Hugging Face's logo

weathon
/
3d2smiles_pretrain

Image-Text-to-Text
Transformers
TensorBoard
Safetensors
git
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
3d2smiles_pretrain / README.md
weathon's picture
weathon
End of training
703a785
preview code
|
raw
history blame
10 kB
metadata
library_name: transformers
license: mit
base_model: microsoft/git-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: 3d2smiles_pretrain
    results: []

3d2smiles_pretrain

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

  • Loss: 0.0246
  • Accuracy: 0.9767

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: 3e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6225 0.0224 100 0.5675 0.0465
0.3462 0.0448 200 0.3916 0.0465
0.2917 0.0672 300 0.3042 0.0814
0.2865 0.0896 400 0.2253 0.1163
0.2157 0.1120 500 0.1793 0.2209
0.1933 0.1344 600 0.1493 0.2442
0.1408 0.1568 700 0.1414 0.2791
0.1277 0.1792 800 0.1093 0.4709
0.1661 0.2016 900 0.1403 0.4012
0.0949 0.2240 1000 0.1010 0.4477
0.1025 0.2464 1100 0.0873 0.5407
0.069 0.2688 1200 0.0771 0.5
0.1072 0.2912 1300 0.0724 0.5465
0.0766 0.3136 1400 0.0780 0.6221
0.0933 0.3360 1500 0.0592 0.6453
0.0676 0.3584 1600 0.0643 0.6628
0.0812 0.3808 1700 0.0667 0.6453
0.0572 0.4032 1800 0.0561 0.6919
0.0459 0.4256 1900 0.0502 0.6221
0.0642 0.4480 2000 0.0538 0.6163
0.0597 0.4704 2100 0.0440 0.6919
0.0575 0.4928 2200 0.0531 0.6919
0.0341 0.5152 2300 0.0526 0.7326
0.0353 0.5376 2400 0.0493 0.6919
0.0569 0.5600 2500 0.0599 0.6977
0.043 0.5824 2600 0.0387 0.7616
0.0356 0.6048 2700 0.0353 0.75
0.0299 0.6272 2800 0.0331 0.8023
0.044 0.6496 2900 0.0497 0.7093
0.0402 0.6720 3000 0.0329 0.8023
0.0354 0.6944 3100 0.0277 0.7849
0.0353 0.7168 3200 0.0416 0.7558
0.0326 0.7392 3300 0.0502 0.7035
0.0434 0.7616 3400 0.0343 0.8140
0.0547 0.7840 3500 0.0303 0.8488
0.0208 0.8064 3600 0.0283 0.8837
0.0162 0.8288 3700 0.0327 0.8198
0.0282 0.8512 3800 0.0200 0.8605
0.0325 0.8736 3900 0.0182 0.8547
0.0238 0.8960 4000 0.0329 0.8314
0.0304 0.9184 4100 0.0330 0.7733
0.0184 0.9408 4200 0.0279 0.8663
0.0142 0.9632 4300 0.0172 0.8837
0.0274 0.9856 4400 0.0297 0.8140
0.0125 1.0078 4500 0.0315 0.8605
0.0205 1.0302 4600 0.0224 0.8779
0.0177 1.0526 4700 0.0307 0.8953
0.0257 1.0750 4800 0.0346 0.8605
0.0119 1.0974 4900 0.0229 0.8430
0.023 1.1198 5000 0.0366 0.8256
0.0262 1.1422 5100 0.0329 0.8198
0.0173 1.1646 5200 0.0523 0.7907
0.0144 1.1870 5300 0.0252 0.8837
0.0137 1.2094 5400 0.0277 0.8837
0.0205 1.2318 5500 0.0216 0.8895
0.0102 1.2542 5600 0.0279 0.8605
0.0079 1.2766 5700 0.0261 0.9244
0.025 1.2990 5800 0.0312 0.8605
0.0109 1.3214 5900 0.0267 0.8605
