_layers": 4,
54
+ "pad_token_id": 0,
55
+ "position_embedding_type": "alibi",
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.34.1",
58
+ "type_vocab_size": 2,
59
+ "use_cache": true,
60
+ "vocab_size": 30528
61
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1abff2a50c614cd8ebad2c1baad5965797e19f74ade57feb55e2d3ebe3463ace
3
+ size 129791747
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
test_results.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_COMMENT": {
4
+ "f1": 0.5654524089306698,
5
+ "number": 7182,
6
+ "precision": 0.4891238056515552,
7
+ "recall": 0.6700083542188805
8
+ },
9
+ "eval_NAME": {
10
+ "f1": 0.7633206840983521,
11
+ "number": 9306,
12
+ "precision": 0.7393011781290907,
13
+ "recall": 0.7889533634214485
14
+ },
15
+ "eval_QTY": {
16
+ "f1": 0.943027601127647,
17
+ "number": 7481,
18
+ "precision": 0.9253731343283582,
19
+ "recall": 0.9613688009624382
20
+ },
21
+ "eval_RANGE_END": {
22
+ "f1": 0.5283018867924528,
23
+ "number": 82,
24
+ "precision": 0.5454545454545454,
25
+ "recall": 0.5121951219512195
26
+ },
27
+ "eval_UNIT": {
28
+ "f1": 0.9350795436284751,
29
+ "number": 6003,
30
+ "precision": 0.9031507061927674,
31
+ "recall": 0.9693486590038314
32
+ },
33
+ "eval_loss": 0.989044725894928,
34
+ "eval_overall_accuracy": 0.7816934549666956,
35
+ "eval_overall_f1": 0.7863058566638071,
36
+ "eval_overall_precision": 0.7400546079093391,
37
+ "eval_overall_recall": 0.8387236307978971,
38
+ "eval_runtime": 11.1131,
39
+ "eval_samples_per_second": 805.715,
40
+ "eval_steps_per_second": 25.195
41
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 8192,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
train_results.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_COMMENT": {
4
+ "f1": 0.5958297497981183,
5
+ "number": 129770,
6
+ "precision": 0.5190033397383109,
7
+ "recall": 0.6993527009324189
8
+ },
9
+ "eval_NAME": {
10
+ "f1": 0.7825777582371811,
11
+ "number": 167635,
12
+ "precision": 0.7596190561757373,
13
+ "recall": 0.8069675187162585
14
+ },
15
+ "eval_QTY": {
16
+ "f1": 0.9442199159942841,
17
+ "number": 135888,
18
+ "precision": 0.9263431419751338,
19
+ "recall": 0.9628002472624514
20
+ },
21
+ "eval_RANGE_END": {
22
+ "f1": 0.5635263612791703,
23
+ "number": 1708,
24
+ "precision": 0.554736245036869,
25
+ "recall": 0.572599531615925
26
+ },
27
+ "eval_UNIT": {
28
+ "f1": 0.9371109281052695,
29
+ "number": 108827,
30
+ "precision": 0.9042646820073313,
31
+ "recall": 0.9724333115862791
32
+ },
33
+ "eval_loss": 0.9513349533081055,
34
+ "eval_overall_accuracy": 0.7994206201354097,
35
+ "eval_overall_f1": 0.8016365751906543,
36
+ "eval_overall_precision": 0.7564202956712965,
37
+ "eval_overall_recall": 0.8526022933721692,
38
+ "eval_runtime": 171.1672,
39
+ "eval_samples_per_second": 944.123,
40
+ "eval_steps_per_second": 29.509
41
+ }
trainer_state.json ADDED
@@ -0,0 +1,1378 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9796159267425537,
3
+ "best_model_checkpoint": "nyt-ingredient-tagger-jina-embeddings-v2-small-en/checkpoint-25000",
4
+ "epoch": 5.0,
5
+ "eval_steps": 1000,
6
+ "global_step": 25255,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.1,
13
+ "learning_rate": 4.9010097010492975e-05,
14
+ "loss": 1.3131,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.2,
19
+ "learning_rate": 4.802019402098595e-05,
20
+ "loss": 1.1585,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.2,
25
+ "eval_COMMENT": {
26
+ "f1": 0.454561770864493,
27
+ "number": 6836,
28
+ "precision": 0.38455309241826097,
29
+ "recall": 0.5557343475716794
30
+ },
31
+ "eval_NAME": {
32
+ "f1": 0.7002083333333333,
33
+ "number": 8859,
34
+ "precision": 0.6500338458563002,
35
+ "recall": 0.7587763855965685
36
+ },
37
+ "eval_QTY": {
38
+ "f1": 0.9200849140587551,
39
+ "number": 7095,
40
+ "precision": 0.8947789025039957,
41
+ "recall": 0.9468639887244539
42
+ },
43
+ "eval_RANGE_END": {
44
+ "f1": 0.0,
45
+ "number": 74,
46
+ "precision": 0.0,
47
+ "recall": 0.0
48
+ },
49
+ "eval_UNIT": {
50
+ "f1": 0.912968864917872,
51
+ "number": 5723,
52
+ "precision": 0.8575376112987412,
53
+ "recall": 0.9760615062030403
54
+ },
55
+ "eval_loss": 1.1246980428695679,
56
+ "eval_overall_accuracy": 0.7220226951438505,
57
+ "eval_overall_f1": 0.7265753075808941,
58
+ "eval_overall_precision": 0.6665790549617429,
59
+ "eval_overall_recall": 0.7984398502815965,
60
+ "eval_runtime": 10.7443,
61
+ "eval_samples_per_second": 791.676,
62
+ "eval_steps_per_second": 24.757,
63
+ "step": 1000
64
+ },
65
+ {
66
+ "epoch": 0.3,
67
+ "learning_rate": 4.703029103147892e-05,
68
+ "loss": 1.1236,
69
+ "step": 1500
70
+ },
71
+ {
72
+ "epoch": 0.4,
73
+ "learning_rate": 4.6040388041971886e-05,
74
+ "loss": 1.1018,
75
+ "step": 2000
76
+ },
77
+ {
78
+ "epoch": 0.4,
79
+ "eval_COMMENT": {
80
+ "f1": 0.48012443829934326,
81
+ "number": 6836,
82
+ "precision": 0.3960273712222011,
83
+ "recall": 0.6095669982445875
84
+ },
85
+ "eval_NAME": {
86
+ "f1": 0.733016860369946,
87
+ "number": 8859,
88
+ "precision": 0.7094423320659062,
89
+ "recall": 0.7582119878090078
90
+ },
91
+ "eval_QTY": {
92
+ "f1": 0.9238356164383562,
93
+ "number": 7095,
94
+ "precision": 0.8986009327115256,
