metadata
license: bsd-3-clause
base_model: Salesforce/codet5p-770m-py
tags:
- generated_from_trainer
datasets:
- mbpp
model-index:
- name: codet5p-770m-py-sanitized-codebleu-1-True-5e-05-0.1-lora-layer_7
results: []
codet5p-770m-py-sanitized-codebleu-1-True-5e-05-0.1-lora-layer_7
This model is a fine-tuned version of Salesforce/codet5p-770m-py on the mbpp dataset. It achieves the following results on the evaluation set:
- Loss: 0.7264
- Codebleu: 0.1085
- Ngram Match Score: 0.0170
- Weighted Ngram Match Score: 0.0403
- Syntax Match Score: 0.1283
- Dataflow Match Score: 0.1285
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 64
Training results
Training Loss | Epoch | Step | Validation Loss | Codebleu | Ngram Match Score | Weighted Ngram Match Score | Syntax Match Score | Dataflow Match Score |
---|---|---|---|---|---|---|---|---|
0.9826 | 1.0 | 15 | 0.9245 | 0.0072 | 0.0000 | 0.0000 | 0.0079 | 0.0100 |
0.9647 | 2.0 | 30 | 0.9231 | 0.0072 | 0.0000 | 0.0000 | 0.0079 | 0.0100 |
0.9702 | 3.0 | 45 | 0.9200 | 0.0072 | 0.0000 | 0.0000 | 0.0079 | 0.0100 |
0.9452 | 4.0 | 60 | 0.9132 | 0.0080 | 0.0000 | 0.0000 | 0.0079 | 0.0120 |
0.9489 | 5.0 | 75 | 0.8993 | 0.0285 | 0.0000 | 0.0125 | 0.0238 | 0.0442 |
0.9429 | 6.0 | 90 | 0.8758 | 0.0629 | 0.0034 | 0.0295 | 0.0728 | 0.0763 |
0.927 | 7.0 | 105 | 0.8542 | 0.0967 | 0.0231 | 0.0527 | 0.1243 | 0.0984 |
0.8731 | 8.0 | 120 | 0.8435 | 0.0949 | 0.0224 | 0.0517 | 0.1243 | 0.0944 |
0.8557 | 9.0 | 135 | 0.8340 | 0.0948 | 0.0212 | 0.0518 | 0.1243 | 0.0944 |
0.8533 | 10.0 | 150 | 0.8233 | 0.0926 | 0.0192 | 0.0500 | 0.1177 | 0.0964 |
0.8407 | 11.0 | 165 | 0.8117 | 0.0991 | 0.0123 | 0.0371 | 0.1270 | 0.1084 |
0.8339 | 12.0 | 180 | 0.7991 | 0.0992 | 0.0113 | 0.0360 | 0.1257 | 0.1104 |
0.814 | 13.0 | 195 | 0.7891 | 0.0998 | 0.0104 | 0.0353 | 0.1217 | 0.1165 |
0.7913 | 14.0 | 210 | 0.7797 | 0.0930 | 0.0096 | 0.0341 | 0.1190 | 0.1024 |
0.803 | 15.0 | 225 | 0.7729 | 0.0968 | 0.0107 | 0.0347 | 0.1283 | 0.1024 |
0.8047 | 16.0 | 240 | 0.7674 | 0.0945 | 0.0110 | 0.0347 | 0.1243 | 0.1004 |
0.7931 | 17.0 | 255 | 0.7632 | 0.1013 | 0.0131 | 0.0368 | 0.1323 | 0.1084 |
0.7705 | 18.0 | 270 | 0.7592 | 0.1006 | 0.0143 | 0.0397 | 0.1296 | 0.1084 |
0.7823 | 19.0 | 285 | 0.7549 | 0.1006 | 0.0143 | 0.0397 | 0.1296 | 0.1084 |
0.765 | 20.0 | 300 | 0.7513 | 0.1042 | 0.0122 | 0.0352 | 0.1323 | 0.1165 |
0.7549 | 21.0 | 315 | 0.7478 | 0.1054 | 0.0127 | 0.0358 | 0.1310 | 0.1205 |
0.7592 | 22.0 | 330 | 0.7451 | 0.1055 | 0.0129 | 0.0358 | 0.1310 | 0.1205 |
0.754 | 23.0 | 345 | 0.7437 | 0.1055 | 0.0129 | 0.0358 | 0.1310 | 0.1205 |
0.7494 | 24.0 | 360 | 0.7425 | 0.1121 | 0.0133 | 0.0358 | 0.1376 | 0.1305 |
0.7748 | 25.0 | 375 | 0.7409 | 0.1121 | 0.0130 | 0.0352 | 0.1376 | 0.1305 |
0.7439 | 26.0 | 390 | 0.7393 | 0.1114 | 0.0138 | 0.0358 | 0.1376 | 0.1285 |
0.7302 | 27.0 | 405 | 0.7377 | 0.1123 | 0.0133 | 0.0352 | 0.1402 | 0.1285 |
