Hugging Face's logo

HuggingFaceTB
/
stack-edu-classifier-markdown

Text Classification
Transformers
Safetensors
code
bert
Generated from Trainer
text-embeddings-inference
Model card Files
xet
Community
stack-edu-classifier-markdown / README.md
loubnabnl's picture
loubnabnl HF Staff
Model save
e83fe43 verified
preview code
|
raw
history blame
16 kB
metadata 
base_model: bigcode/starencoder
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: classifier-llama3-markdown-500k
    results: []

classifier-llama3-markdown-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: 0.3043
  • Precision: 0.6007
  • Recall: 0.5268
  • F1 Macro: 0.5255
  • Accuracy: 0.6694
  • F1 Binary Minimum3: 0.8552
  • F1 Binary Minimum2: 0.9316

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 9.3055 0.0541 0.2 0.0852 0.2707 0 0
0.3498 0.2946 1000 0.3644 0.5104 0.4920 0.4872 0.6121 0.8302 0.9179
0.3395 0.5891 2000 0.3504 0.5128 0.4955 0.4867 0.6306 0.8445 0.9174
0.3426 0.8837 3000 0.3387 0.5200 0.5029 0.5005 0.6349 0.8435 0.9229
0.3533 1.1782 4000 0.3362 0.5195 0.5056 0.5032 0.6367 0.8441 0.9225
0.3308 1.4728 5000 0.3306 0.5229 0.5065 0.5040 0.6456 0.8483 0.9242
0.3453 1.7673 6000 0.3297 0.5248 0.5034 0.5020 0.6443 0.8495 0.9239
0.3339 2.0619 7000 0.3258 0.5225 0.5106 0.5078 0.6494 0.8492 0.9247
0.3442 2.3564 8000 0.3339 0.5441 0.5139 0.5114 0.6418 0.8423 0.9256
0.323 2.6510 9000 0.3235 0.5249 0.5114 0.5090 0.6528 0.8503 0.9260
0.3077 2.9455 10000 0.3230 0.5268 0.5096 0.5088 0.6495 0.8501 0.9260
0.3215 3.2401 11000 0.3435 0.5266 0.5127 0.5092 0.6343 0.8378 0.9239
0.3183 3.5346 12000 0.3287 0.5397 0.5205 0.5174 0.6495 0.8453 0.9268
0.3205 3.8292 13000 0.3223 0.5430 0.5178 0.5155 0.6514 0.8481 0.9270
0.3325 4.1237 14000 0.3200 0.5601 0.5166 0.5154 0.6533 0.8490 0.9273
0.3057 4.4183 15000 0.3183 0.5282 0.5156 0.5141 0.6560 0.8507 0.9278
0.2949 4.7128 16000 0.3176 0.5551 0.5184 0.5166 0.6577 0.8511 0.9276
0.3045 5.0074 17000 0.3230 0.5263 0.5123 0.5086 0.6574 0.8517 0.9262
0.3115 5.3019 18000 0.3200 0.5487 0.5209 0.5165 0.6569 0.8506 0.9262
0.3065 5.5965 19000 0.3170 0.5813 0.5166 0.5157 0.6589 0.8522 0.9288
0.3133 5.8910 20000 0.3201 0.5424 0.5243 0.5215 0.6580 0.8498 0.9285
0.3099 6.1856 21000 0.3198 0.5389 0.5208 0.5184 0.6540 0.8487 0.9277
0.3055 6.4801 22000 0.3190 0.5391 0.5247 0.5205 0.6585 0.8497 0.9286
0.3176 6.7747 23000 0.3163 0.5574 0.5196 0.5185 0.6571 0.8511 0.9281
0.3135 7.0692 24000 0.3149 0.5579 0.5192 0.5181 0.6590 0.8519 0.9287
0.3061 7.3638 25000 0.3140 0.5540 0.5197 0.5185 0.6601 0.8520 0.9287
0.3214 7.6583 26000 0.3149 0.5534 0.5228 0.5205 0.6597 0.8513 0.9287
