Hugging Face's logo

HuggingFaceTB
/
stack-edu-classifier-python

Safetensors
bert
Generated from Trainer
Model card Files
xet
Community
stack-edu-classifier-python / README.md
loubnabnl's picture
loubnabnl HF Staff
Model save
0480c11 verified
preview code
|
raw
history blame
16.4 kB
metadata
base_model: bigcode/starencoder
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: classifier-llama3-python-500k
    results: []

classifier-llama3-python-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: 0.4216
  • Precision: 0.5822
  • Recall: 0.4400
  • F1 Macro: 0.4652
  • Accuracy: 0.5749
  • F1 Binary Minimum3: 0.8018
  • F1 Binary Minimum2: 0.9438

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 7.4938 0.0265 0.2 0.0468 0.1326 0 0
0.4898 0.2880 1000 0.4759 0.5180 0.3927 0.4073 0.5474 0.7902 0.9395
0.4733 0.5760 2000 0.4696 0.5338 0.3984 0.4131 0.5493 0.7767 0.9400
0.4761 0.8641 3000 0.4571 0.5266 0.4080 0.4235 0.5573 0.7893 0.9410
0.4473 1.1521 4000 0.4589 0.5171 0.4061 0.4202 0.5522 0.7806 0.9389
0.4472 1.4401 5000 0.4686 0.5557 0.4135 0.4280 0.5503 0.7991 0.9416
0.4399 1.7281 6000 0.4508 0.5605 0.4184 0.4359 0.5599 0.7988 0.9421
0.4511 2.0161 7000 0.4467 0.5597 0.4183 0.4357 0.5615 0.7927 0.9402
0.445 2.3041 8000 0.4441 0.5395 0.4167 0.4326 0.5618 0.7913 0.9412
0.4554 2.5922 9000 0.4486 0.5640 0.4197 0.4372 0.5609 0.7999 0.9415
0.4499 2.8802 10000 0.4428 0.5702 0.4225 0.4405 0.5633 0.7990 0.9419
0.4492 3.1682 11000 0.4651 0.5714 0.4279 0.4459 0.5549 0.8015 0.9420
0.4459 3.4562 12000 0.4401 0.5690 0.4190 0.4368 0.5646 0.7986 0.9418
0.4469 3.7442 13000 0.4428 0.5753 0.4189 0.4378 0.5625 0.7976 0.9405
0.459 4.0323 14000 0.4385 0.5729 0.4229 0.4415 0.5661 0.7955 0.9421
0.4543 4.3203 15000 0.4418 0.5721 0.4220 0.4404 0.5629 0.8009 0.9408
0.442 4.6083 16000 0.4488 0.5803 0.4207 0.4388 0.5595 0.8028 0.9412
0.4525 4.8963 17000 0.4469 0.5712 0.4174 0.4338 0.5597 0.8005 0.9399
0.4539 5.1843 18000 0.4371 0.5867 0.4183 0.4372 0.5659 0.7996 0.9421
0.4527 5.4724 19000 0.4371 0.5707 0.4269 0.4450 0.5653 0.7920 0.9413
0.455 5.7604 20000 0.4364 0.5712 0.4288 0.4494 0.5664 0.7982 0.9416
0.4519 6.0484 21000 0.4371 0.5805 0.4274 0.4479 0.5667 0.8012 0.9421
0.4293 6.3364 22000 0.4351 0.5841 0.4214 0.4412 0.5669 0.7989 0.9424
0.4441 6.6244 23000 0.4360 0.5707 0.4272 0.4456 0.5667 0.7933 0.9413
0.4376 6.9124 24000 0.4360 0.5652 0.4262 0.4450 0.5652 0.7933 0.9412
0.4357 7.2005 25000 0.4382 0.5716 0.4244 0.4441 0.5647 0.8009 0.9411
0.4513 7.4885 26000 0.4382 0.5764 0.4245 0.4425 0.5629 0.7857 0.9410
0.422 7.7765 27000 0.4344 0.5736 0.4256 0.4456 0.5670 0.7967 0.9418
0.4317 8.0645 28000 0.4343 0.5799 0.4209 0.4406 0.5658 0.7995 0.9413
