t5-base-finetuned-en-to-fr

This model is a fine-tuned version of Demosthene-OR/t5-base-finetuned-en-to-fr on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0189
  • Bleu: 99.5662
  • Gen Len: 5.9091

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: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 150

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
No log 1.0 2 0.9476 91.8099 6.2727
No log 2.0 4 0.8626 91.2967 6.3636
No log 3.0 6 0.7928 91.2967 6.3636
No log 4.0 8 0.7310 91.2967 6.3636
No log 5.0 10 0.6734 91.2967 6.3636
No log 6.0 12 0.6195 91.7685 6.4545
No log 7.0 14 0.5711 92.7714 6.3636
No log 8.0 16 0.5311 92.7714 6.3636
No log 9.0 18 0.4952 93.8091 6.2727
No log 10.0 20 0.4598 93.8091 6.2727
No log 11.0 22 0.4246 93.8091 6.2727
No log 12.0 24 0.3919 93.8091 6.2727
No log 13.0 26 0.3602 93.8091 6.2727
No log 14.0 28 0.3308 93.8091 6.2727
No log 15.0 30 0.3050 93.8091 6.2727
No log 16.0 32 0.2828 94.9904 6.2273
No log 17.0 34 0.2633 95.4618 5.8636
No log 18.0 36 0.2452 95.4618 5.8636
No log 19.0 38 0.2278 96.624 5.8182
No log 20.0 40 0.2143 96.0313 5.9545
No log 21.0 42 0.2027 96.0313 5.9545
No log 22.0 44 0.1921 96.4897 5.9091
No log 23.0 46 0.1831 96.4897 5.9091
No log 24.0 48 0.1739 96.4897 5.9091
No log 25.0 50 0.1648 96.4897 5.9091
No log 26.0 52 0.1570 96.4897 5.9091
No log 27.0 54 0.1511 96.9417 5.8182
No log 28.0 56 0.1458 96.9417 5.8182
No log 29.0 58 0.1409 96.9417 5.8182
No log 30.0 60 0.1377 96.9417 5.8182
No log 31.0 62 0.1342 96.9417 5.8182
No log 32.0 64 0.1313 96.9417 5.8182
No log 33.0 66 0.1282 96.9417 5.8182
No log 34.0 68 0.1252 96.9417 5.8182
No log 35.0 70 0.1216 96.9417 5.8182
No log 36.0 72 0.1172 96.9417 5.8182
No log 37.0 74 0.1128 96.9417 5.8182
No log 38.0 76 0.1094 96.9417 5.8182
No log 39.0 78 0.1060 96.9417 5.8182
No log 40.0 80 0.1033 97.3874 5.8636
No log 41.0 82 0.1007 97.3874 5.8636
No log 42.0 84 0.0978 97.3874 5.8636
No log 43.0 86 0.0945 97.3874 5.8636
No log 44.0 88 0.0912 97.3874 5.8636
No log 45.0 90 0.0882 97.3874 5.8636
No log 46.0 92 0.0855 97.3874 5.8636
No log 47.0 94 0.0835 97.3874 5.8636
No log 48.0 96 0.0818 97.3874 6.1364
No log 49.0 98 0.0800 97.3874 6.1364
No log 50.0 100 0.0781 97.3874 6.1364
No log 51.0 102 0.0764 97.3874 6.1364
No log 52.0 104 0.0748 97.3874 5.8636
No log 53.0 106 0.0735 97.3874 5.8636
No log 54.0 108 0.0726 97.3874 5.8636
No log 55.0 110 0.0716 97.3874 5.8636
No log 56.0 112 0.0707 97.3874 5.8636
No log 57.0 114 0.0702 97.3874 5.8636
No log 58.0 116 0.0705 97.3874 5.8636
No log 59.0 118 0.0705 97.3874 5.8636
No log 60.0 120 0.0703 97.3874 5.8636
No log 61.0 122 0.0708 97.3874 5.8636
No log 62.0 124 0.0707 97.3874 5.8636
No log 63.0 126 0.0714 97.3874 5.8636
No log 64.0 128 0.0717 97.3874 5.8636
No log 65.0 130 0.0717 97.3874 5.8636
No log 66.0 132 0.0707 97.3874 5.8636
No log 67.0 134 0.0700 97.3874 5.8636
No log 68.0 136 0.0682 97.3874 5.8636
No log 69.0 138 0.0669 97.3874 5.8636
No log 70.0 140 0.0652 97.3874 5.8636
No log 71.0 142 0.0630 97.3874 5.8636
No log 72.0 144 0.0615 97.3874 5.8636
No log 73.0 146 0.0595 97.3874 5.8636
No log 74.0 148 0.0580 97.3874 5.8636
No log 75.0 150 0.0566 97.3874 5.8636
No log 76.0 152 0.0557 97.3874 5.8636
