codebyzeb's picture
Final model for experiment Farsi
7504750 verified
|
raw
history blame
25.7 kB
metadata
library_name: transformers
tags:
  - Farsi
  - generated_from_trainer
model-index:
  - name: childes-segmentation-100k-gpt2_lm-model
    results: []

childes-segmentation-100k-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: 2.5822
  • Model Preparation Time: 0.0008
  • Perplexity: 13.2257
  • Bpc: 3.7253
  • Spike Seg Type Fscore Entropy: 0.2147
  • Spike Seg Boundary Fscore Entropy: 0.3699
  • Absolute Seg Type Fscore Entropy: 0.2664
  • Absolute Seg Boundary Fscore Entropy: 0.3205
  • Spike Seg Type Fscore Increase in entropy: 0.2026
  • Spike Seg Boundary Fscore Increase in entropy: 0.3669
  • Absolute Seg Type Fscore Increase in entropy: 0.3012
  • Absolute Seg Boundary Fscore Increase in entropy: 0.3784
  • Spike Seg Type Fscore Loss: 0.2730
  • Spike Seg Boundary Fscore Loss: 0.4662
  • Absolute Seg Type Fscore Loss: 0.3005
  • Absolute Seg Boundary Fscore Loss: 0.4539
  • Spike Seg Type Fscore Increase in loss: 0.2485
  • Spike Seg Boundary Fscore Increase in loss: 0.4556
  • Absolute Seg Type Fscore Increase in loss: 0.3506
  • Absolute Seg Boundary Fscore Increase in loss: 0.4909
  • Spike Seg Type Fscore Rank: 0.3340
  • Spike Seg Boundary Fscore Rank: 0.5171
  • Absolute Seg Type Fscore Rank: 0.3037
  • Absolute Seg Boundary Fscore Rank: 0.4734
  • Spike Seg Type Fscore Increase in rank: 0.2880
  • Spike Seg Boundary Fscore Increase in rank: 0.4970
  • Absolute Seg Type Fscore Increase in rank: 0.3533
  • Absolute Seg Boundary Fscore Increase in rank: 0.5186
  • Spike Seg Type Fscore Boundary prediction: 0.2822
  • Spike Seg Boundary Fscore Boundary prediction: 0.4492
  • Absolute Seg Type Fscore Boundary prediction: 0.3900
  • Absolute Seg Boundary Fscore Boundary prediction: 0.4825
  • Spike Seg Type Fscore Increase in boundary prediction: 0.2559
  • Spike Seg Boundary Fscore Increase in boundary prediction: 0.4277
  • Absolute Seg Type Fscore Increase in boundary prediction: 0.3831
  • Absolute Seg Boundary Fscore Increase in boundary prediction: 0.5119
  • Spike Seg Type Fscore Majority vote cutoff: 0.3535
  • Spike Seg Type Fscore Majority vote spike: 0.2767
  • Absolute Seg Type Fscore Majority vote cutoff: 0.3948
  • Absolute Seg Type Fscore Majority vote spike: 0.3462
  • Spike Seg Boundary Fscore Majority vote cutoff: 0.5237
  • Spike Seg Boundary Fscore Majority vote spike: 0.4786
  • Absolute Seg Boundary Fscore Majority vote cutoff: 0.5578
  • Absolute Seg Boundary Fscore Majority vote spike: 0.5272

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: 30000
  • training_steps: 100000

