Hugging Face's logo

HuggingFaceTB
/
stack-edu-classifier-csharp

Safetensors
bert
Generated from Trainer
Model card Files
xet
Community
stack-edu-classifier-csharp / README.md
loubnabnl's picture
loubnabnl HF Staff
Model save
0b1c48c verified
preview code
|
raw
history blame
15.9 kB
metadata
base_model: bigcode/starencoder
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: classifier-llama3-c-sharp-500k
    results: []

classifier-llama3-c-sharp-500k

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

  • Loss: 0.3976
  • Precision: 0.4844
  • Recall: 0.3567
  • F1 Macro: 0.3780
  • Accuracy: 0.5809
  • F1 Binary Minimum3: 0.6447
  • F1 Binary Minimum2: 0.9079

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: 0.0001
  • train_batch_size: 16
  • eval_batch_size: 256
  • seed: 0
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 128
  • total_eval_batch_size: 2048
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 200
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Macro Accuracy F1 Binary Minimum3 F1 Binary Minimum2
No log 0 0 5.5906 0.0374 0.2 0.0630 0.1870 0 0
0.4303 0.2991 1000 0.4344 0.4626 0.3128 0.3229 0.5565 0.5903 0.9038
0.4238 0.5983 2000 0.4268 0.4721 0.3366 0.3523 0.5667 0.6412 0.9052
0.4281 0.8974 3000 0.4229 0.4743 0.3374 0.3523 0.5685 0.6456 0.9051
0.4149 1.1965 4000 0.4171 0.4751 0.3347 0.3506 0.5700 0.6347 0.9056
0.4229 1.4957 5000 0.4192 0.4738 0.3493 0.3657 0.5699 0.6557 0.9058
0.4321 1.7948 6000 0.4126 0.4755 0.3452 0.3648 0.5736 0.6351 0.9063
0.408 2.0939 7000 0.4157 0.4677 0.3417 0.3617 0.5693 0.6088 0.9045
0.4117 2.3931 8000 0.4125 0.4783 0.3465 0.3635 0.5726 0.6492 0.9060
0.4172 2.6922 9000 0.4219 0.4819 0.3495 0.3611 0.5671 0.6665 0.9051
0.4061 2.9913 10000 0.4140 0.4737 0.3336 0.3515 0.5689 0.6022 0.9053
0.4192 3.2905 11000 0.4098 0.4805 0.3458 0.3634 0.5728 0.6487 0.9061
0.4218 3.5896 12000 0.4082 0.4836 0.3324 0.3484 0.5723 0.6237 0.9068
0.4007 3.8887 13000 0.4122 0.4829 0.3514 0.3671 0.5730 0.6594 0.9062
0.4122 4.1879 14000 0.4061 0.4789 0.3466 0.3666 0.5754 0.6369 0.9068
0.406 4.4870 15000 0.4070 0.4810 0.3521 0.3717 0.5753 0.6480 0.9068
0.4184 4.7861 16000 0.4060 0.4809 0.3460 0.3656 0.5752 0.6390 0.9069
0.4124 5.0853 17000 0.4057 0.4782 0.3491 0.3698 0.5763 0.6358 0.9065
0.4038 5.3844 18000 0.4130 0.4841 0.3473 0.3608 0.5697 0.6569 0.9054
0.4182 5.6835 19000 0.4048 0.4799 0.3431 0.3622 0.5748 0.6352 0.9066
0.4067 5.9827 20000 0.4047 0.4801 0.3520 0.3736 0.5766 0.6364 0.9073
0.4106 6.2818 21000 0.4096 0.4741 0.3349 0.3521 0.5708 0.6032 0.9056
0.4046 6.5809 22000 0.4043 0.4834 0.3408 0.3597 0.5753 0.6285 0.9069
0.3939 6.8800 23000 0.4075 0.4798 0.3597 0.3779 0.5763 0.6575 0.9072
0.4154 7.1792 24000 0.4057 0.4756 0.3465 0.3672 0.5754 0.6242 0.9063
