childes-segmentation-800k-2-gpt2_lm-model

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.6290
  • Model Preparation Time: 0.001
  • Perplexity: 13.8603
  • Bpc: 3.7929
  • Spike Seg Type Fscore Entropy: 0.2980
  • Spike Seg Boundary Fscore Entropy: 0.6260
  • Absolute Seg Type Fscore Entropy: 0.2485
  • Absolute Seg Boundary Fscore Entropy: 0.5617
  • Spike Seg Type Fscore Increase in entropy: 0.2963
  • Spike Seg Boundary Fscore Increase in entropy: 0.6503
  • Absolute Seg Type Fscore Increase in entropy: 0.3176
  • Absolute Seg Boundary Fscore Increase in entropy: 0.6553
  • Spike Seg Type Fscore Loss: 0.2602
  • Spike Seg Boundary Fscore Loss: 0.5793
  • Absolute Seg Type Fscore Loss: 0.2205
  • Absolute Seg Boundary Fscore Loss: 0.5444
  • Spike Seg Type Fscore Increase in loss: 0.3181
  • Spike Seg Boundary Fscore Increase in loss: 0.6476
  • Absolute Seg Type Fscore Increase in loss: 0.3157
  • Absolute Seg Boundary Fscore Increase in loss: 0.6526
  • Spike Seg Type Fscore Rank: 0.3241
  • Spike Seg Boundary Fscore Rank: 0.6053
  • Absolute Seg Type Fscore Rank: 0.2705
  • Absolute Seg Boundary Fscore Rank: 0.5571
  • Spike Seg Type Fscore Increase in rank: 0.3497
  • Spike Seg Boundary Fscore Increase in rank: 0.6572
  • Absolute Seg Type Fscore Increase in rank: 0.3818
  • Absolute Seg Boundary Fscore Increase in rank: 0.6636
  • Spike Seg Type Fscore Boundary prediction: 0.4136
  • Spike Seg Boundary Fscore Boundary prediction: 0.7427
  • Absolute Seg Type Fscore Boundary prediction: 0.2589
  • Absolute Seg Boundary Fscore Boundary prediction: 0.6639
  • Spike Seg Type Fscore Increase in boundary prediction: 0.3999
  • Spike Seg Boundary Fscore Increase in boundary prediction: 0.7307
  • Absolute Seg Type Fscore Increase in boundary prediction: 0.2801
  • Absolute Seg Boundary Fscore Increase in boundary prediction: 0.6821
  • Spike Seg Type Fscore Majority vote cutoff: 0.3951
  • Spike Seg Type Fscore Majority vote spike: 0.3563
  • Absolute Seg Type Fscore Majority vote cutoff: 0.3641
  • Absolute Seg Type Fscore Majority vote spike: 0.4161
  • Spike Seg Boundary Fscore Majority vote cutoff: 0.7457
  • Spike Seg Boundary Fscore Majority vote spike: 0.7125
  • Absolute Seg Boundary Fscore Majority vote cutoff: 0.7296
  • Absolute Seg Boundary Fscore Majority vote spike: 0.7315

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: 0.001
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 60000
  • training_steps: 200000

