Hugging Face's logo

ManojAlexender
/
final_roberta_with_new_400k_plus_37k

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
Model card Files Community
final_roberta_with_new_400k_plus_37k / README.md
ManojAlexender's picture
ManojAlexender
End of training
cf587bc verified
preview code
|
raw
history blame contribute delete
9.13 kB
metadata
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: final_roberta_with_new_400k_plus_37k
    results: []

final_roberta_with_new_400k_plus_37k

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2668
  • Accuracy: 0.9031
  • F1: 0.9027
  • Precision: 0.9042
  • Recall: 0.9031

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: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.3174 0.01 100 0.3254 0.8928 0.8919 0.8964 0.8928
0.3285 0.01 200 0.2578 0.8956 0.8951 0.8968 0.8956
0.247 0.02 300 0.3913 0.8621 0.8588 0.8783 0.8621
0.2853 0.03 400 0.3394 0.8737 0.8711 0.8871 0.8737
0.3031 0.04 500 0.3924 0.8537 0.8491 0.8770 0.8537
0.2747 0.04 600 0.2532 0.9079 0.9079 0.9080 0.9079
0.2797 0.05 700 0.3607 0.8637 0.8607 0.8781 0.8637
0.2211 0.06 800 0.2910 0.8880 0.8872 0.8909 0.8880
0.2769 0.06 900 0.2834 0.8824 0.8810 0.8884 0.8824
0.2412 0.07 1000 0.2394 0.9063 0.9061 0.9069 0.9063
0.3386 0.08 1100 0.2400 0.9016 0.9013 0.9020 0.9016
0.2743 0.09 1200 0.2421 0.9047 0.9048 0.9048 0.9047
0.2682 0.09 1300 0.2833 0.8768 0.8752 0.8839 0.8768
0.3219 0.1 1400 0.2383 0.9071 0.9070 0.9071 0.9071
0.2211 0.11 1500 0.2454 0.9047 0.9047 0.9047 0.9047
0.2606 0.11 1600 0.2083 0.9223 0.9221 0.9228 0.9223
0.1966 0.12 1700 0.2688 0.9004 0.9001 0.9007 0.9004
0.2205 0.13 1800 0.3076 0.8776 0.8752 0.8911 0.8776
0.2242 0.14 1900 0.2171 0.9151 0.9150 0.9153 0.9151
0.257 0.14 2000 0.2643 0.8912 0.8905 0.8932 0.8912
0.2238 0.15 2100 0.2165 0.9131 0.9128 0.9141 0.9131
0.2313 0.16 2200 0.2312 0.8996 0.8996 0.8996 0.8996
0.1856 0.16 2300 0.2269 0.9107 0.9108 0.9109 0.9107
0.2201 0.17 2400 0.2425 0.9059 0.9056 0.9065 0.9059
0.3332 0.18 2500 0.2254 0.9043 0.9044 0.9048 0.9043
0.1843 0.19 2600 0.2524 0.8980 0.8971 0.9020 0.8980
0.2728 0.19 2700 0.2348 0.8968 0.8957 0.9017 0.8968
0.2131 0.2 2800 0.2210 0.9135 0.9136 0.9138 0.9135
0.19 0.21 2900 0.2259 0.9123 0.9120 0.9130 0.9123
0.2099 0.21 3000 0.2814 0.9024 0.9016 0.9054 0.9024
0.2209 0.22 3100 0.2473 0.9051 0.9046 0.9070 0.9051
0.2366 0.23 3200 0.2561 0.8992 0.8983 0.9029 0.8992
0.3156 0.24 3300 0.2192 0.9095 0.9094 0.9095 0.9095
0.197 0.24 3400 0.2382 0.9063 0.9057 0.9093 0.9063
0.2371 0.25 3500 0.2243 0.9139 0.9141 0.9166 0.9139