0.4033 7.4783 25000 0.4054 0.4785 0.3449 0.3657 0.5749 0.6156 0.9067
0.4152 7.7774 26000 0.4033 0.4770 0.3500 0.3713 0.5765 0.6341 0.9067
0.4093 8.0766 27000 0.4046 0.4826 0.3528 0.3727 0.5769 0.6516 0.9075
0.404 8.3757 28000 0.4038 0.4835 0.3491 0.3683 0.5757 0.6448 0.9074
0.4173 8.6748 29000 0.4149 0.4835 0.3529 0.3642 0.5700 0.6695 0.9052
0.4199 8.9740 30000 0.4045 0.4829 0.3547 0.3735 0.5776 0.6551 0.9072
0.4053 9.2731 31000 0.4091 0.4681 0.3463 0.3670 0.5712 0.6069 0.9040
0.4072 9.5722 32000 0.4027 0.4801 0.3464 0.3671 0.5765 0.6280 0.9072
0.3984 9.8714 33000 0.4029 0.4786 0.3568 0.3772 0.5779 0.6508 0.9075
0.4075 10.1705 34000 0.4084 0.4716 0.3434 0.3639 0.5710 0.6008 0.9052
0.4016 10.4696 35000 0.4021 0.4817 0.3527 0.3732 0.5781 0.6445 0.9078
0.4077 10.7688 36000 0.4066 0.4759 0.3627 0.3824 0.5759 0.6582 0.9076
0.4039 11.0679 37000 0.4069 0.4707 0.3473 0.3683 0.5731 0.6108 0.9052
0.4107 11.3670 38000 0.4021 0.4807 0.3522 0.3741 0.5784 0.6346 0.9075
0.4208 11.6662 39000 0.4046 0.4872 0.3498 0.3674 0.5763 0.6531 0.9072
0.4028 11.9653 40000 0.4019 0.4788 0.3501 0.3716 0.5772 0.6292 0.9070
0.4084 12.2644 41000 0.4067 0.4809 0.3613 0.3789 0.5761 0.6635 0.9075
0.397 12.5636 42000 0.4023 0.4864 0.3506 0.3697 0.5775 0.6500 0.9077
0.4122 12.8627 43000 0.4012 0.4791 0.3516 0.3732 0.5781 0.6370 0.9066
0.3996 13.1618 44000 0.4046 0.4829 0.3565 0.3747 0.5766 0.6589 0.9075
0.4065 13.4610 45000 0.4015 0.4853 0.3487 0.3681 0.5782 0.6420 0.9073
0.4099 13.7601 46000 0.4044 0.4824 0.3576 0.3758 0.5773 0.6605 0.9074
0.3996 14.0592 47000 0.4007 0.4839 0.3476 0.3687 0.5782 0.6298 0.9074
0.4141 14.3584 48000 0.4022 0.4816 0.3584 0.3777 0.5781 0.6554 0.9074
0.4148 14.6575 49000 0.4021 0.4841 0.3401 0.3593 0.5760 0.6213 0.9071
0.399 14.9566 50000 0.4004 0.4815 0.3569 0.3784 0.5793 0.6447 0.9078
0.4095 15.2558 51000 0.4052 0.4750 0.3464 0.3675 0.5739 0.6091 0.9056
0.407 15.5549 52000 0.4006 0.4829 0.3557 0.3767 0.5792 0.6466 0.9078
0.3992 15.8540 53000 0.4014 0.4836 0.3535 0.3726 0.5781 0.6494 0.9070
0.4021 16.1532 54000 0.4037 0.4820 0.3580 0.3759 0.5770 0.6597 0.9073
0.4098 16.4523 55000 0.4034 0.4853 0.3519 0.3693 0.5757 0.6536 0.9066
0.4091 16.7514 56000 0.4000 0.4830 0.3523 0.3726 0.5782 0.6419 0.9075
0.3989 17.0506 57000 0.3997 0.4800 0.3546 0.3763 0.5786 0.6380 0.9076
0.3974 17.3497 58000 0.4038 0.4847 0.3574 0.3745 0.5770 0.6588 0.9070
0.4046 17.6488 59000 0.3997 0.4837 0.3484 0.3692 0.5785 0.6328 0.9074
0.4033 17.9480 60000 0.4028 0.4849 0.3571 0.3752 0.5780 0.6599 0.9076
0.3988 18.2471 61000 0.4002 0.4767 0.3544 0.3768 0.5784 0.6307 0.9069
0.4064 18.5462 62000 0.3995 0.4853 0.3551 0.3761 0.5799 0.6450 0.9079
0.4107 18.8453 63000 0.4002 0.4855 0.3546 0.3744 0.5791 0.6522 0.9078
