loubnabnl's picture
loubnabnl HF Staff
Model save
65d945d verified
|
raw
history blame
16.2 kB
metadata
base_model: bigcode/starencoder
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: classifier-llama3-java-500k
    results: []

classifier-llama3-java-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: 0.3339
  • Precision: 0.7086
  • Recall: 0.4035
  • F1 Macro: 0.4297
  • Accuracy: 0.6370
  • F1 Binary Minimum3: 0.7202
  • F1 Binary Minimum2: 0.9386

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 6.4612 0.0266 0.2 0.0470 0.1330 0 0
0.3818 0.2936 1000 0.3828 0.4885 0.3550 0.3697 0.6056 0.7105 0.9351
0.3773 0.5872 2000 0.3656 0.4854 0.3527 0.3704 0.6141 0.7005 0.9347
0.3586 0.8808 3000 0.3592 0.4868 0.3677 0.3898 0.6197 0.7037 0.9356
0.3726 1.1744 4000 0.3573 0.4863 0.3705 0.3927 0.6188 0.7089 0.9349
0.3624 1.4680 5000 0.3531 0.4963 0.3833 0.4059 0.6237 0.7191 0.9373
0.3586 1.7616 6000 0.3512 0.4953 0.3817 0.4051 0.6251 0.7153 0.9367
0.3584 2.0552 7000 0.3543 0.4865 0.3709 0.3941 0.6219 0.7010 0.9353
0.3628 2.3488 8000 0.3493 0.4976 0.3836 0.4070 0.6265 0.7179 0.9368
0.36 2.6424 9000 0.3543 0.4846 0.3710 0.3944 0.6212 0.6952 0.9349
0.3502 2.9360 10000 0.3520 0.4925 0.4040 0.4219 0.6258 0.7274 0.9369
0.3583 3.2296 11000 0.3478 0.4891 0.3895 0.4143 0.6257 0.7105 0.9360
0.3594 3.5232 12000 0.3483 0.4909 0.3958 0.4191 0.6268 0.7185 0.9368
0.3523 3.8168 13000 0.3510 0.4874 0.3773 0.4018 0.6237 0.7005 0.9352
0.3419 4.1104 14000 0.3528 0.4823 0.3869 0.4114 0.6234 0.7001 0.9341
0.3457 4.4040 15000 0.3485 0.4988 0.3956 0.4162 0.6274 0.7250 0.9369
0.3575 4.6976 16000 0.3500 0.5014 0.4013 0.4210 0.6268 0.7281 0.9377
0.3476 4.9912 17000 0.3499 0.5020 0.4029 0.4217 0.6275 0.7299 0.9379
0.3562 5.2848 18000 0.3454 0.4958 0.3933 0.4169 0.6288 0.7190 0.9371
0.3461 5.5784 19000 0.3451 0.4910 0.3904 0.4153 0.6281 0.7108 0.9359
0.3444 5.8720 20000 0.3436 0.4971 0.3864 0.4110 0.6287 0.7152 0.9364
0.3403 6.1656 21000 0.3450 0.4986 0.3982 0.4203 0.6298 0.7259 0.9373
0.3502 6.4592 22000 0.3452 0.4931 0.3828 0.4079 0.6276 0.7077 0.9361
0.3476 6.7528 23000 0.3542 0.7095 0.4026 0.4183 0.6243 0.7321 0.9376
0.3403 7.0464 24000 0.3431 0.4992 0.3905 0.4153 0.6308 0.7156 0.9369
0.3543 7.3400 25000 0.3443 0.5061 0.3950 0.4164 0.6303 0.7252 0.9379
0.3478 7.6336 26000 0.3430 0.4985 0.3962 0.4189 0.6305 0.7227 0.9372
0.3523 7.9272 27000 0.3425 0.4958 0.3941 0.4191 0.6309 0.7138 0.9367
0.3495 8.2208 28000 0.3429 0.4963 0.3880 0.4134 0.6300 0.7118 0.9368
0.3492 8.5144 29000 0.3492 0.7052 0.4057 0.4231 0.6278 0.7315 0.9376