0.737 | 28.0 | 420 | 0.7369 | 0.1124 | 0.0135 | 0.0352 | 0.1402 | 0.1285 |
0.7246 | 29.0 | 435 | 0.7369 | 0.1103 | 0.0138 | 0.0358 | 0.1349 | 0.1285 |
0.7269 | 30.0 | 450 | 0.7361 | 0.1103 | 0.0137 | 0.0352 | 0.1349 | 0.1285 |
0.7233 | 31.0 | 465 | 0.7346 | 0.1103 | 0.0137 | 0.0352 | 0.1349 | 0.1285 |
0.7276 | 32.0 | 480 | 0.7331 | 0.1077 | 0.0143 | 0.0356 | 0.1323 | 0.1245 |
0.7161 | 33.0 | 495 | 0.7323 | 0.1077 | 0.0141 | 0.0356 | 0.1323 | 0.1245 |
0.7294 | 34.0 | 510 | 0.7313 | 0.1077 | 0.0141 | 0.0356 | 0.1323 | 0.1245 |
0.7071 | 35.0 | 525 | 0.7303 | 0.1039 | 0.0125 | 0.0340 | 0.1257 | 0.1225 |
0.7358 | 36.0 | 540 | 0.7299 | 0.1039 | 0.0125 | 0.0340 | 0.1257 | 0.1225 |
0.7174 | 37.0 | 555 | 0.7299 | 0.1050 | 0.0125 | 0.0340 | 0.1283 | 0.1225 |
0.723 | 38.0 | 570 | 0.7304 | 0.1066 | 0.0129 | 0.0340 | 0.1283 | 0.1265 |
0.7125 | 39.0 | 585 | 0.7300 | 0.1066 | 0.0129 | 0.0340 | 0.1283 | 0.1265 |
0.7215 | 40.0 | 600 | 0.7294 | 0.1066 | 0.0129 | 0.0340 | 0.1283 | 0.1265 |
0.7039 | 41.0 | 615 | 0.7291 | 0.1066 | 0.0129 | 0.0340 | 0.1283 | 0.1265 |
0.7034 | 42.0 | 630 | 0.7287 | 0.1050 | 0.0128 | 0.0340 | 0.1283 | 0.1225 |
0.7063 | 43.0 | 645 | 0.7292 | 0.1050 | 0.0131 | 0.0340 | 0.1283 | 0.1225 |
0.7128 | 44.0 | 660 | 0.7297 | 0.1059 | 0.0136 | 0.0340 | 0.1283 | 0.1245 |
0.7065 | 45.0 | 675 | 0.7297 | 0.1059 | 0.0136 | 0.0340 | 0.1283 | 0.1245 |
0.7037 | 46.0 | 690 | 0.7289 | 0.1059 | 0.0136 | 0.0340 | 0.1283 | 0.1245 |
0.7103 | 47.0 | 705 | 0.7277 | 0.1059 | 0.0136 | 0.0340 | 0.1283 | 0.1245 |
0.7055 | 48.0 | 720 | 0.7275 | 0.1059 | 0.0134 | 0.0340 | 0.1283 | 0.1245 |
0.7018 | 49.0 | 735 | 0.7276 | 0.1075 | 0.0140 | 0.0340 | 0.1283 | 0.1285 |
0.6889 | 50.0 | 750 | 0.7281 | 0.1074 | 0.0134 | 0.0336 | 0.1283 | 0.1285 |
0.7159 | 51.0 | 765 | 0.7279 | 0.1074 | 0.0132 | 0.0336 | 0.1283 | 0.1285 |
0.7091 | 52.0 | 780 | 0.7276 | 0.1074 | 0.0132 | 0.0336 | 0.1283 | 0.1285 |
0.7064 | 53.0 | 795 | 0.7270 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.6773 | 54.0 | 810 | 0.7267 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.7052 | 55.0 | 825 | 0.7262 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.7047 | 56.0 | 840 | 0.7265 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.7119 | 57.0 | 855 | 0.7263 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.6952 | 58.0 | 870 | 0.7263 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.709 | 59.0 | 885 | 0.7262 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.6899 | 60.0 | 900 | 0.7262 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.7147 | 61.0 | 915 | 0.7264 | 0.1085 | 0.0169 | 0.0403 | 0.1283 | 0.1285 |
0.699 | 62.0 | 930 | 0.7264 | 0.1085 | 0.0170 | 0.0403 | 0.1283 | 0.1285 |
0.6958 | 63.0 | 945 | 0.7264 | 0.1085 | 0.0170 | 0.0403 | 0.1283 | 0.1285 |
0.69 | 64.0 | 960 | 0.7264 | 0.1085 | 0.0170 | 0.0403 | 0.1283 | 0.1285 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1
- Datasets 2.14.4
- Tokenizers 0.13.3