95
+ "recall": 0.9505285412262157
96
+ },
97
+ "eval_RANGE_END": {
98
+ "f1": 0.047619047619047616,
99
+ "number": 74,
100
+ "precision": 0.2,
101
+ "recall": 0.02702702702702703
102
+ },
103
+ "eval_UNIT": {
104
+ "f1": 0.9199999999999999,
105
+ "number": 5723,
106
+ "precision": 0.8794009877329935,
107
+ "recall": 0.9645290931329722
108
+ },
109
+ "eval_loss": 1.0676864385604858,
110
+ "eval_overall_accuracy": 0.7414702173996103,
111
+ "eval_overall_f1": 0.7423559781301609,
112
+ "eval_overall_precision": 0.6852761825824404,
113
+ "eval_overall_recall": 0.8098086542834155,
114
+ "eval_runtime": 8.0746,
115
+ "eval_samples_per_second": 1053.43,
116
+ "eval_steps_per_second": 32.943,
117
+ "step": 2000
118
+ },
119
+ {
120
+ "epoch": 0.49,
121
+ "learning_rate": 4.505048505246486e-05,
122
+ "loss": 1.0885,
123
+ "step": 2500
124
+ },
125
+ {
126
+ "epoch": 0.59,
127
+ "learning_rate": 4.406058206295783e-05,
128
+ "loss": 1.0676,
129
+ "step": 3000
130
+ },
131
+ {
132
+ "epoch": 0.59,
133
+ "eval_COMMENT": {
134
+ "f1": 0.4957528957528958,
135
+ "number": 6836,
136
+ "precision": 0.41734173417341736,
137
+ "recall": 0.6104447045055588
138
+ },
139
+ "eval_NAME": {
140
+ "f1": 0.7383297644539616,
141
+ "number": 8859,
142
+ "precision": 0.7021688219122288,
143
+ "recall": 0.7784174286036799
144
+ },
145
+ "eval_QTY": {
146
+ "f1": 0.9259863945578231,
147
+ "number": 7095,
148
+ "precision": 0.8949375410913872,
149
+ "recall": 0.9592670894996477
150
+ },
151
+ "eval_RANGE_END": {
152
+ "f1": 0.3047619047619048,
153
+ "number": 74,
154
+ "precision": 0.5161290322580645,
155
+ "recall": 0.21621621621621623
156
+ },
157
+ "eval_UNIT": {
158
+ "f1": 0.9193198866477745,
159
+ "number": 5723,
160
+ "precision": 0.8788844621513944,
161
+ "recall": 0.9636554254761489
162
+ },
163
+ "eval_loss": 1.0472389459609985,
164
+ "eval_overall_accuracy": 0.7540595269934666,
165
+ "eval_overall_f1": 0.7511794345133026,
166
+ "eval_overall_precision": 0.6939017520974771,
167
+ "eval_overall_recall": 0.8187637737433099,
168
+ "eval_runtime": 7.2751,
169
+ "eval_samples_per_second": 1169.193,
170
+ "eval_steps_per_second": 36.563,
171
+ "step": 3000
172
+ },
173
+ {
174
+ "epoch": 0.69,
175
+ "learning_rate": 4.3070679073450804e-05,
176
+ "loss": 1.0667,
177
+ "step": 3500
178
+ },
179
+ {
180
+ "epoch": 0.79,
181
+ "learning_rate": 4.208077608394378e-05,
182
+ "loss": 1.0613,
183
+ "step": 4000
184
+ },
185
+ {
186
+ "epoch": 0.79,
187
+ "eval_COMMENT": {
188
+ "f1": 0.5095226317091268,
189
+ "number": 6836,
190
+ "precision": 0.4413024850042845,
191
+ "recall": 0.6026916325336454
192
+ },
193
+ "eval_NAME": {
194
+ "f1": 0.7497392545424604,
195
+ "number": 8859,
196
+ "precision": 0.7297499465697799,
197
+ "recall": 0.7708544982503669
198
+ },
199
+ "eval_QTY": {
200
+ "f1": 0.9260893546607832,
201
+ "number": 7095,
202
+ "precision": 0.9064651100013497,
203
+ "recall": 0.9465821000704722
204
+ },
205
+ "eval_RANGE_END": {
206
+ "f1": 0.23214285714285715,
207
+ "number": 74,
208
+ "precision": 0.34210526315789475,
209
+ "recall": 0.17567567567567569
210
+ },
211
+ "eval_UNIT": {
212
+ "f1": 0.9216135881104034,
213
+ "number": 5723,
214
+ "precision": 0.8965631196298744,
215
+ "recall": 0.9481041411846933
216
+ },
217
+ "eval_loss": 1.045949101448059,
218
+ "eval_overall_accuracy": 0.7502196920490581,
219
+ "eval_overall_f1": 0.7602500822639027,
220
+ "eval_overall_precision": 0.717671543503246,
221
+ "eval_overall_recall": 0.8081995312554657,
222
+ "eval_runtime": 9.0484,
223
+ "eval_samples_per_second": 940.052,
224
+ "eval_steps_per_second": 29.397,
225
+ "step": 4000
226
+ },
227
+ {
228
+ "epoch": 0.89,
229
+ "learning_rate": 4.109087309443675e-05,
230
+ "loss": 1.0532,
231
+ "step": 4500
232
+ },
233
+ {
234
+ "epoch": 0.99,
235
+ "learning_rate": 4.010097010492972e-05,
236
+ "loss": 1.045,
237
+ "step": 5000
238
+ },
239
+ {
240
+ "epoch": 0.99,
241
+ "eval_COMMENT": {
242
+ "f1": 0.5115396388123661,
243
+ "number": 6836,
244
+ "precision": 0.43983577218654596,
245
+ "recall": 0.6111761263897015
246
+ },
247
+ "eval_NAME": {
248
+ "f1": 0.7500674909562118,
249
+ "number": 8859,
250
+ "precision": 0.7188987787207618,
251
+ "recall": 0.7840614064792866
252
+ },
253
+ "eval_QTY": {
254
+ "f1": 0.9248702917593155,
255
+ "number": 7095,
256
+ "precision": 0.886005680351149,
257
+ "recall": 0.9673009161381254
258
+ },
259
+ "eval_RANGE_END": {
260
+ "f1": 0.27868852459016397,
261
+ "number": 74,
262
+ "precision": 0.3541666666666667,
263
+ "recall": 0.22972972972972974
264
+ },
265
+ "eval_UNIT": {
266
+ "f1": 0.9236628580890875,
267
+ "number": 5723,
268
+ "precision": 0.8777340676632572,
269
+ "recall": 0.974663637952123
270
+ },
271
+ "eval_loss": 1.029239296913147,
272
+ "eval_overall_accuracy": 0.7610323615940091,
273
+ "eval_overall_f1": 0.7619755400100165,
274
+ "eval_overall_precision": 0.7079555688982287,
275
+ "eval_overall_recall": 0.8249204183719873,
276
+ "eval_runtime": 10.4805,
277
+ "eval_samples_per_second": 811.605,
278
+ "eval_steps_per_second": 25.381,
279
+ "step": 5000
280
+ },
281
+ {
282
+ "epoch": 1.09,
283
+ "learning_rate": 3.911106711542269e-05,
284