No log 77.0 154 0.0545 97.3874 5.8636
No log 78.0 156 0.0535 97.3874 5.8636
No log 79.0 158 0.0523 97.3874 5.8636
No log 80.0 160 0.0511 97.3874 5.8636
No log 81.0 162 0.0499 97.3874 5.8636
No log 82.0 164 0.0490 97.3874 5.8636
No log 83.0 166 0.0482 97.3874 5.8636
No log 84.0 168 0.0474 97.3874 5.8636
No log 85.0 170 0.0466 97.3874 5.8636
No log 86.0 172 0.0458 97.3874 5.8636
No log 87.0 174 0.0449 97.3874 5.8636
No log 88.0 176 0.0439 97.3874 5.8636
No log 89.0 178 0.0428 97.3874 5.8636
No log 90.0 180 0.0423 97.3874 5.8636
No log 91.0 182 0.0419 97.3874 5.8636
No log 92.0 184 0.0415 97.3874 5.8636
No log 93.0 186 0.0411 97.3874 5.8636
No log 94.0 188 0.0409 97.3874 5.8636
No log 95.0 190 0.0404 97.3874 5.8636
No log 96.0 192 0.0398 98.4309 5.9091
No log 97.0 194 0.0394 98.4309 5.9091
No log 98.0 196 0.0391 98.4309 5.9091
No log 99.0 198 0.0388 97.3874 5.8636
No log 100.0 200 0.0385 97.3874 5.8636
No log 101.0 202 0.0381 99.5662 5.9091
No log 102.0 204 0.0373 99.5662 5.9091
No log 103.0 206 0.0365 99.5662 5.9091
No log 104.0 208 0.0356 99.5662 5.9091
No log 105.0 210 0.0345 99.5662 5.9091
No log 106.0 212 0.0334 99.5662 5.9091
No log 107.0 214 0.0324 99.5662 5.9091
No log 108.0 216 0.0315 99.5662 5.9091
No log 109.0 218 0.0305 99.5662 5.9091
No log 110.0 220 0.0294 99.5662 5.9091
No log 111.0 222 0.0283 99.5662 5.9091
No log 112.0 224 0.0273 99.5662 5.9091
No log 113.0 226 0.0264 99.5662 5.9091
No log 114.0 228 0.0257 99.5662 5.9091
No log 115.0 230 0.0251 99.5662 5.9091
No log 116.0 232 0.0246 99.5662 5.9091
No log 117.0 234 0.0241 99.5662 5.9091
No log 118.0 236 0.0237 99.5662 5.9091
No log 119.0 238 0.0233 99.5662 5.9091
No log 120.0 240 0.0228 99.5662 5.9091
No log 121.0 242 0.0224 99.5662 5.9091
No log 122.0 244 0.0221 99.5662 5.9091
No log 123.0 246 0.0217 99.5662 5.9091
No log 124.0 248 0.0212 99.5662 5.9091
No log 125.0 250 0.0208 99.5662 5.9091
No log 126.0 252 0.0204 99.5662 5.9091
No log 127.0 254 0.0201 99.5662 5.9091
No log 128.0 256 0.0199 99.5662 5.9091
No log 129.0 258 0.0197 99.5662 5.9091
No log 130.0 260 0.0197 99.5662 5.9091
No log 131.0 262 0.0196 99.5662 5.9091
No log 132.0 264 0.0197 99.5662 5.9091
No log 133.0 266 0.0197 99.5662 5.9091
No log 134.0 268 0.0196 99.5662 5.9091
No log 135.0 270 0.0195 99.5662 5.9091
No log 136.0 272 0.0193 99.5662 5.9091
No log 137.0 274 0.0191 99.5662 5.9091
No log 138.0 276 0.0190 99.5662 5.9091
No log 139.0 278 0.0190 99.5662 5.9091
No log 140.0 280 0.0189 99.5662 5.9091
No log 141.0 282 0.0189 99.5662 5.9091
No log 142.0 284 0.0189 99.5662 5.9091
No log 143.0 286 0.0189 99.5662 5.9091
No log 144.0 288 0.0189 99.5662 5.9091
No log 145.0 290 0.0189 99.5662 5.9091
No log 146.0 292 0.0189 99.5662 5.9091
No log 147.0 294 0.0189 99.5662 5.9091
No log 148.0 296 0.0189 99.5662 5.9091
No log 149.0 298 0.0189 99.5662 5.9091
No log 150.0 300 0.0189 99.5662 5.9091

Framework versions

  • Transformers 4.33.1
  • Pytorch 2.0.1
  • Datasets 2.13.0
  • Tokenizers 0.13.2
Downloads last month
32
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Demosthene-OR/t5-base-finetuned-en-to-fr

Unable to build the model tree, the base model loops to the model itself. Learn more.