0.3024 7.9529 27000 0.3255 0.5368 0.5158 0.5149 0.6475 0.8486 0.9248
0.2981 8.2474 28000 0.3234 0.5361 0.5195 0.5176 0.6510 0.8476 0.9271
0.3057 8.5420 29000 0.3144 0.5445 0.5232 0.5212 0.6597 0.8508 0.9289
0.3041 8.8365 30000 0.3149 0.5629 0.5201 0.5189 0.6572 0.8517 0.9283
0.3082 9.1311 31000 0.3146 0.5477 0.5240 0.5221 0.6606 0.8518 0.9292
0.3125 9.4256 32000 0.3120 0.5709 0.5190 0.5178 0.6625 0.8531 0.9296
0.3072 9.7202 33000 0.3132 0.5721 0.5184 0.5186 0.6596 0.8530 0.9289
0.3141 10.0147 34000 0.3121 0.5493 0.5242 0.5224 0.6628 0.8523 0.9294
0.3022 10.3093 35000 0.3114 0.5696 0.5225 0.5206 0.6646 0.8530 0.9291
0.3042 10.6038 36000 0.3132 0.5511 0.5264 0.5234 0.6629 0.8521 0.9293
0.3063 10.8984 37000 0.3116 0.5688 0.5216 0.5189 0.6632 0.8531 0.9283
0.307 11.1929 38000 0.3109 0.5645 0.5220 0.5210 0.6635 0.8530 0.9294
0.314 11.4875 39000 0.3140 0.5618 0.5225 0.5189 0.6641 0.8530 0.9281
0.3111 11.7820 40000 0.3143 0.5288 0.5181 0.5144 0.6627 0.8528 0.9278
0.3279 12.0766 41000 0.3144 0.5602 0.5266 0.5242 0.6604 0.8506 0.9295
0.3121 12.3711 42000 0.3097 0.5611 0.5221 0.5210 0.6655 0.8536 0.9303
0.3055 12.6657 43000 0.3100 0.5571 0.5233 0.5220 0.6633 0.8535 0.9299
0.3162 12.9602 44000 0.3090 0.5650 0.5243 0.5230 0.6658 0.8537 0.9301
0.3056 13.2548 45000 0.3108 0.5991 0.5182 0.5180 0.6639 0.8537 0.9298
0.2897 13.5493 46000 0.3090 0.5596 0.5256 0.5234 0.6658 0.8541 0.9301
0.2988 13.8439 47000 0.3094 0.5728 0.5218 0.5207 0.6646 0.8541 0.9304
0.2959 14.1384 48000 0.3096 0.5538 0.5259 0.5243 0.6648 0.8529 0.9304
0.2986 14.4330 49000 0.3097 0.6318 0.5200 0.5185 0.6653 0.8534 0.9305
0.3047 14.7275 50000 0.3086 0.5668 0.5244 0.5230 0.6663 0.8539 0.9309
0.2932 15.0221 51000 0.3090 0.5612 0.5251 0.5237 0.6647 0.8537 0.9303
0.3075 15.3166 52000 0.3080 0.5818 0.5245 0.5226 0.6676 0.8545 0.9302
0.302 15.6112 53000 0.3108 0.5529 0.5272 0.5244 0.6629 0.8521 0.9300
0.3179 15.9057 54000 0.3081 0.5837 0.5225 0.5220 0.6645 0.8539 0.9302
0.3177 16.2003 55000 0.3100 0.5665 0.5238 0.5231 0.6624 0.8532 0.9299
0.3074 16.4948 56000 0.3079 0.5715 0.5240 0.5221 0.6672 0.8542 0.9303
0.2997 16.7894 57000 0.3074 0.5825 0.5235 0.5221 0.6670 0.8540 0.9307
0.3074 17.0839 58000 0.3074 0.5995 0.5216 0.5211 0.6658 0.8540 0.9308
0.3012 17.3785 59000 0.3070 0.5656 0.5259 0.5246 0.6671 0.8542 0.9303
0.3006 17.6730 60000 0.3072 0.5569 0.5275 0.5254 0.6675 0.8544 0.9306
0.3101 17.9676 61000 0.3069 0.5836 0.5254 0.5242 0.6668 0.8545 0.9308
0.2887 18.2622 62000 0.3094 0.5714 0.5256 0.5248 0.6627 0.8533 0.9295
0.3043 18.5567 63000 0.3068 0.5662 0.5279 0.5259 0.6690 0.8542 0.9310
0.2998 18.8513 64000 0.3087 0.5764 0.5264 0.5254 0.6641 0.8533 0.9304