0.4458 8.3525 29000 0.4339 0.5793 0.4307 0.4521 0.5686 0.8015 0.9431
0.4591 8.6406 30000 0.4382 0.5869 0.4260 0.4470 0.5655 0.8031 0.9420
0.4313 8.9286 31000 0.4364 0.5717 0.4352 0.4577 0.5681 0.8022 0.9426
0.4201 9.2166 32000 0.4328 0.5686 0.4326 0.4540 0.5691 0.7958 0.9420
0.4433 9.5046 33000 0.4378 0.5778 0.4339 0.4554 0.5674 0.8036 0.9428
0.4404 9.7926 34000 0.4339 0.5855 0.4292 0.4516 0.5692 0.8021 0.9434
0.4324 10.0806 35000 0.4318 0.5695 0.4316 0.4533 0.5685 0.7985 0.9424
0.4393 10.3687 36000 0.4365 0.5804 0.4307 0.4529 0.5672 0.8040 0.9425
0.4334 10.6567 37000 0.4304 0.5780 0.4308 0.4525 0.5700 0.7996 0.9424
0.4396 10.9447 38000 0.4311 0.5691 0.4329 0.4547 0.5708 0.8001 0.9427
0.4398 11.2327 39000 0.4362 0.5732 0.4356 0.4579 0.5681 0.8040 0.9426
0.4568 11.5207 40000 0.4305 0.5814 0.4299 0.4516 0.5700 0.7998 0.9424
0.4459 11.8088 41000 0.4307 0.5793 0.4339 0.4562 0.5705 0.8017 0.9427
0.4326 12.0968 42000 0.4326 0.5821 0.4331 0.4559 0.5693 0.8026 0.9431
0.4343 12.3848 43000 0.4320 0.5751 0.4347 0.4578 0.5702 0.8026 0.9430
0.4247 12.6728 44000 0.4292 0.5768 0.4360 0.4592 0.5713 0.8011 0.9429
0.4285 12.9608 45000 0.4414 0.5789 0.4342 0.4566 0.5652 0.8054 0.9425
0.4304 13.2488 46000 0.4305 0.5767 0.4354 0.4584 0.5709 0.8028 0.9432
0.4211 13.5369 47000 0.4279 0.5759 0.4323 0.4542 0.5712 0.7996 0.9425
0.4451 13.8249 48000 0.4280 0.5906 0.4282 0.4507 0.5723 0.8002 0.9437
0.4298 14.1129 49000 0.4295 0.5799 0.4278 0.4488 0.5680 0.7921 0.9424
0.4328 14.4009 50000 0.4283 0.5804 0.4349 0.4587 0.5717 0.8014 0.9431
0.4366 14.6889 51000 0.4276 0.5777 0.4367 0.4606 0.5718 0.8007 0.9433
0.4225 14.9770 52000 0.4333 0.5690 0.4382 0.4614 0.5693 0.8039 0.9421
0.4411 15.2650 53000 0.4280 0.5738 0.4327 0.4559 0.5711 0.8013 0.9428
0.4279 15.5530 54000 0.4273 0.5787 0.4349 0.4589 0.5720 0.8016 0.9433
0.418 15.8410 55000 0.4283 0.5747 0.4328 0.4542 0.5694 0.7920 0.9423
0.4472 16.1290 56000 0.4276 0.5761 0.4350 0.4560 0.5712 0.7953 0.9426
0.426 16.4171 57000 0.4260 0.5910 0.4283 0.4514 0.5725 0.7983 0.9438
0.437 16.7051 58000 0.4298 0.5777 0.4354 0.4589 0.5708 0.8040 0.9430
0.4289 16.9931 59000 0.4272 0.5741 0.4382 0.4619 0.5725 0.8019 0.9435
0.4454 17.2811 60000 0.4254 0.5921 0.4311 0.4542 0.5734 0.8013 0.9433
0.4367 17.5691 61000 0.4273 0.5792 0.4377 0.4625 0.5726 0.8021 0.9432
0.4555 17.8571 62000 0.4259 0.5746 0.4379 0.4616 0.5725 0.7998 0.9430
0.4351 18.1452 63000 0.4257 0.5776 0.4334 0.4566 0.5719 0.7972 0.9431
0.4334 18.4332 64000 0.4247 0.5813 0.4378 0.4622 0.5739 0.7988 0.9437
0.423 18.7212 65000 0.4261 0.5783 0.4343 0.4573 0.5713 0.7934 0.9426
0.4433 19.0092 66000 0.4248 0.5756 0.4352 0.4591 0.5730 0.7996 0.9433
0.4355 19.2972 67000 0.4241 0.5822 0.4378 0.4623 0.5738 0.8012 0.9436