0.2273 0.26 3600 0.2362 0.9135 0.9131 0.9153 0.9135
0.2504 0.26 3700 0.2671 0.8888 0.8873 0.8965 0.8888
0.1978 0.27 3800 0.2049 0.9171 0.9170 0.9172 0.9171
0.2189 0.28 3900 0.2268 0.9099 0.9099 0.9099 0.9099
0.2171 0.29 4000 0.2135 0.9163 0.9162 0.9164 0.9163
0.2325 0.29 4100 0.2624 0.8916 0.8905 0.8966 0.8916
0.1888 0.3 4200 0.2878 0.8924 0.8911 0.8987 0.8924
0.2345 0.31 4300 0.2444 0.8964 0.8953 0.9013 0.8964
0.1688 0.31 4400 0.2479 0.9083 0.9077 0.9109 0.9083
0.2083 0.32 4500 0.2200 0.9135 0.9131 0.9150 0.9135
0.2475 0.33 4600 0.2353 0.9035 0.9030 0.9052 0.9035
0.1928 0.34 4700 0.2987 0.8944 0.8933 0.8992 0.8944
0.2008 0.34 4800 0.2993 0.8760 0.8735 0.8897 0.8760
0.22 0.35 4900 0.2431 0.9035 0.9033 0.9039 0.9035
0.1844 0.36 5000 0.2590 0.9171 0.9171 0.9171 0.9171
0.2235 0.36 5100 0.2421 0.9047 0.9041 0.9072 0.9047
0.2222 0.37 5200 0.2958 0.8948 0.8941 0.8973 0.8948
0.2241 0.38 5300 0.2031 0.9211 0.9209 0.9216 0.9211
0.2307 0.39 5400 0.2277 0.9043 0.9036 0.9076 0.9043
0.1926 0.39 5500 0.2817 0.8900 0.8887 0.8959 0.8900
0.2119 0.4 5600 0.2151 0.9175 0.9174 0.9176 0.9175
0.1747 0.41 5700 0.2404 0.9123 0.9121 0.9126 0.9123
0.1809 0.41 5800 0.3013 0.8920 0.8908 0.8980 0.8920
0.1748 0.42 5900 0.3084 0.9063 0.9056 0.9097 0.9063
0.2101 0.43 6000 0.2129 0.9175 0.9173 0.9180 0.9175
0.202 0.44 6100 0.3794 0.8848 0.8834 0.8914 0.8848
0.1671 0.44 6200 0.2678 0.9043 0.9041 0.9046 0.9043
0.2808 0.45 6300 0.2613 0.9075 0.9070 0.9098 0.9075
0.2853 0.46 6400 0.2270 0.9087 0.9088 0.9088 0.9087
0.187 0.46 6500 0.2400 0.9111 0.9112 0.9115 0.9111
0.1382 0.47 6600 0.2454 0.9139 0.9136 0.9146 0.9139
0.2259 0.48 6700 0.3165 0.8904 0.8890 0.8976 0.8904
0.164 0.49 6800 0.3091 0.9031 0.9023 0.9074 0.9031
0.2557 0.49 6900 0.2708 0.9024 0.9015 0.9064 0.9024
0.1586 0.5 7000 0.2139 0.9247 0.9246 0.9247 0.9247
0.2391 0.51 7100 0.2087 0.9143 0.9141 0.9147 0.9143
0.1974 0.51 7200 0.2438 0.9171 0.9171 0.9171 0.9171
0.2507 0.52 7300 0.2323 0.9051 0.9044 0.9084 0.9051
0.226 0.53 7400 0.2465 0.9063 0.9056 0.9096 0.9063
0.1859 0.54 7500 0.2762 0.8960 0.8949 0.9012 0.8960
0.2208 0.54 7600 0.2705 0.8948 0.8937 0.9002 0.8948
0.2073 0.55 7700 0.2419 0.9008 0.9006 0.9007 0.9008
0.1557 0.56 7800 0.3004 0.8900 0.8899 0.8899 0.8900
0.1872 0.56 7900 0.2520 0.9059 0.9056 0.9066 0.9059
0.1749 0.57 8000 0.2757 0.9067 0.9067 0.9067 0.9067
0.2298 0.58 8100 0.2617 0.9075 0.9071 0.9092 0.9075
0.1781 0.59 8200 0.2380 0.9139 0.9137 0.9144 0.9139
0.1448 0.59 8300 0.2668 0.9031 0.9027 0.9042 0.9031

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1