0.3014 19.1458 65000 0.3061 0.5621 0.5281 0.5262 0.6686 0.8542 0.9312
0.3109 19.4404 66000 0.3098 0.5328 0.5182 0.5161 0.6663 0.8547 0.9304
0.3084 19.7349 67000 0.3107 0.5749 0.5265 0.5249 0.6615 0.8513 0.9298
0.2999 20.0295 68000 0.3061 0.5997 0.5248 0.5233 0.6681 0.8543 0.9310
0.3053 20.3240 69000 0.3073 0.5664 0.5256 0.5246 0.6652 0.8544 0.9304
0.3033 20.6186 70000 0.3102 0.5811 0.5270 0.5254 0.6623 0.8518 0.9297
0.3124 20.9131 71000 0.3057 0.6007 0.5250 0.5238 0.6685 0.8550 0.9313
0.3113 21.2077 72000 0.3080 0.5908 0.5257 0.5254 0.6643 0.8541 0.9300
0.2879 21.5022 73000 0.3067 0.5679 0.5240 0.5238 0.6664 0.8548 0.9307
0.2977 21.7968 74000 0.3064 0.5671 0.5285 0.5265 0.6677 0.8549 0.9311
0.2967 22.0913 75000 0.3054 0.5725 0.5269 0.5250 0.6697 0.8548 0.9309
0.2993 22.3859 76000 0.3060 0.6008 0.5234 0.5225 0.6686 0.8552 0.9314
0.2874 22.6804 77000 0.3054 0.6007 0.5237 0.5231 0.6679 0.8548 0.9313
0.3046 22.9750 78000 0.3053 0.6011 0.5269 0.5253 0.6689 0.8550 0.9314
0.2963 23.2695 79000 0.3056 0.7346 0.5232 0.5225 0.6682 0.8554 0.9314
0.3113 23.5641 80000 0.3052 0.5737 0.5275 0.5256 0.6698 0.8553 0.9314
0.3072 23.8586 81000 0.3054 0.5848 0.5246 0.5242 0.6679 0.8549 0.9312
0.3004 24.1532 82000 0.3055 0.5742 0.5254 0.5247 0.6671 0.8549 0.9310
0.2939 24.4477 83000 0.3055 0.5656 0.5289 0.5264 0.6700 0.8545 0.9312
0.3127 24.7423 84000 0.3050 0.5666 0.5288 0.5265 0.6693 0.8551 0.9313
0.3148 25.0368 85000 0.3046 0.5842 0.5265 0.5255 0.6698 0.8548 0.9316
0.3029 25.3314 86000 0.3048 0.6004 0.5269 0.5253 0.6693 0.8549 0.9315
0.3098 25.6259 87000 0.3070 0.5745 0.5253 0.5250 0.6657 0.8550 0.9302
0.2965 25.9205 88000 0.3046 0.5666 0.5278 0.5260 0.6699 0.8550 0.9314
0.2929 26.2150 89000 0.3047 0.5661 0.5278 0.5258 0.6703 0.8546 0.9314
0.2995 26.5096 90000 0.3048 0.6003 0.5255 0.5242 0.6695 0.8549 0.9313
0.2888 26.8041 91000 0.3057 0.5670 0.5278 0.5264 0.6674 0.8551 0.9307
0.2903 27.0987 92000 0.3045 0.6009 0.5264 0.5251 0.6690 0.8552 0.9315
0.2928 27.3932 93000 0.3049 0.5738 0.5269 0.5259 0.6681 0.8548 0.9310
0.2933 27.6878 94000 0.3044 0.5836 0.5267 0.5254 0.6688 0.8548 0.9314
0.305 27.9823 95000 0.3044 0.6011 0.5263 0.5252 0.6693 0.8552 0.9315
0.2952 28.2769 96000 0.3050 0.5671 0.5264 0.5255 0.6674 0.8548 0.9308
0.3055 28.5714 97000 0.3045 0.5737 0.5273 0.5259 0.6689 0.8548 0.9313
0.2978 28.8660 98000 0.3045 0.6007 0.5266 0.5254 0.6685 0.8548 0.9314
0.3093 29.1605 99000 0.3049 0.5741 0.5271 0.5259 0.6679 0.8550 0.9310
0.2949 29.4551 100000 0.3045 0.5839 0.5271 0.5258 0.6685 0.8550 0.9313
0.3012 29.7496 101000 0.3043 0.6007 0.5268 0.5255 0.6694 0.8552 0.9316

Framework versions

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