0.4268 19.5853 68000 0.4308 0.5814 0.4356 0.4589 0.5706 0.8044 0.9426
0.4291 19.8733 69000 0.4297 0.5802 0.4351 0.4587 0.5696 0.8044 0.9427
0.4291 20.1613 70000 0.4247 0.5799 0.4361 0.4598 0.5738 0.8020 0.9433
0.419 20.4493 71000 0.4241 0.5820 0.4363 0.4606 0.5739 0.8012 0.9434
0.4264 20.7373 72000 0.4282 0.5817 0.4369 0.4613 0.5714 0.8042 0.9430
0.4259 21.0253 73000 0.4239 0.5794 0.4353 0.4588 0.5729 0.7969 0.9431
0.422 21.3134 74000 0.4230 0.5843 0.4376 0.4622 0.5744 0.7990 0.9437
0.4312 21.6014 75000 0.4247 0.5835 0.4340 0.4585 0.5725 0.8012 0.9430
0.4103 21.8894 76000 0.4245 0.5804 0.4409 0.4664 0.5732 0.8026 0.9436
0.4473 22.1774 77000 0.4235 0.5831 0.4360 0.4603 0.5738 0.8008 0.9434
0.4205 22.4654 78000 0.4244 0.5807 0.4357 0.4600 0.5733 0.8021 0.9433
0.4294 22.7535 79000 0.4229 0.5862 0.4342 0.4584 0.5741 0.8011 0.9434
0.4467 23.0415 80000 0.4232 0.5749 0.4401 0.4649 0.5740 0.8009 0.9432
0.4296 23.3295 81000 0.4229 0.5812 0.4381 0.4629 0.5743 0.8006 0.9433
0.4308 23.6175 82000 0.4235 0.5758 0.4442 0.4698 0.5746 0.8022 0.9438
0.4251 23.9055 83000 0.4219 0.5862 0.4358 0.4602 0.5747 0.8003 0.9436
0.4383 24.1935 84000 0.4229 0.5784 0.4381 0.4626 0.5743 0.8015 0.9432
0.4309 24.4816 85000 0.4219 0.5837 0.4372 0.4618 0.5748 0.8001 0.9434
0.4245 24.7696 86000 0.4222 0.5810 0.4391 0.4633 0.5748 0.7982 0.9435
0.4227 25.0576 87000 0.4221 0.5836 0.4362 0.4606 0.5743 0.7991 0.9435
0.4224 25.3456 88000 0.4220 0.5775 0.4416 0.4664 0.5744 0.7998 0.9433
0.4247 25.6336 89000 0.4227 0.5854 0.4382 0.4632 0.5743 0.8028 0.9434
0.416 25.9217 90000 0.4230 0.5760 0.4427 0.4675 0.5746 0.8018 0.9434
0.4221 26.2097 91000 0.4215 0.5820 0.4390 0.4638 0.5752 0.8012 0.9439
0.4126 26.4977 92000 0.4263 0.5836 0.4391 0.4641 0.5723 0.8050 0.9431
0.424 26.7857 93000 0.4215 0.5828 0.4381 0.4627 0.5750 0.7986 0.9438
0.4272 27.0737 94000 0.4239 0.5853 0.4403 0.4660 0.5740 0.8040 0.9438
0.4306 27.3618 95000 0.4235 0.5801 0.4399 0.4651 0.5736 0.8029 0.9435
0.4164 27.6498 96000 0.4219 0.5801 0.4405 0.4656 0.5747 0.8015 0.9437
0.431 27.9378 97000 0.4213 0.5791 0.4393 0.4637 0.5748 0.7993 0.9434
0.4284 28.2258 98000 0.4227 0.5826 0.4402 0.4658 0.5744 0.8033 0.9435
0.4289 28.5138 99000 0.4216 0.5845 0.4384 0.4634 0.5749 0.8009 0.9438
0.4244 28.8018 100000 0.4221 0.5813 0.4377 0.4623 0.5746 0.8020 0.9436
0.4314 29.0899 101000 0.4216 0.5829 0.4402 0.4654 0.5751 0.8013 0.9437
0.4269 29.3779 102000 0.4212 0.5852 0.4405 0.4658 0.5754 0.8006 0.9438
0.4367 29.6659 103000 0.4215 0.5841 0.4400 0.4653 0.5749 0.8013 0.9437
0.4223 29.9539 104000 0.4216 0.5822 0.4400 0.4652 0.5749 0.8018 0.9438

Framework versions

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