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
1.6826 384.6154 10000 1.7302 0.0008 5.6416 2.4961 0.2332 0.4283 0.294 0.3923 0.1835 0.4010 0.3243 0.4165 0.2734 0.4959 0.3570 0.4712 0.2582 0.4915 0.3736 0.5280 0.3512 0.5176 0.3576 0.4885 0.2703 0.4800 0.4004 0.4945 0.2874 0.4678 0.3595 0.4466 0.2533 0.4417 0.3677 0.4670 0.3824 0.2551 0.3844 0.3634 0.5378 0.4915 0.5556 0.5485
1.2519 769.2308 20000 1.9310 0.0008 6.8966 2.7859 0.2336 0.4211 0.2940 0.3732 0.2122 0.4251 0.2895 0.4324 0.2639 0.4672 0.2822 0.4313 0.2561 0.4750 0.3049 0.4958 0.3398 0.5081 0.3398 0.4833 0.2744 0.4846 0.3640 0.4978 0.2805 0.4713 0.4 0.5025 0.2554 0.4591 0.3921 0.4812 0.3480 0.2516 0.3945 0.3533 0.5479 0.4874 0.5636 0.5456
1.103 1153.8462 30000 2.1048 0.0008 8.2053 3.0366 0.2435 0.4124 0.2877 0.3427 0.2166 0.4002 0.3004 0.4311 0.2608 0.4719 0.3249 0.4287 0.2513 0.4762 0.3092 0.4925 0.3202 0.5149 0.3899 0.4426 0.2950 0.4961 0.3649 0.5244 0.2719 0.4621 0.3801 0.5106 0.2507 0.4447 0.3962 0.5509 0.3538 0.2866 0.3761 0.3337 0.5365 0.4971 0.5496 0.5406
0.9883 1538.4615 40000 2.2380 0.0008 9.3742 3.2287 0.2224 0.3862 0.2940 0.3445 0.2167 0.3904 0.2743 0.4030 0.2780 0.4676 0.3298 0.4298 0.2465 0.4583 0.3443 0.4889 0.3291 0.5216 0.3575 0.4506 0.2717 0.4902 0.3488 0.5229 0.2822 0.4524 0.3686 0.4866 0.2621 0.4354 0.3662 0.4680 0.3406 0.2749 0.3548 0.3527 0.5290 0.4790 0.5251 0.5324
0.9244 1923.0769 50000 2.3482 0.0008 10.4671 3.3878 0.2264 0.3775 0.2530 0.3560 0.2108 0.3881 0.2659 0.4046 0.2613 0.4648 0.3373 0.4432 0.252 0.4637 0.3496 0.4915 0.3295 0.5205 0.3060 0.4438 0.2815 0.4893 0.3608 0.5148 0.2760 0.4474 0.3833 0.4605 0.2521 0.4380 0.3862 0.4467 0.3386 0.2571 0.3885 0.3671 0.5364 0.4747 0.5370 0.5416
0.8837 2307.6923 60000 2.4229 0.0008 11.2783 3.4955 0.2284 0.3627 0.2575 0.3255 0.2032 0.3790 0.2890 0.3915 0.2731 0.4666 0.2921 0.4476 0.2455 0.4535 0.3594 0.4932 0.3410 0.5251 0.3104 0.4766 0.2789 0.4848 0.3421 0.5058 0.2854 0.4603 0.3971 0.4923 0.2636 0.4444 0.4021 0.4650 0.3515 0.2529 0.3790 0.3591 0.5312 0.4750 0.5330 0.5336
0.8519 2692.3077 70000 2.4612 0.0008 11.7190 3.5508 0.2246 0.3689 0.2718 0.3261 0.2092 0.3721 0.2849 0.3755 0.2749 0.4653 0.3509 0.4594 0.2534 0.4558 0.3469 0.4849 0.3189 0.5143 0.3079 0.4711 0.2951 0.5 0.3635 0.5247 0.2971 0.4582 0.3996 0.4986 0.2662 0.4419 0.4049 0.5126 0.3244 0.2640 0.3818 0.3604 0.5321 0.4809 0.5681 0.5347
0.8278 3076.9231 80000 2.5177 0.0008 12.4001 3.6323 0.2254 0.3696 0.2586 0.3270 0.2085 0.3773 0.2848 0.3861 0.2833 0.4766 0.3310 0.4503 0.2429 0.4492 0.3580 0.4916 0.3314 0.5184 0.3237 0.4797 0.2823 0.4935 0.3527 0.5269 0.2782 0.4476 0.4016 0.5046 0.2565 0.4330 0.4065 0.4622 0.3471 0.2689 0.3861 0.3340 0.5368 0.4792 0.5540 0.5321
0.8061 3461.5385 90000 2.5572 0.0008 12.8992 3.6892 0.2118 0.3622 0.2697 0.3221 0.2004 0.3627 0.2740 0.3728 0.2806 0.4686 0.3246 0.4505 0.2568 0.4491 0.3599 0.4903 0.3190 0.5155 0.3007 0.4738 0.2964 0.4979 0.3523 0.5199 0.2830 0.4449 0.3893 0.4766 0.2533 0.4277 0.3793 0.4798 0.3538 0.2623 0.3920 0.3604 0.5315 0.4729 0.5392 0.5293
0.7891 3846.1538 100000 2.5822 0.0008 13.2257 3.7253 0.2147 0.3699 0.2664 0.3205 0.2026 0.3669 0.3012 0.3784 0.2730 0.4662 0.3005 0.4539 0.2485 0.4556 0.3506 0.4909 0.3340 0.5171 0.3037 0.4734 0.2880 0.4970 0.3533 0.5186 0.2822 0.4492 0.3900 0.4825 0.2559 0.4277 0.3831 0.5119 0.3535 0.2767 0.3948 0.3462 0.5237 0.4786 0.5578 0.5272

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.18.0
  • Tokenizers 0.19.1