+ "loss": 1.0391,
285
+ "step": 5500
286
+ },
287
+ {
288
+ "epoch": 1.19,
289
+ "learning_rate": 3.812116412591566e-05,
290
+ "loss": 1.0334,
291
+ "step": 6000
292
+ },
293
+ {
294
+ "epoch": 1.19,
295
+ "eval_COMMENT": {
296
+ "f1": 0.5539883268482491,
297
+ "number": 6836,
298
+ "precision": 0.47399084477736164,
299
+ "recall": 0.6664716208308953
300
+ },
301
+ "eval_NAME": {
302
+ "f1": 0.7479798253701395,
303
+ "number": 8859,
304
+ "precision": 0.7198329853862213,
305
+ "recall": 0.7784174286036799
306
+ },
307
+ "eval_QTY": {
308
+ "f1": 0.92860166502046,
309
+ "number": 7095,
310
+ "precision": 0.9296510806611104,
311
+ "recall": 0.927554615926709
312
+ },
313
+ "eval_RANGE_END": {
314
+ "f1": 0.3283582089552239,
315
+ "number": 74,
316
+ "precision": 0.36666666666666664,
317
+ "recall": 0.2972972972972973
318
+ },
319
+ "eval_UNIT": {
320
+ "f1": 0.9286739258884106,
321
+ "number": 5723,
322
+ "precision": 0.8982691051600261,
323
+ "recall": 0.9612091560370435
324
+ },
325
+ "eval_loss": 1.034406065940857,
326
+ "eval_overall_accuracy": 0.759465861765942,
327
+ "eval_overall_f1": 0.7717964680056355,
328
+ "eval_overall_precision": 0.7258049607148359,
329
+ "eval_overall_recall": 0.8240109140518418,
330
+ "eval_runtime": 8.1039,
331
+ "eval_samples_per_second": 1049.615,
332
+ "eval_steps_per_second": 32.824,
333
+ "step": 6000
334
+ },
335
+ {
336
+ "epoch": 1.29,
337
+ "learning_rate": 3.7131261136408633e-05,
338
+ "loss": 1.026,
339
+ "step": 6500
340
+ },
341
+ {
342
+ "epoch": 1.39,
343
+ "learning_rate": 3.6141358146901606e-05,
344
+ "loss": 1.0187,
345
+ "step": 7000
346
+ },
347
+ {
348
+ "epoch": 1.39,
349
+ "eval_COMMENT": {
350
+ "f1": 0.5136793031529874,
351
+ "number": 6836,
352
+ "precision": 0.4423198816818086,
353
+ "recall": 0.6124926857811586
354
+ },
355
+ "eval_NAME": {
356
+ "f1": 0.751153062318996,
357
+ "number": 8859,
358
+ "precision": 0.7155410238070911,
359
+ "recall": 0.7904955412574782
360
+ },
361
+ "eval_QTY": {
362
+ "f1": 0.9245735607675907,
363
+ "number": 7095,
364
+ "precision": 0.8767850372804247,
365
+ "recall": 0.9778717406624383
366
+ },
367
+ "eval_RANGE_END": {
368
+ "f1": 0.23853211009174313,
369
+ "number": 74,
370
+ "precision": 0.37142857142857144,
371
+ "recall": 0.17567567567567569
372
+ },
373
+ "eval_UNIT": {
374
+ "f1": 0.927461139896373,
375
+ "number": 5723,
376
+ "precision": 0.8888354957552459,
377
+ "recall": 0.9695963655425476
378
+ },
379
+ "eval_loss": 1.021018385887146,
380
+ "eval_overall_accuracy": 0.7651014404156956,
381
+ "eval_overall_f1": 0.7638116425658138,
382
+ "eval_overall_precision": 0.7083482836981222,
383
+ "eval_overall_recall": 0.8286983593941302,
384
+ "eval_runtime": 7.3061,
385
+ "eval_samples_per_second": 1164.235,
386
+ "eval_steps_per_second": 36.408,
387
+ "step": 7000
388
+ },
389
+ {
390
+ "epoch": 1.48,
391
+ "learning_rate": 3.515145515739457e-05,
392
+ "loss": 1.0186,
393
+ "step": 7500
394
+ },
395
+ {
396
+ "epoch": 1.58,
397
+ "learning_rate": 3.416155216788755e-05,
398
+ "loss": 1.0319,
399
+ "step": 8000
400
+ },
401
+ {
402
+ "epoch": 1.58,
403
+ "eval_COMMENT": {
404
+ "f1": 0.5439773496645535,
405
+ "number": 6836,
406
+ "precision": 0.46955690149824675,
407
+ "recall": 0.6464306612053833
408
+ },
409
+ "eval_NAME": {
410
+ "f1": 0.7617639003012875,
411
+ "number": 8859,
412
+ "precision": 0.7399957428693061,
413
+ "recall": 0.7848515633818716
414
+ },
415
+ "eval_QTY": {
416
+ "f1": 0.9297781396295542,
417
+ "number": 7095,
418
+ "precision": 0.8963893249607535,
419
+ "recall": 0.9657505285412262
420
+ },
421
+ "eval_RANGE_END": {
422
+ "f1": 0.368,
423
+ "number": 74,
424
+ "precision": 0.45098039215686275,
425
+ "recall": 0.3108108108108108
426
+ },
427
+ "eval_UNIT": {
428
+ "f1": 0.9286800743620078,
429
+ "number": 5723,
430
+ "precision": 0.8991981672394044,
431
+ "recall": 0.9601607548488555
432
+ },
433
+ "eval_loss": 1.0136394500732422,
434
+ "eval_overall_accuracy": 0.7699919764642953,
435
+ "eval_overall_f1": 0.7758823529411765,
436
+ "eval_overall_precision": 0.727991905068531,
437
+ "eval_overall_recall": 0.8305173680344212,
438
+ "eval_runtime": 9.7068,
439
+ "eval_samples_per_second": 876.293,
440
+ "eval_steps_per_second": 27.403,
441
+ "step": 8000
442
+ },
443
+ {
444
+ "epoch": 1.68,
445
+ "learning_rate": 3.317164917838052e-05,
446
+ "loss": 1.0243,
447
+ "step": 8500
448
+ },
449
+ {
450
+ "epoch": 1.78,
451
+ "learning_rate": 3.218174618887349e-05,
452
+ "loss": 1.0154,
453
+ "step": 9000
454
+ },
455
+ {
456
+ "epoch": 1.78,
457
+ "eval_COMMENT": {
458
+ "f1": 0.5554884675763797,
459
+ "number": 6836,
460
+ "precision": 0.47295907875796833,
461
+ "recall": 0.6729081334113517
462
+ },
463
+ "eval_NAME": {
464
+ "f1": 0.7644151565074134,
465
+ "number": 8859,
466
+ "precision": 0.7443054218800128,
467
+ "recall": 0.7856417202844564
468
+ },
469
+ "eval_QTY": {
470
+ "f1": 0.9317274604267033,
471
+ "number": 7095,
472
+ "precision": 0.9104236718224613,
473
+ "recall": 0.9540521494009866
474
+ },
475
+ "eval_RANGE_END": {
476
+ "f1": 0.3795620437956204,
477
+ "number": 74,
478
+ "precision": 0.4126984126984127,
479
+ "recall": 0.35135135135135137
480
+ },
481