0.4047 19.1445 64000 0.3992 0.4864 0.3526 0.3729 0.5798 0.6456 0.9079
0.405 19.4436 65000 0.3991 0.4860 0.3483 0.3679 0.5786 0.6416 0.9075
0.4002 19.7427 66000 0.4000 0.4853 0.3516 0.3709 0.5781 0.6473 0.9072
0.393 20.0419 67000 0.4005 0.4854 0.3569 0.3763 0.5799 0.6552 0.9078
0.3946 20.3410 68000 0.4027 0.4822 0.3587 0.3765 0.5770 0.6609 0.9074
0.4107 20.6401 69000 0.4041 0.4848 0.3575 0.3740 0.5759 0.6605 0.9072
0.4044 20.9393 70000 0.3985 0.4832 0.3554 0.3771 0.5801 0.6423 0.9078
0.3865 21.2384 71000 0.3987 0.4828 0.3595 0.3812 0.5807 0.6462 0.9083
0.3958 21.5375 72000 0.3985 0.4831 0.3559 0.3779 0.5799 0.6385 0.9078
0.4097 21.8367 73000 0.3992 0.4886 0.3533 0.3734 0.5796 0.6464 0.9074
0.4098 22.1358 74000 0.3986 0.4839 0.3558 0.3774 0.5803 0.6428 0.9079
0.4058 22.4349 75000 0.3988 0.4805 0.3553 0.3776 0.5794 0.6366 0.9075
0.389 22.7341 76000 0.3993 0.4852 0.3574 0.3775 0.5797 0.6530 0.9079
0.3903 23.0332 77000 0.4034 0.6830 0.3632 0.3816 0.5771 0.6624 0.9076
0.4029 23.3323 78000 0.3996 0.4812 0.3492 0.3707 0.5777 0.6267 0.9070
0.3989 23.6315 79000 0.3987 0.4817 0.3599 0.3809 0.5815 0.6518 0.9081
0.4032 23.9306 80000 0.3983 0.4849 0.3550 0.3760 0.5807 0.6451 0.9081
0.3981 24.2297 81000 0.3980 0.4820 0.3570 0.3783 0.5805 0.6444 0.9077
0.3913 24.5289 82000 0.3981 0.4825 0.3533 0.3750 0.5799 0.6365 0.9078
0.3964 24.8280 83000 0.3985 0.4883 0.3532 0.3733 0.5802 0.6448 0.9078
0.3942 25.1271 84000 0.3978 0.4843 0.3526 0.3740 0.5800 0.6394 0.9079
0.4057 25.4263 85000 0.3984 0.4870 0.3580 0.3787 0.5812 0.6500 0.9082
0.4076 25.7254 86000 0.4014 0.4862 0.3555 0.3726 0.5777 0.6590 0.9072
0.4003 26.0245 87000 0.3979 0.4804 0.3602 0.3820 0.5809 0.6442 0.9079
0.3979 26.3237 88000 0.3981 0.4845 0.3570 0.3778 0.5803 0.6472 0.9077
0.4201 26.6228 89000 0.3998 0.4827 0.3603 0.3799 0.5793 0.6561 0.9079
0.4014 26.9219 90000 0.3977 0.4844 0.3569 0.3782 0.5810 0.6457 0.9081
0.4031 27.2211 91000 0.3977 0.4838 0.3584 0.3802 0.5816 0.6442 0.9081
0.3843 27.5202 92000 0.3985 0.4876 0.3551 0.3751 0.5803 0.6500 0.9079
0.405 27.8193 93000 0.3978 0.4846 0.3566 0.3776 0.5809 0.6456 0.9079
0.394 28.1185 94000 0.3978 0.4828 0.3596 0.3811 0.5812 0.6483 0.9080
0.4047 28.4176 95000 0.3976 0.4856 0.3553 0.3768 0.5808 0.6426 0.9080
0.3874 28.7167 96000 0.3976 0.4844 0.3572 0.3788 0.5813 0.6447 0.9082
0.3974 29.0159 97000 0.3976 0.4852 0.3570 0.3786 0.5810 0.6441 0.9081
0.4096 29.3150 98000 0.3978 0.4855 0.3581 0.3791 0.5809 0.6473 0.9080
0.397 29.6141 99000 0.3976 0.4850 0.3586 0.3801 0.5815 0.6462 0.9081
0.4048 29.9133 100000 0.3976 0.4844 0.3567 0.3780 0.5809 0.6447 0.9079

Framework versions

  • Transformers 4.43.4
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1