Training results

Training Loss Epoch Step Validation Loss Model Preparation Time Perplexity Bpc Spike Seg Type Fscore Entropy Spike Seg Boundary Fscore Entropy Absolute Seg Type Fscore Entropy Absolute Seg Boundary Fscore Entropy Spike Seg Type Fscore Increase in entropy Spike Seg Boundary Fscore Increase in entropy Absolute Seg Type Fscore Increase in entropy Absolute Seg Boundary Fscore Increase in entropy Spike Seg Type Fscore Loss Spike Seg Boundary Fscore Loss Absolute Seg Type Fscore Loss Absolute Seg Boundary Fscore Loss Spike Seg Type Fscore Increase in loss Spike Seg Boundary Fscore Increase in loss Absolute Seg Type Fscore Increase in loss Absolute Seg Boundary Fscore Increase in loss Spike Seg Type Fscore Rank Spike Seg Boundary Fscore Rank Absolute Seg Type Fscore Rank Absolute Seg Boundary Fscore Rank Spike Seg Type Fscore Increase in rank Spike Seg Boundary Fscore Increase in rank Absolute Seg Type Fscore Increase in rank Absolute Seg Boundary Fscore Increase in rank Spike Seg Type Fscore Boundary prediction Spike Seg Boundary Fscore Boundary prediction Absolute Seg Type Fscore Boundary prediction Absolute Seg Boundary Fscore Boundary prediction Spike Seg Type Fscore Increase in boundary prediction Spike Seg Boundary Fscore Increase in boundary prediction Absolute Seg Type Fscore Increase in boundary prediction Absolute Seg Boundary Fscore Increase in boundary prediction Spike Seg Type Fscore Majority vote cutoff Spike Seg Type Fscore Majority vote spike Absolute Seg Type Fscore Majority vote cutoff Absolute Seg Type Fscore Majority vote spike Spike Seg Boundary Fscore Majority vote cutoff Spike Seg Boundary Fscore Majority vote spike Absolute Seg Boundary Fscore Majority vote cutoff Absolute Seg Boundary Fscore Majority vote spike
1.8651 454.5455 20000 2.0874 0.001 8.0640 3.0115 0.3263 0.6601 0.2752 0.5959 0.3176 0.6853 0.3379 0.6883 0.2959 0.6244 0.2416 0.5645 0.3535 0.6937 0.3706 0.6917 0.3132 0.6027 0.2718 0.5530 0.3514 0.6561 0.3839 0.6656 0.4018 0.7333 0.2765 0.6572 0.3782 0.7210 0.2325 0.6770 0.4298 0.3697 0.3931 0.4421 0.7635 0.7324 0.7376 0.7577
1.5625 909.0909 40000 2.2034 0.001 9.0558 3.1788 0.3292 0.6672 0.2759 0.5938 0.3178 0.6859 0.3462 0.6900 0.2804 0.6049 0.2290 0.5588 0.3411 0.6775 0.3549 0.6742 0.3306 0.6113 0.2786 0.5473 0.3527 0.6563 0.3794 0.6638 0.4332 0.7623 0.3306 0.6813 0.3938 0.7449 0.3307 0.6991 0.4348 0.3605 0.4386 0.4414 0.7644 0.7272 0.7465 0.7593
1.4479 1363.6364 60000 2.2910 0.001 9.8847 3.3052 0.3379 0.6655 0.2968 0.5959 0.3108 0.6807 0.3515 0.6793 0.2677 0.5957 0.2302 0.5620 0.3247 0.6691 0.3407 0.6509 0.3255 0.6078 0.2737 0.5456 0.3515 0.6599 0.3853 0.6667 0.4139 0.7547 0.2746 0.6873 0.3960 0.7465 0.2732 0.6972 0.4149 0.3549 0.3883 0.4265 0.7666 0.7283 0.7260 0.7569
1.3536 1818.1818 80000 2.3933 0.001 10.9498 3.4528 0.3237 0.6546 0.2728 0.5732 0.3087 0.6780 0.3305 0.6815 0.2751 0.5899 0.2244 0.5547 0.3312 0.6673 0.3326 0.6629 0.3280 0.6094 0.2747 0.5556 0.3529 0.6600 0.3798 0.6670 0.4242 0.7535 0.2807 0.6809 0.4072 0.7436 0.2394 0.6733 0.4069 0.3614 0.3655 0.4342 0.7606 0.7249 0.7368 0.7543
1.3014 2272.7273 100000 2.4635 0.001 11.7458 3.5541 0.3108 0.6408 0.2776 0.5602 0.2945 0.6651 0.3293 0.6602 0.2659 0.5809 0.2129 0.5521 0.3275 0.6579 0.3312 0.6590 0.3260 0.6072 0.2712 0.5478 0.3582 0.6605 0.3823 0.6654 0.4175 0.7463 0.2418 0.6713 0.3911 0.7370 0.2354 0.6848 0.4073 0.3650 0.3910 0.4225 0.7486 0.7195 0.7301 0.7445
1.2661 2727.2727 120000 2.5116 0.001 12.3251 3.6235 0.3171 0.6384 0.2771 0.5690 0.3050 0.6623 0.3258 0.6648 0.2671 0.5845 0.2255 0.5534 0.3274 0.6581 0.3285 0.6596 0.3270 0.6098 0.2622 0.5563 0.3522 0.6595 0.3852 0.6691 0.4171 0.7536 0.2788 0.6714 0.3992 0.7411 0.3206 0.6836 0.4117 0.3537 0.4124 0.4199 0.7548 0.7194 0.7365 0.7439
1.2387 3181.8182 140000 2.5537 0.001 12.8543 3.6842 0.3096 0.6350 0.2773 0.5630 0.2976 0.6584 0.3292 0.6500 0.2604 0.5838 0.2242 0.5480 0.3229 0.6534 0.3216 0.6523 0.3216 0.6063 0.2638 0.5582 0.3503 0.6558 0.3870 0.6656 0.4184 0.7477 0.3065 0.6704 0.3944 0.7349 0.3327 0.6848 0.4143 0.3538 0.4064 0.4294 0.7501 0.7155 0.7257 0.7391
1.218 3636.3636 160000 2.5832 0.001 13.2399 3.7268 0.3096 0.6375 0.2732 0.5688 0.2932 0.6570 0.3206 0.6599 0.2629 0.5810 0.2168 0.5485 0.3149 0.6505 0.3237 0.6538 0.3238 0.6074 0.2578 0.5455 0.3491 0.6577 0.3852 0.6656 0.4102 0.7414 0.2484 0.6490 0.3989 0.7304 0.3284 0.6682 0.4036 0.3518 0.3991 0.4169 0.7458 0.7165 0.7261 0.7354
1.1982 4090.9091 180000 2.6101 0.001 13.6006 3.7656 0.3080 0.6330 0.2704 0.5646 0.2936 0.6530 0.3202 0.6523 0.2598 0.5800 0.2208 0.5487 0.3212 0.6507 0.3008 0.6554 0.3275 0.6064 0.2645 0.5489 0.3539 0.6581 0.3863 0.6649 0.4104 0.7413 0.2051 0.6573 0.3963 0.7312 0.3284 0.6694 0.4041 0.3512 0.3855 0.4158 0.7443 0.7138 0.7224 0.7350
1.1838 4545.4545 200000 2.6290 0.001 13.8603 3.7929 0.2980 0.6260 0.2485 0.5617 0.2963 0.6503 0.3176 0.6553 0.2602 0.5793 0.2205 0.5444 0.3181 0.6476 0.3157 0.6526 0.3241 0.6053 0.2705 0.5571 0.3497 0.6572 0.3818 0.6636 0.4136 0.7427 0.2589 0.6639 0.3999 0.7307 0.2801 0.6821 0.3951 0.3563 0.3641 0.4161 0.7457 0.7125 0.7296 0.7315

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.18.0
  • Tokenizers 0.19.1
Downloads last month
8
Safetensors
Model size
832k params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support