+ "eval_UNIT": {
482
+ "f1": 0.9292205085452273,
483
+ "number": 5723,
484
+ "precision": 0.8885522959183674,
485
+ "recall": 0.9737899702952997
486
+ },
487
+ "eval_loss": 1.0071077346801758,
488
+ "eval_overall_accuracy": 0.7732204944026286,
489
+ "eval_overall_f1": 0.7789823224924309,
490
+ "eval_overall_precision": 0.7284683532742716,
491
+ "eval_overall_recall": 0.8370238220170008,
492
+ "eval_runtime": 10.018,
493
+ "eval_samples_per_second": 849.074,
494
+ "eval_steps_per_second": 26.552,
495
+ "step": 9000
496
+ },
497
+ {
498
+ "epoch": 1.88,
499
+ "learning_rate": 3.119184319936647e-05,
500
+ "loss": 1.0185,
501
+ "step": 9500
502
+ },
503
+ {
504
+ "epoch": 1.98,
505
+ "learning_rate": 3.0201940209859435e-05,
506
+ "loss": 1.011,
507
+ "step": 10000
508
+ },
509
+ {
510
+ "epoch": 1.98,
511
+ "eval_COMMENT": {
512
+ "f1": 0.5534477566997892,
513
+ "number": 6836,
514
+ "precision": 0.4703654417033473,
515
+ "recall": 0.6721767115272089
516
+ },
517
+ "eval_NAME": {
518
+ "f1": 0.7638995503892971,
519
+ "number": 8859,
520
+ "precision": 0.742723104808615,
521
+ "recall": 0.7863189976295293
522
+ },
523
+ "eval_QTY": {
524
+ "f1": 0.9353837666367836,
525
+ "number": 7095,
526
+ "precision": 0.915743991358358,
527
+ "recall": 0.9558844256518675
528
+ },
529
+ "eval_RANGE_END": {
530
+ "f1": 0.3384615384615385,
531
+ "number": 74,
532
+ "precision": 0.39285714285714285,
533
+ "recall": 0.2972972972972973
534
+ },
535
+ "eval_UNIT": {
536
+ "f1": 0.9314305016044586,
537
+ "number": 5723,
538
+ "precision": 0.901291060630822,
539
+ "recall": 0.9636554254761489
540
+ },
541
+ "eval_loss": 1.0126872062683105,
542
+ "eval_overall_accuracy": 0.7711955068200054,
543
+ "eval_overall_f1": 0.7789157805466762,
544
+ "eval_overall_precision": 0.7296281585138562,
545
+ "eval_overall_recall": 0.8353447371182705,
546
+ "eval_runtime": 7.6383,
547
+ "eval_samples_per_second": 1113.597,
548
+ "eval_steps_per_second": 34.824,
549
+ "step": 10000
550
+ },
551
+ {
552
+ "epoch": 2.08,
553
+ "learning_rate": 2.9212037220352405e-05,
554
+ "loss": 1.0062,
555
+ "step": 10500
556
+ },
557
+ {
558
+ "epoch": 2.18,
559
+ "learning_rate": 2.8222134230845377e-05,
560
+ "loss": 0.9958,
561
+ "step": 11000
562
+ },
563
+ {
564
+ "epoch": 2.18,
565
+ "eval_COMMENT": {
566
+ "f1": 0.5524928818077179,
567
+ "number": 6836,
568
+ "precision": 0.4715127701375246,
569
+ "recall": 0.6670567583382094
570
+ },
571
+ "eval_NAME": {
572
+ "f1": 0.7657111356119073,
573
+ "number": 8859,
574
+ "precision": 0.7483029845921776,
575
+ "recall": 0.7839485269217744
576
+ },
577
+ "eval_QTY": {
578
+ "f1": 0.9343237989447376,
579
+ "number": 7095,
580
+ "precision": 0.9206457791763579,
581
+ "recall": 0.9484143763213531
582
+ },
583
+ "eval_RANGE_END": {
584
+ "f1": 0.36363636363636365,
585
+ "number": 74,
586
+ "precision": 0.35,
587
+ "recall": 0.3783783783783784
588
+ },
589
+ "eval_UNIT": {
590
+ "f1": 0.9317363263254879,
591
+ "number": 5723,
592
+ "precision": 0.8947876447876448,
593
+ "recall": 0.9718679014502883
594
+ },
595
+ "eval_loss": 1.0023564100265503,
596
+ "eval_overall_accuracy": 0.7762770794330035,
597
+ "eval_overall_f1": 0.7792751537354442,
598
+ "eval_overall_precision": 0.7317627545535522,
599
+ "eval_overall_recall": 0.8333858047364187,
600
+ "eval_runtime": 8.2445,
601
+ "eval_samples_per_second": 1031.712,
602
+ "eval_steps_per_second": 32.264,
603
+ "step": 11000
604
+ },
605
+ {
606
+ "epoch": 2.28,
607
+ "learning_rate": 2.7232231241338353e-05,
608
+ "loss": 0.9969,
609
+ "step": 11500
610
+ },
611
+ {
612
+ "epoch": 2.38,
613
+ "learning_rate": 2.6242328251831323e-05,
614
+ "loss": 1.0042,
615
+ "step": 12000
616
+ },
617
+ {
618
+ "epoch": 2.38,
619
+ "eval_COMMENT": {
620
+ "f1": 0.5542747950440732,
621
+ "number": 6836,
622
+ "precision": 0.4789602641951635,
623
+ "recall": 0.657694558221182
624
+ },
625
+ "eval_NAME": {
626
+ "f1": 0.7601839329185826,
627
+ "number": 8859,
628
+ "precision": 0.7298981923955953,
629
+ "recall": 0.7930917710802574
630
+ },
631
+ "eval_QTY": {
632
+ "f1": 0.9347491965907503,
633
+ "number": 7095,
634
+ "precision": 0.9267211525141986,
635
+ "recall": 0.9429175475687104
636
+ },
637
+ "eval_RANGE_END": {
638
+ "f1": 0.3902439024390244,
639
+ "number": 74,
640
+ "precision": 0.35555555555555557,
641
+ "recall": 0.43243243243243246
642
+ },
643
+ "eval_UNIT": {
644
+ "f1": 0.932516381584546,
645
+ "number": 5723,
646
+ "precision": 0.9089250165892502,
647
+ "recall": 0.9573650183470208
648
+ },
649
+ "eval_loss": 1.0006664991378784,
650
+ "eval_overall_accuracy": 0.7768883964390785,
651
+ "eval_overall_f1": 0.7786074142146806,
652
+ "eval_overall_precision": 0.733323029366306,
653
+ "eval_overall_recall": 0.8298527302620072,
654
+ "eval_runtime": 10.2957,
655
+ "eval_samples_per_second": 826.172,
656
+ "eval_steps_per_second": 25.836,
657
+ "step": 12000
658
+ },
659
+ {
660
+ "epoch": 2.47,
661
+ "learning_rate": 2.5252425262324292e-05,
662
+ "loss": 1.0,
663
+ "step": 12500
664
+ },
665
+ {
666
+ "epoch": 2.57,
667
+ "learning_rate": 2.4262522272817265e-05,
668
+ "loss": 1.0048,
669
+ "step": 13000
670
+ },
671
+ {
672
+ "epoch": 2.57,
673
+ "eval_COMMENT": {
674