0.3426 8.8080 30000 0.3498 0.7065 0.4029 0.4211 0.6267 0.7306 0.9379
0.3471 9.1016 31000 0.3440 0.5101 0.3913 0.4132 0.6302 0.7269 0.9378
0.3468 9.3952 32000 0.3429 0.5062 0.3969 0.4179 0.6315 0.7265 0.9376
0.3484 9.6888 33000 0.3410 0.4969 0.3975 0.4210 0.6312 0.7195 0.9370
0.3482 9.9824 34000 0.3406 0.5032 0.3898 0.4147 0.6314 0.7161 0.9375
0.3408 10.2760 35000 0.3409 0.4990 0.3990 0.4227 0.6320 0.7206 0.9376
0.3476 10.5696 36000 0.3404 0.5055 0.3874 0.4119 0.6314 0.7173 0.9379
0.3391 10.8632 37000 0.3429 0.7050 0.4040 0.4246 0.6315 0.7276 0.9381
0.3403 11.1568 38000 0.3409 0.7022 0.4016 0.4244 0.6331 0.7245 0.9377
0.3354 11.4504 39000 0.3425 0.7096 0.3982 0.4192 0.6317 0.7276 0.9378
0.3462 11.7440 40000 0.3412 0.4983 0.3915 0.4175 0.6324 0.7088 0.9371
0.3374 12.0376 41000 0.3437 0.4857 0.4019 0.4264 0.6297 0.7074 0.9350
0.3537 12.3312 42000 0.3430 0.7070 0.4052 0.4246 0.6323 0.7298 0.9378
0.3466 12.6248 43000 0.3388 0.7051 0.3952 0.4197 0.6336 0.7211 0.9378
0.342 12.9184 44000 0.3471 0.4908 0.3793 0.4047 0.6261 0.6961 0.9353
0.3437 13.2120 45000 0.3396 0.7069 0.3922 0.4161 0.6331 0.7193 0.9374
0.335 13.5056 46000 0.3395 0.7111 0.3953 0.4186 0.6338 0.7246 0.9382
0.3466 13.7992 47000 0.3384 0.7042 0.3986 0.4224 0.6333 0.7219 0.9378
0.353 14.0928 48000 0.3407 0.6951 0.3941 0.4209 0.6316 0.7058 0.9368
0.3444 14.3864 49000 0.3395 0.7014 0.3893 0.4159 0.6330 0.7119 0.9374
0.3328 14.6800 50000 0.3424 0.7149 0.4036 0.4235 0.6333 0.7313 0.9382
0.3374 14.9736 51000 0.3378 0.7023 0.3953 0.4205 0.6337 0.7168 0.9375
0.335 15.2672 52000 0.3385 0.7048 0.4025 0.4252 0.6342 0.7265 0.9379
0.3441 15.5608 53000 0.3384 0.7116 0.3918 0.4154 0.6336 0.7223 0.9380
0.3395 15.8544 54000 0.3409 0.6944 0.3944 0.4213 0.6318 0.7034 0.9367
0.3558 16.1480 55000 0.3373 0.7042 0.3983 0.4235 0.6354 0.7175 0.9378
0.328 16.4416 56000 0.3389 0.7083 0.4069 0.4292 0.6350 0.7280 0.9386
0.3416 16.7352 57000 0.3373 0.7048 0.4045 0.4284 0.6353 0.7227 0.9383
0.3275 17.0288 58000 0.3375 0.7096 0.4008 0.4250 0.6354 0.7241 0.9384
0.3528 17.3224 59000 0.3366 0.7077 0.3967 0.4213 0.6354 0.7196 0.9382
0.3504 17.6160 60000 0.3365 0.7039 0.3970 0.4220 0.6341 0.7166 0.9379
0.3292 17.9096 61000 0.3367 0.7064 0.4033 0.4280 0.6359 0.7229 0.9385
0.3382 18.2032 62000 0.3476 0.7160 0.4088 0.4247 0.6299 0.7363 0.9378
0.3349 18.4968 63000 0.3376 0.7017 0.3927 0.4195 0.6337 0.7074 0.9379
0.3401 18.7904 64000 0.3361 0.7030 0.3994 0.4251 0.6352 0.7154 0.9379
0.3365 19.0840 65000 0.3362 0.7016 0.4026 0.4276 0.6354 0.7168 0.9378
0.3385 19.3776 66000 0.3361 0.7031 0.3992 0.4256 0.6347 0.7134 0.9381
