Hugging Face's logo

galkowskim
/
longformer_base_4096_QA_SQUAD

Question Answering
Transformers
Safetensors
longformer
Generated from Trainer
Inference Endpoints
Model card Files Community
galkowskim commited on
Commit
ef7f77c
•
1 Parent(s): d03354e

Training complete

Browse files
Files changed (3) hide show
  1. README.md +53 -0
  2. test_metrics.json +4 -0
  3. train_losses.csv +111 -0
README.md ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: allenai/longformer-base-4096
4
+ tags:
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: longformer_base_4096_QA_SQUAD
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ # longformer_base_4096_QA_SQUAD
15
+
16
+ This model is a fine-tuned version of [allenai/longformer-base-4096](https://huggingface.co/allenai/longformer-base-4096) on an unknown dataset.
17
+
18
+ ## Model description
19
+
20
+ More information needed
21
+
22
+ ## Intended uses & limitations
23
+
24
+ More information needed
25
+
26
+ ## Training and evaluation data
27
+
28
+ More information needed
29
+
30
+ ## Training procedure
31
+
32
+ ### Training hyperparameters
33
+
34
+ The following hyperparameters were used during training:
35
+ - learning_rate: 2e-05
36
+ - train_batch_size: 8
37
+ - eval_batch_size: 8
38
+ - seed: 42
39
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
+ - lr_scheduler_type: linear
41
+ - num_epochs: 5
42
+ - mixed_precision_training: Native AMP
43
+
44
+ ### Training results
45
+
46
+
47
+
48
+ ### Framework versions
49
+
50
+ - Transformers 4.40.0
51
+ - Pytorch 2.2.1
52
+ - Datasets 2.19.0
53
+ - Tokenizers 0.19.1
test_metrics.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "exact_match": 84.68306527909176,
3
+ "f1": 91.79238618680928
4
+ }
train_losses.csv ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ loss,epoch
2
+ 2.2082,0.045662100456621
3
+ 1.2748,0.091324200913242
4
+ 1.1634,0.136986301369863
5
+ 1.0932,0.182648401826484
6
+ 1.0833,0.228310502283105
7
+ 1.0057,0.273972602739726
8
+ 0.981,0.319634703196347
9
+ 1.0403,0.365296803652968
10
+ 0.9965,0.410958904109589
11
+ 0.983,0.45662100456621
12
+ 0.9687,0.502283105022831
13
+ 1.0,0.547945205479452
14
+ 0.9663,0.593607305936073
15
+ 0.9629,0.639269406392694
16
+ 0.8853,0.684931506849315
17
+ 0.9325,0.730593607305936
18
+ 0.9136,0.776255707762557
19
+ 0.9216,0.821917808219178
20
+ 0.9099,0.867579908675799
21
+ 0.8797,0.91324200913242
22
+ 0.8632,0.958904109589041
23
+ 0.8619,1.004566210045662
24
+ 0.7005,1.0502283105022832
25
+ 0.6824,1.095890410958904
26
+ 0.7055,1.1415525114155252
27
+ 0.6921,1.187214611872146
28
+ 0.7235,1.2328767123287672
29
+ 0.6818,1.278538812785388
30
+ 0.7069,1.3242009132420092
31
+ 0.679,1.36986301369863
32
+ 0.6634,1.4155251141552512
33
+ 0.7179,1.461187214611872
34
+ 0.6772,1.5068493150684932
35
+ 0.7062,1.5525114155251143
36
+ 0.7144,1.5981735159817352
37
+ 0.6827,1.643835616438356
38
+ 0.6945,1.6894977168949772
39
+ 0.6747,1.7351598173515983
40
+ 0.6775,1.7808219178082192
41
+ 0.681,1.82648401826484
42
+ 0.7223,1.8721461187214612
43
+ 0.6732,1.9178082191780823
44
+ 0.7043,1.9634703196347032
45
+ 0.668,2.009132420091324
46
+ 0.5202,2.0547945205479454
47
+ 0.5198,2.1004566210045663
48
+ 0.5111,2.146118721461187
49
+ 0.4963,2.191780821917808
50
+ 0.5084,2.237442922374429
51
+ 0.5117,2.2831050228310503
52
+ 0.5256,2.328767123287671
53
+ 0.5166,2.374429223744292
54
+ 0.5223,2.4200913242009134
55
+ 0.5246,2.4657534246575343
56
+ 0.5214,2.5114155251141552
57
+ 0.5277,2.557077625570776
58
+ 0.4875,2.602739726027397
59
+ 0.517,2.6484018264840183
60
+ 0.5523,2.6940639269406392
61
+ 0.5067,2.73972602739726
62
+ 0.522,2.7853881278538815
63
+ 0.5136,2.8310502283105023
64
+ 0.517,2.8767123287671232
65
+ 0.4843,2.922374429223744
66
+ 0.5241,2.968036529680365
67
+ 0.4638,3.0136986301369864
68
+ 0.3786,3.0593607305936072
69
+ 0.3915,3.105022831050228
70
+ 0.384,3.1506849315068495
71
+ 0.3935,3.1963470319634704
72
+ 0.3693,3.2420091324200913
73
+ 0.3741,3.287671232876712
74
+ 0.3624,3.3333333333333335
75
+ 0.3793,3.3789954337899544
76
+ 0.3854,3.4246575342465753
77
+ 0.3777,3.470319634703196
78
+ 0.3866,3.5159817351598175
79
+ 0.4034,3.5616438356164384
80
+ 0.409,3.6073059360730593
81
+ 0.3762,3.65296803652968
82
+ 0.3732,3.6986301369863015
83
+ 0.3622,3.7442922374429224
84
+ 0.3641,3.7899543378995433
85
+ 0.3814,3.8356164383561646
86
+ 0.3862,3.8812785388127855
87
+ 0.3941,3.9269406392694064
88
+ 0.3603,3.9726027397260273
89
+ 0.3511,4.018264840182648
90
+ 0.2943,4.063926940639269
91
+ 0.2776,4.109589041095891
92
+ 0.2889,4.155251141552512
93
+ 0.2925,4.200913242009133
94
+ 0.29,4.2465753424657535
95
+ 0.2784,4.292237442922374
96
+ 0.2877,4.337899543378995
97
+ 0.2766,4.383561643835616
98
+ 0.3058,4.429223744292237
99
+ 0.2767,4.474885844748858
100
+ 0.2683,4.52054794520548
101
+ 0.2954,4.566210045662101
102
+ 0.2652,4.6118721461187215
103
+ 0.2898,4.657534246575342
104
+ 0.2987,4.703196347031963
105
+ 0.3012,4.748858447488584
106
+ 0.2832,4.794520547945205
107
+ 0.2981,4.840182648401827
108
+ 0.3067,4.885844748858448
109
+ 0.2899,4.931506849315069
110
+ 0.3128,4.9771689497716896
111
+ 0.5844105368139537,5.0