+ "f1": 0.542151434082001,
675
+ "number": 6836,
676
+ "precision": 0.47168994262206343,
677
+ "recall": 0.6373610298420129
678
+ },
679
+ "eval_NAME": {
680
+ "f1": 0.7605572485851111,
681
+ "number": 8859,
682
+ "precision": 0.7342649994746243,
683
+ "recall": 0.7888023478947963
684
+ },
685
+ "eval_QTY": {
686
+ "f1": 0.9334144527817211,
687
+ "number": 7095,
688
+ "precision": 0.8968563263185243,
689
+ "recall": 0.9730796335447498
690
+ },
691
+ "eval_RANGE_END": {
692
+ "f1": 0.4028776978417266,
693
+ "number": 74,
694
+ "precision": 0.4307692307692308,
695
+ "recall": 0.3783783783783784
696
+ },
697
+ "eval_UNIT": {
698
+ "f1": 0.9308501753799899,
699
+ "number": 5723,
700
+ "precision": 0.8915373540233562,
701
+ "recall": 0.9737899702952997
702
+ },
703
+ "eval_loss": 0.9942804574966431,
704
+ "eval_overall_accuracy": 0.7787414511137432,
705
+ "eval_overall_f1": 0.777442751202021,
706
+ "eval_overall_precision": 0.72784423828125,
707
+ "eval_overall_recall": 0.8342953090565641,
708
+ "eval_runtime": 9.9772,
709
+ "eval_samples_per_second": 852.54,
710
+ "eval_steps_per_second": 26.661,
711
+ "step": 13000
712
+ },
713
+ {
714
+ "epoch": 2.67,
715
+ "learning_rate": 2.3272619283310237e-05,
716
+ "loss": 0.9991,
717
+ "step": 13500
718
+ },
719
+ {
720
+ "epoch": 2.77,
721
+ "learning_rate": 2.2282716293803206e-05,
722
+ "loss": 0.9911,
723
+ "step": 14000
724
+ },
725
+ {
726
+ "epoch": 2.77,
727
+ "eval_COMMENT": {
728
+ "f1": 0.5497025285076846,
729
+ "number": 6836,
730
+ "precision": 0.4768817204301075,
731
+ "recall": 0.6487712112346401
732
+ },
733
+ "eval_NAME": {
734
+ "f1": 0.7610256410256411,
735
+ "number": 8859,
736
+ "precision": 0.729257190151045,
737
+ "recall": 0.7956880009030365
738
+ },
739
+ "eval_QTY": {
740
+ "f1": 0.9355014287658184,
741
+ "number": 7095,
742
+ "precision": 0.9042483230303827,
743
+ "recall": 0.9689922480620154
744
+ },
745
+ "eval_RANGE_END": {
746
+ "f1": 0.4132231404958678,
747
+ "number": 74,
748
+ "precision": 0.5319148936170213,
749
+ "recall": 0.33783783783783783
750
+ },
751
+ "eval_UNIT": {
752
+ "f1": 0.9322147651006711,
753
+ "number": 5723,
754
+ "precision": 0.8965628529933839,
755
+ "recall": 0.9708195002621003
756
+ },
757
+ "eval_loss": 0.9951051473617554,
758
+ "eval_overall_accuracy": 0.7786077255186643,
759
+ "eval_overall_f1": 0.7798045602605863,
760
+ "eval_overall_precision": 0.7295888824551245,
761
+ "eval_overall_recall": 0.8374435932416833,
762
+ "eval_runtime": 11.0903,
763
+ "eval_samples_per_second": 766.978,
764
+ "eval_steps_per_second": 23.985,
765
+ "step": 14000
766
+ },
767
+ {
768
+ "epoch": 2.87,
769
+ "learning_rate": 2.129281330429618e-05,
770
+ "loss": 0.9912,
771
+ "step": 14500
772
+ },
773
+ {
774
+ "epoch": 2.97,
775
+ "learning_rate": 2.0302910314789152e-05,
776
+ "loss": 0.9991,
777
+ "step": 15000
778
+ },
779
+ {
780
+ "epoch": 2.97,
781
+ "eval_COMMENT": {
782
+ "f1": 0.5583836904107916,
783
+ "number": 6836,
784
+ "precision": 0.4791033832617576,
785
+ "recall": 0.6691047396138092
786
+ },
787
+ "eval_NAME": {
788
+ "f1": 0.7690366469168313,
789
+ "number": 8859,
790
+ "precision": 0.7481054541573273,
791
+ "recall": 0.7911728186025511
792
+ },
793
+ "eval_QTY": {
794
+ "f1": 0.935168616655196,
795
+ "number": 7095,
796
+ "precision": 0.9137861466039005,
797
+ "recall": 0.9575757575757575
798
+ },
799
+ "eval_RANGE_END": {
800
+ "f1": 0.4109589041095891,
801
+ "number": 74,
802
+ "precision": 0.4166666666666667,
803
+ "recall": 0.40540540540540543
804
+ },
805
+ "eval_UNIT": {
806
+ "f1": 0.9313129631171426,
807
+ "number": 5723,
808
+ "precision": 0.8894720101781171,
809
+ "recall": 0.977284640922593
810
+ },
811
+ "eval_loss": 0.9920729398727417,
812
+ "eval_overall_accuracy": 0.7806709204141673,
813
+ "eval_overall_f1": 0.783059806192698,
814
+ "eval_overall_precision": 0.7336981443551099,
815
+ "eval_overall_recall": 0.839542449365096,
816
+ "eval_runtime": 9.9635,
817
+ "eval_samples_per_second": 853.712,
818
+ "eval_steps_per_second": 26.697,
819
+ "step": 15000
820
+ },
821
+ {
822
+ "epoch": 3.07,
823
+ "learning_rate": 1.9313007325282124e-05,
824
+ "loss": 0.9903,
825
+ "step": 15500
826
+ },
827
+ {
828
+ "epoch": 3.17,
829
+ "learning_rate": 1.8323104335775097e-05,
830
+ "loss": 0.9805,
831
+ "step": 16000
832
+ },
833
+ {
834
+ "epoch": 3.17,
835
+ "eval_COMMENT": {
836
+ "f1": 0.5583219220714553,
837
+ "number": 6836,
838
+ "precision": 0.4859154929577465,
839
+ "recall": 0.6560854300760679
840
+ },
841
+ "eval_NAME": {
842
+ "f1": 0.7669049828084921,
843
+ "number": 8859,
844
+ "precision": 0.7423922231614539,
845
+ "recall": 0.7930917710802574
846
+ },
847
+ "eval_QTY": {
848
+ "f1": 0.9368610113290964,
849
+ "number": 7095,
850
+ "precision": 0.9187102018696653,
851
+ "recall": 0.9557434813248766
852
+ },
853
+ "eval_RANGE_END": {
854
+ "f1": 0.4087591240875913,
855
+ "number": 74,
856
+ "precision": 0.4444444444444444,
857
+ "recall": 0.3783783783783784
858
+ },
859
+ "eval_UNIT": {
860
+ "f1": 0.9333669863705198,
861
+ "number": 5723,
862
+ "precision": 0.9000486775920817,
863
+ "recall": 0.9692468984798183
864
+ },
865
+ "eval_loss": 0.987960696220398,
866