0.3395 19.6712 67000 0.3379 0.7092 0.4054 0.4282 0.6354 0.7263 0.9380
0.3383 19.9648 68000 0.3417 0.7138 0.4116 0.4313 0.6343 0.7337 0.9384
0.3356 20.2584 69000 0.3359 0.7086 0.3992 0.4235 0.6356 0.7221 0.9380
0.3363 20.5520 70000 0.3365 0.7088 0.4062 0.4305 0.6367 0.7260 0.9384
0.3333 20.8456 71000 0.3365 0.7018 0.3933 0.4201 0.6340 0.7091 0.9380
0.3298 21.1392 72000 0.3351 0.7074 0.3976 0.4239 0.6358 0.7160 0.9385
0.3372 21.4328 73000 0.3349 0.7061 0.4002 0.4261 0.6364 0.7191 0.9383
0.3424 21.7264 74000 0.3352 0.7052 0.3990 0.4256 0.6356 0.7153 0.9384
0.3393 22.0200 75000 0.3371 0.7007 0.3948 0.4223 0.6340 0.7072 0.9378
0.3305 22.3136 76000 0.3377 0.6977 0.3956 0.4222 0.6336 0.7065 0.9375
0.3343 22.6072 77000 0.3374 0.7115 0.4086 0.4310 0.6360 0.7283 0.9385
0.3378 22.9008 78000 0.3349 0.7064 0.4023 0.4286 0.6363 0.7180 0.9383
0.3443 23.1944 79000 0.3349 0.7061 0.4039 0.4309 0.6367 0.7170 0.9382
0.3464 23.4880 80000 0.3390 0.7120 0.4123 0.4332 0.6353 0.7313 0.9385
0.3355 23.7816 81000 0.3350 0.7107 0.4041 0.4287 0.6369 0.7237 0.9383
0.3312 24.0752 82000 0.3347 0.7080 0.3984 0.4258 0.6366 0.7152 0.9387
0.3526 24.3688 83000 0.3349 0.7087 0.4059 0.4309 0.6368 0.7237 0.9386
0.3438 24.6624 84000 0.3344 0.7090 0.3990 0.4248 0.6362 0.7201 0.9382
0.3365 24.9560 85000 0.3344 0.7092 0.4026 0.4293 0.6369 0.7194 0.9386
0.3529 25.2496 86000 0.3354 0.7081 0.4090 0.4337 0.6372 0.7263 0.9385
0.3398 25.5432 87000 0.3343 0.7114 0.4018 0.4280 0.6371 0.7215 0.9386
0.3349 25.8368 88000 0.3342 0.7074 0.4036 0.4305 0.6368 0.7173 0.9385
0.343 26.1304 89000 0.3348 0.7051 0.4011 0.4289 0.6365 0.7131 0.9382
0.3408 26.4240 90000 0.3345 0.7064 0.4002 0.4274 0.6368 0.7144 0.9384
0.3264 26.7176 91000 0.3344 0.7111 0.4038 0.4297 0.6374 0.7223 0.9387
0.3308 27.0112 92000 0.3341 0.7077 0.4025 0.4297 0.6369 0.7171 0.9387
0.3354 27.3048 93000 0.3364 0.7153 0.4059 0.4292 0.6368 0.7296 0.9384
0.3448 27.5984 94000 0.3340 0.7083 0.4024 0.4293 0.6367 0.7184 0.9386
0.3404 27.8920 95000 0.3340 0.7086 0.4014 0.4287 0.6372 0.7165 0.9386
0.3382 28.1856 96000 0.3339 0.7079 0.4023 0.4288 0.6367 0.7190 0.9384
0.3318 28.4792 97000 0.3339 0.7069 0.4049 0.4314 0.6369 0.7187 0.9385
0.331 28.7728 98000 0.3345 0.7106 0.4060 0.4310 0.6377 0.7249 0.9387
0.3372 29.0664 99000 0.3342 0.7104 0.4040 0.4293 0.6372 0.7228 0.9385
0.3401 29.3600 100000 0.3339 0.7084 0.4033 0.4295 0.6367 0.7196 0.9386
0.331 29.6536 101000 0.3341 0.7102 0.4045 0.4299 0.6372 0.7226 0.9386
0.3378 29.9472 102000 0.3339 0.7086 0.4035 0.4297 0.6370 0.7202 0.9386

Framework versions

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