0.0155 1.3438 6000 0.0305 0.8140
0.0113 1.3662 6100 0.0225 0.8953
0.0119 1.3886 6200 0.0265 0.9244
0.0079 1.4110 6300 0.0194 0.9012
0.0131 1.4334 6400 0.0173 0.8837
0.0064 1.4558 6500 0.0245 0.8721
0.015 1.4782 6600 0.0142 0.8895
0.0141 1.5006 6700 0.0211 0.9651
0.0136 1.5230 6800 0.0278 0.8895
0.0197 1.5454 6900 0.0262 0.9012
0.0248 1.5678 7000 0.0192 0.8605
0.0076 1.5902 7100 0.0233 0.9186
0.0056 1.6126 7200 0.0234 0.8953
0.0097 1.6350 7300 0.0211 0.8430
0.0075 1.6574 7400 0.0098 0.9360
0.0103 1.6798 7500 0.0273 0.8721
0.0123 1.7022 7600 0.0058 0.9709
0.0088 1.7246 7700 0.0182 0.9070
0.014 1.7470 7800 0.0334 0.8779
0.0056 1.7694 7900 0.0223 0.8837
0.0096 1.7918 8000 0.0193 0.8837
0.0086 1.8142 8100 0.0285 0.9244
0.0047 1.8366 8200 0.0155 0.9535
0.0031 1.8590 8300 0.0254 0.9302
0.0168 1.8814 8400 0.0132 0.9186
0.0081 1.9038 8500 0.0210 0.9128
0.0069 1.9262 8600 0.0142 0.8663
0.0038 1.9486 8700 0.0082 0.9302
0.0077 1.9710 8800 0.0133 0.9419
0.0025 1.9934 8900 0.0174 0.9012
0.0075 2.0157 9000 0.0201 0.9477
0.0008 2.0381 9100 0.0199 0.9302
0.0081 2.0605 9200 0.0255 0.9012
0.003 2.0829 9300 0.0096 0.9535
0.0027 2.1053 9400 0.0203 0.9128
0.0028 2.1277 9500 0.0249 0.9244
0.0023 2.1501 9600 0.0240 0.9128
0.003 2.1725 9700 0.0305 0.9767
0.0063 2.1949 9800 0.0305 0.9593
0.0041 2.2173 9900 0.0225 0.9419
0.0021 2.2397 10000 0.0193 0.9360
0.0065 2.2621 10100 0.0171 0.8895
0.0018 2.2845 10200 0.0170 0.9535
0.011 2.3069 10300 0.0257 0.9360
0.0047 2.3293 10400 0.0128 0.9651
0.0028 2.3517 10500 0.0226 0.9360
0.0034 2.3741 10600 0.0250 0.8488
0.0032 2.3965 10700 0.0211 0.9535
0.001 2.4189 10800 0.0233 0.9767
0.0021 2.4413 10900 0.0201 0.9477
0.0032 2.4637 11000 0.0205 0.9419
0.002 2.4861 11100 0.0222 0.9767
0.0029 2.5085 11200 0.0230 0.9651
0.0024 2.5309 11300 0.0220 0.9767
0.003 2.5533 11400 0.0224 0.9767
0.0034 2.5757 11500 0.0222 0.9593
0.0002 2.5981 11600 0.0202 0.9709
0.0015 2.6205 11700 0.0188 0.9651
0.0035 2.6428 11800 0.0218 0.9651
0.0075 2.6652 11900 0.0268 0.9302
0.0008 2.6876 12000 0.0219 0.9651
0.0014 2.7100 12100 0.0247 0.9767
0.0013 2.7324 12200 0.0234 0.9767
0.0006 2.7548 12300 0.0254 0.9767
0.0008 2.7772 12400 0.0247 0.9709
0.0005 2.7996 12500 0.0252 0.9651
0.0011 2.8220 12600 0.0222 0.9767
0.0002 2.8444 12700 0.0248 0.9767
0.0042 2.8668 12800 0.0251 0.9709
0.0012 2.8892 12900 0.0235 0.9767
0.0004 2.9116 13000 0.0254 0.9593
0.0013 2.9340 13100 0.0257 0.9593
0.0029 2.9564 13200 0.0264 0.9651
0.0008 2.9788 13300 0.0246 0.9767

Framework versions

  • Transformers 4.48.1
  • Pytorch 2.5.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.21.0