+ "eval_overall_accuracy": 0.7821992129293547,
867
+ "eval_overall_f1": 0.783963999474445,
868
+ "eval_overall_precision": 0.7388935327079657,
869
+ "eval_overall_recall": 0.8348899849581978,
870
+ "eval_runtime": 9.4297,
871
+ "eval_samples_per_second": 902.044,
872
+ "eval_steps_per_second": 28.209,
873
+ "step": 16000
874
+ },
875
+ {
876
+ "epoch": 3.27,
877
+ "learning_rate": 1.7333201346268066e-05,
878
+ "loss": 0.9868,
879
+ "step": 16500
880
+ },
881
+ {
882
+ "epoch": 3.37,
883
+ "learning_rate": 1.634329835676104e-05,
884
+ "loss": 0.9848,
885
+ "step": 17000
886
+ },
887
+ {
888
+ "epoch": 3.37,
889
+ "eval_COMMENT": {
890
+ "f1": 0.5623249735515589,
891
+ "number": 6836,
892
+ "precision": 0.48933174482833314,
893
+ "recall": 0.6609128145114102
894
+ },
895
+ "eval_NAME": {
896
+ "f1": 0.7673142355394577,
897
+ "number": 8859,
898
+ "precision": 0.7466623945316672,
899
+ "recall": 0.7891409865673327
900
+ },
901
+ "eval_QTY": {
902
+ "f1": 0.936261013215859,
903
+ "number": 7095,
904
+ "precision": 0.9149737656397148,
905
+ "recall": 0.9585623678646934
906
+ },
907
+ "eval_RANGE_END": {
908
+ "f1": 0.3795620437956204,
909
+ "number": 74,
910
+ "precision": 0.4126984126984127,
911
+ "recall": 0.35135135135135137
912
+ },
913
+ "eval_UNIT": {
914
+ "f1": 0.9321663019693655,
915
+ "number": 5723,
916
+ "precision": 0.899171943497321,
917
+ "recall": 0.9676742966975362
918
+ },
919
+ "eval_loss": 0.9842170476913452,
920
+ "eval_overall_accuracy": 0.7823902494937531,
921
+ "eval_overall_f1": 0.7848384233538249,
922
+ "eval_overall_precision": 0.7402561160894235,
923
+ "eval_overall_recall": 0.8351348515059293,
924
+ "eval_runtime": 7.8716,
925
+ "eval_samples_per_second": 1080.599,
926
+ "eval_steps_per_second": 33.793,
927
+ "step": 17000
928
+ },
929
+ {
930
+ "epoch": 3.46,
931
+ "learning_rate": 1.535339536725401e-05,
932
+ "loss": 0.9841,
933
+ "step": 17500
934
+ },
935
+ {
936
+ "epoch": 3.56,
937
+ "learning_rate": 1.4363492377746981e-05,
938
+ "loss": 0.9771,
939
+ "step": 18000
940
+ },
941
+ {
942
+ "epoch": 3.56,
943
+ "eval_COMMENT": {
944
+ "f1": 0.5630382256365777,
945
+ "number": 6836,
946
+ "precision": 0.4883396023643203,
947
+ "recall": 0.6647162083089526
948
+ },
949
+ "eval_NAME": {
950
+ "f1": 0.7652202248411449,
951
+ "number": 8859,
952
+ "precision": 0.7373874816830647,
953
+ "recall": 0.795236482672988
954
+ },
955
+ "eval_QTY": {
956
+ "f1": 0.9356418569359177,
957
+ "number": 7095,
958
+ "precision": 0.9162388543636855,
959
+ "recall": 0.9558844256518675
960
+ },
961
+ "eval_RANGE_END": {
962
+ "f1": 0.4520547945205479,
963
+ "number": 74,
964
+ "precision": 0.4583333333333333,
965
+ "recall": 0.44594594594594594
966
+ },
967
+ "eval_UNIT": {
968
+ "f1": 0.9327950206072841,
969
+ "number": 5723,
970
+ "precision": 0.8992864093415505,
971
+ "recall": 0.968897431417089
972
+ },
973
+ "eval_loss": 0.983447253704071,
974
+ "eval_overall_accuracy": 0.7836510908187827,
975
+ "eval_overall_f1": 0.7841076515077104,
976
+ "eval_overall_precision": 0.7369149819994462,
977
+ "eval_overall_recall": 0.8377584216601952,
978
+ "eval_runtime": 9.9588,
979
+ "eval_samples_per_second": 854.123,
980
+ "eval_steps_per_second": 26.71,
981
+ "step": 18000
982
+ },
983
+ {
984
+ "epoch": 3.66,
985
+ "learning_rate": 1.3373589388239954e-05,
986
+ "loss": 0.9815,
987
+ "step": 18500
988
+ },
989
+ {
990
+ "epoch": 3.76,
991
+ "learning_rate": 1.2383686398732925e-05,
992
+ "loss": 0.9787,
993
+ "step": 19000
994
+ },
995
+ {
996
+ "epoch": 3.76,
997
+ "eval_COMMENT": {
998
+ "f1": 0.5682919349892671,
999
+ "number": 6836,
1000
+ "precision": 0.4892808110676946,
1001
+ "recall": 0.677735517846694
1002
+ },
1003
+ "eval_NAME": {
1004
+ "f1": 0.7695169319984682,
1005
+ "number": 8859,
1006
+ "precision": 0.7466029723991507,
1007
+ "recall": 0.7938819279828423
1008
+ },
1009
+ "eval_QTY": {
1010
+ "f1": 0.9372361774271676,
1011
+ "number": 7095,
1012
+ "precision": 0.9206090266449157,
1013
+ "recall": 0.9544749823819592
1014
+ },
1015
+ "eval_RANGE_END": {
1016
+ "f1": 0.4189189189189189,
1017
+ "number": 74,
1018
+ "precision": 0.4189189189189189,
1019
+ "recall": 0.4189189189189189
1020
+ },
1021
+ "eval_UNIT": {
1022
+ "f1": 0.9332317847169331,
1023
+ "number": 5723,
1024
+ "precision": 0.9048244174597965,
1025
+ "recall": 0.9634806919447843
1026
+ },
1027
+ "eval_loss": 0.9832035899162292,
1028
+ "eval_overall_accuracy": 0.7843961334199365,
1029
+ "eval_overall_f1": 0.786327868852459,
1030
+ "eval_overall_precision": 0.7399191682349675,
1031
+ "eval_overall_recall": 0.8389477734634624,
1032
+ "eval_runtime": 10.5172,
1033
+ "eval_samples_per_second": 808.772,
1034
+ "eval_steps_per_second": 25.292,
1035
+ "step": 19000
1036
+ },
1037
+ {
1038
+ "epoch": 3.86,
1039
+ "learning_rate": 1.1393783409225896e-05,
1040
+ "loss": 0.9744,
1041
+ "step": 19500
1042
+ },
1043
+ {
1044
+ "epoch": 3.96,
1045
+ "learning_rate": 1.0403880419718868e-05,
1046
+ "loss": 0.9746,
1047
+ "step": 20000
1048
+ },
1049
+ {
1050
+ "epoch": 3.96,
1051
+ "eval_COMMENT": {
1052
+ "f1": 0.5697782746413266,
1053
+ "number": 6836,
1054
+ "precision": 0.4950890447922288,
1055
+ "recall": 0.6710064365125804
1056
+ },
1057
+ "eval_NAME": {
1058
+ "f1": 0.768101654069449,
1059
+ "number": 8859,
1060
+ "precision": 0.7460368124268539,
1061
+ "recall": 0.7915114572750874
1062
+ },
1063
+ "eval_QTY": {
1064
+ "f1": 0.9370073342929603,
1065
+ "number": 7095,
1066
+ "precision": 0.9120629837203096,
1067
+ "recall": 0.9633544749823819
1068
+ },
1069
+ "eval_RANGE_END": {
1070
+ "f1": 0.43312101910828027,
1071
+ "number": 74,
1072
+ "precision": 0.40963855421686746,
1073
+ "recall": 0.4594594594594595
1074
+ },
1075
+ "eval_UNIT": {
1076
+ "f1": 0.9337932194834694,
1077
+ "number": 5723,
1078
+ "precision": 0.9003893575600259,
1079
+ "recall": 0.9697710990739122
1080
+ },
1081
+ "eval_loss": 0.9827048778533936,
1082
+ "eval_overall_accuracy": 0.7846444809536546,
1083
+ "eval_overall_f1": 0.7875786988457503,
1084
+ "eval_overall_precision": 0.7411819163709304,
1085
+ "eval_overall_recall": 0.8401721062021198,
1086
+ "eval_runtime": 8.1048,
1087
+ "eval_samples_per_second": 1049.503,
1088
+ "eval_steps_per_second": 32.82,
1089
+ "step": 20000
1090
+ },
1091
+ {
1092
+ "epoch": 4.06,
1093
+ "learning_rate": 9.41397743021184e-06,
1094
+ "loss": 0.9818,
1095
+ "step": 20500
1096
+ },
1097
+ {
1098
+ "epoch": 4.16,
1099
+ "learning_rate": 8.424074440704812e-06,
1100
+ "loss": 0.976,
1101
+ "step": 21000
1102
+ },
1103
+ {
1104
+ "epoch": 4.16,
1105
+ "eval_COMMENT": {
1106
+ "f1": 0.5667608401916225,
1107
+ "number": 6836,
1108
+ "precision": 0.4884607241160279,
1109
+ "recall": 0.6749561146869514
1110
+ },
1111
+ "eval_NAME": {
1112
+ "f1": 0.7705115565779385,
1113
+ "number": 8859,
1114
+ "precision": 0.7483774869666986,
1115
+ "recall": 0.7939948075403545
1116
+ },
1117
+ "eval_QTY": {
1118
+ "f1": 0.936799394814662,
1119
+ "number": 7095,
1120
+ "precision": 0.9147193123824873,
1121
+ "recall": 0.9599718111346018
1122
+ },
1123
+ "eval_RANGE_END": {
1124
+ "f1": 0.42857142857142855,
1125
+ "number": 74,
1126
+ "precision": 0.4125,
1127
+ "recall": 0.44594594594594594
1128
+ },
1129
+ "eval_UNIT": {
1130
+ "f1": 0.9335247909451813,
1131
+ "number": 5723,
1132
+ "precision": 0.9035317200784827,
1133
+ "recall": 0.9655774943211602
1134
+ },
1135
+ "eval_loss": 0.9835863709449768,
1136
+ "eval_overall_accuracy": 0.7854086272112483,
1137
+ "eval_overall_f1": 0.7865212692798899,
1138
+ "eval_overall_precision": 0.7393111090590082,
1139
+ "eval_overall_recall": 0.8401721062021198,
1140
+ "eval_runtime": 10.4319,
1141
+ "eval_samples_per_second": 815.38,
1142
+ "eval_steps_per_second": 25.499,
1143
+ "step": 21000
1144
+ },
1145
+ {
1146
+ "epoch": 4.26,
1147
+ "learning_rate": 7.434171451197783e-06,
1148
+ "loss": 0.9902,
1149
+ "step": 21500
1150
+ },
1151
+ {
1152
+ "epoch": 4.36,
1153
+ "learning_rate": 6.444268461690754e-06,
1154
+ "loss": 0.9635,
1155
+ "step": 22000
1156
+ },
1157
+ {
1158
+ "epoch": 4.36,
1159
+ "eval_COMMENT": {
1160
+ "f1": 0.5716760502381983,
1161
+ "number": 6836,
1162
+ "precision": 0.49533612093920876,
1163
+ "recall": 0.6758338209479228
1164
+ },
1165
+ "eval_NAME": {
1166
+ "f1": 0.7705016685814322,
1167
+ "number": 8859,
1168
+ "precision": 0.7475583864118895,
1169
+ "recall": 0.7948978440004515
1170
+ },
1171
+ "eval_QTY": {
1172
+ "f1": 0.9371633752244165,
1173
+ "number": 7095,
1174
+ "precision": 0.9186408555570597,
1175
+ "recall": 0.9564482029598309
1176
+ },
1177
+ "eval_RANGE_END": {
1178
+ "f1": 0.43373493975903615,
1179
+ "number": 74,
1180
+ "precision": 0.391304347826087,
1181
+ "recall": 0.4864864864864865
1182
+ },
1183
+ "eval_UNIT": {
1184
+ "f1": 0.9337832138338253,
1185
+ "number": 5723,
1186
+ "precision": 0.9026418786692759,
1187
+ "recall": 0.9671500961034423
1188
+ },
1189
+ "eval_loss": 0.9831692576408386,
1190
+ "eval_overall_accuracy": 0.7850838650517709,
1191
+ "eval_overall_f1": 0.7882188858807121,
1192
+ "eval_overall_precision": 0.7422893874775944,
1193
+ "eval_overall_recall": 0.84020708713751,
1194
+ "eval_runtime": 9.9258,
1195
+ "eval_samples_per_second": 856.957,
1196
+ "eval_steps_per_second": 26.799,
1197
+ "step": 22000
1198
+ },
1199
+ {
1200
+ "epoch": 4.45,
1201
+ "learning_rate": 5.4543654721837265e-06,
1202
+ "loss": 0.967,
1203
+ "step": 22500
1204
+ },
1205
+ {
1206
+ "epoch": 4.55,
1207
+ "learning_rate": 4.4644624826766974e-06,
1208
+ "loss": 0.9688,
1209
+ "step": 23000
1210
+ },
1211
+ {
1212
+ "epoch": 4.55,
1213
+ "eval_COMMENT": {
1214
+ "f1": 0.5723930522310194,
1215
+ "number": 6836,
1216
+ "precision": 0.4930739135032251,
1217
+ "recall": 0.6821240491515506
1218
+ },
1219
+ "eval_NAME": {
1220
+ "f1": 0.7698282463625424,
1221
+ "number": 8859,
1222
+ "precision": 0.7467897697124058,
1223
+ "recall": 0.7943334462128908
1224
+ },
1225
+ "eval_QTY": {
1226
+ "f1": 0.9373659076752716,
1227
+ "number": 7095,
1228
+ "precision": 0.9208593962469405,
1229
+ "recall": 0.9544749823819592
1230
+ },
1231
+ "eval_RANGE_END": {
1232
+ "f1": 0.4301075268817204,
1233
+ "number": 74,
1234
+ "precision": 0.35714285714285715,
1235
+ "recall": 0.5405405405405406
1236
+ },
1237
+ "eval_UNIT": {
1238
+ "f1": 0.9337379868487607,
1239
+ "number": 5723,
1240
+ "precision": 0.9021013194331324,
1241
+ "recall": 0.9676742966975362
1242
+ },
1243
+ "eval_loss": 0.9836147427558899,
1244
+ "eval_overall_accuracy": 0.7860008405608834,
1245
+ "eval_overall_f1": 0.7875949698716269,
1246
+ "eval_overall_precision": 0.7403416961674619,
1247
+ "eval_overall_recall": 0.8412914961346066,
1248
+ "eval_runtime": 8.0875,
1249
+ "eval_samples_per_second": 1051.747,
1250
+ "eval_steps_per_second": 32.89,
1251
+ "step": 23000
1252
+ },
1253
+ {
1254
+ "epoch": 4.65,
1255
+ "learning_rate": 3.4745594931696697e-06,
1256
+ "loss": 0.9686,
1257
+ "step": 23500
1258
+ },
1259
+ {
1260
+ "epoch": 4.75,
1261
+ "learning_rate": 2.484656503662641e-06,
1262
+ "loss": 0.9669,
1263
+ "step": 24000
1264
+ },
1265
+ {
1266
+ "epoch": 4.75,
1267
+ "eval_COMMENT": {
1268
+ "f1": 0.5733910891089109,
1269
+ "number": 6836,
1270
+ "precision": 0.4968897468897469,
1271
+ "recall": 0.677735517846694
1272
+ },
1273
+ "eval_NAME": {
1274
+ "f1": 0.7695764151460354,
1275
+ "number": 8859,
1276
+ "precision": 0.7478168264110756,
1277
+ "recall": 0.7926402528502088
1278
+ },
1279
+ "eval_QTY": {
1280
+ "f1": 0.9377799214497348,
1281
+ "number": 7095,
1282
+ "precision": 0.9173631706659477,
1283
+ "recall": 0.9591261451726568
1284
+ },
1285
+ "eval_RANGE_END": {
1286
+ "f1": 0.43902439024390244,
1287
+ "number": 74,
1288
+ "precision": 0.4,
1289
+ "recall": 0.4864864864864865
1290
+ },
1291
+ "eval_UNIT": {
1292
+ "f1": 0.9331987211845869,
1293
+ "number": 5723,
1294
+ "precision": 0.8998864189518092,
1295
+ "recall": 0.9690721649484536
1296
+ },
1297
+ "eval_loss": 0.9803335070610046,
1298
+ "eval_overall_accuracy": 0.7873189928552325,
1299
+ "eval_overall_f1": 0.7886242865577643,
1300
+ "eval_overall_precision": 0.7423807318202872,
1301
+ "eval_overall_recall": 0.8410116486514849,
1302
+ "eval_runtime": 11.2838,
1303
+ "eval_samples_per_second": 753.825,
1304
+ "eval_steps_per_second": 23.574,
1305
+ "step": 24000
1306
+ },
1307
+ {
1308
+ "epoch": 4.85,
1309
+ "learning_rate": 1.4947535141556129e-06,
1310
+ "loss": 0.965,
1311
+ "step": 24500
1312
+ },
1313
+ {
1314
+ "epoch": 4.95,
1315
+ "learning_rate": 5.048505246485845e-07,
1316
+ "loss": 0.9691,
1317
+ "step": 25000
1318
+ },
1319
+ {
1320
+ "epoch": 4.95,
1321
+ "eval_COMMENT": {
1322
+ "f1": 0.5725781491798204,
1323
+ "number": 6836,
1324
+ "precision": 0.4962978860392746,
1325
+ "recall": 0.6765652428320655
1326
+ },
1327
+ "eval_NAME": {
1328
+ "f1": 0.7694330320460149,
1329
+ "number": 8859,
1330
+ "precision": 0.7474457215836526,
1331
+ "recall": 0.7927531324077209
1332
+ },
1333
+ "eval_QTY": {
1334
+ "f1": 0.9377026560883062,
1335
+ "number": 7095,
1336
+ "precision": 0.9183783783783783,
1337
+ "recall": 0.9578576462297392
1338
+ },
1339
+ "eval_RANGE_END": {
1340
+ "f1": 0.4484848484848485,
1341
+ "number": 74,
1342
+ "precision": 0.4065934065934066,
1343
+ "recall": 0.5
1344
+ },
1345
+ "eval_UNIT": {
1346
+ "f1": 0.933075174678003,
1347
+ "number": 5723,
1348
+ "precision": 0.9002599090318388,
1349
+ "recall": 0.968373230822995
1350
+ },
1351
+ "eval_loss": 0.9796159267425537,
1352
+ "eval_overall_accuracy": 0.7870897489779544,
1353
+ "eval_overall_f1": 0.7882984134276199,
1354
+ "eval_overall_precision": 0.7423212409616217,
1355
+ "eval_overall_recall": 0.8403470108790709,
1356
+ "eval_runtime": 8.9053,
1357
+ "eval_samples_per_second": 955.162,
1358
+ "eval_steps_per_second": 29.87,
1359
+ "step": 25000
1360
+ },
1361
+ {
1362
+ "epoch": 5.0,
1363
+ "step": 25255,
1364
+ "total_flos": 1840716078790068.0,
1365
+ "train_loss": 1.016139645029875,
1366
+ "train_runtime": 1364.1405,
1367
+ "train_samples_per_second": 592.325,
1368
+ "train_steps_per_second": 18.513
1369
+ }
1370
+ ],
1371
+ "logging_steps": 500,
1372
+ "max_steps": 25255,
1373
+ "num_train_epochs": 5,
1374
+ "save_steps": 1000,
1375
+ "total_flos": 1840716078790068.0,
1376
+ "trial_name": null,
1377
+ "trial_params": null
1378
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81859d1ac0fb671d403bd6d35f3e6e21c0bc76f408fd774fe29eb961bde29f34
3
+ size 4600
validation_results.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_COMMENT": {
4
+ "f1": 0.5725781491798204,
5
+ "number": 6836,
6
+ "precision": 0.4962978860392746,
7
+ "recall": 0.6765652428320655
8
+ },
9
+ "eval_NAME": {
10
+ "f1": 0.7694330320460149,
11
+ "number": 8859,
12
+ "precision": 0.7474457215836526,
13
+ "recall": 0.7927531324077209
14
+ },
15
+ "eval_QTY": {
16
+ "f1": 0.9377026560883062,
17
+ "number": 7095,
18
+ "precision": 0.9183783783783783,
19
+ "recall": 0.9578576462297392
20
+ },
21
+ "eval_RANGE_END": {
22
+ "f1": 0.4484848484848485,
23
+ "number": 74,
24
+ "precision": 0.4065934065934066,
25
+ "recall": 0.5
26
+ },
27
+ "eval_UNIT": {
28
+ "f1": 0.933075174678003,
29
+ "number": 5723,
30
+ "precision": 0.9002599090318388,
31
+ "recall": 0.968373230822995
32
+ },
33
+ "eval_loss": 0.9796159267425537,
34
+ "eval_overall_accuracy": 0.7870897489779544,
35
+ "eval_overall_f1": 0.7882984134276199,
36
+ "eval_overall_precision": 0.7423212409616217,
37
+ "eval_overall_recall": 0.8403470108790709,
38
+ "eval_runtime": 8.7725,
39
+ "eval_samples_per_second": 969.621,
40
+ "eval_steps_per_second": 30.322
41
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff