ONNX Model Zoo

carted-nlp/categorization-finetuned-20220721-164940-distilled-20220811-013354

By carted-nlp · ↓ 4 · ♥ 0 · Updated 2022-08-12

What's in this repo

This page lists the files in carted-nlp/categorization-finetuned-20220721-164940-distilled-20220811-013354, including categorization-finetuned-20220721-164940-distilled-20220811-013354-default.onnx, categorization-finetuned-20220721-164940-distilled-20220811-013354-int8.onnx.

  • 2 ONNX models (.onnx files)
  • 20 files, including 9 .json, 2 .onnx, 2 .bin, 1 .md, 1 .txt, 1 .1
  • Last updated 2022-08-12

ONNX files

File Size Download
categorization-finetuned-20220721-164940-distilled-20220811-013354-default.onnx — Download
categorization-finetuned-20220721-164940-distilled-20220811-013354-int8.onnx — Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

categorization-finetuned-20220721-164940-distilled-20220811-013354

This model is a fine-tuned version of carted-nlp/categorization-finetuned-20220721-164940 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0645
  • Accuracy: 0.8776
  • F1: 0.8768

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: 64
  • eval_batch_size: 96
  • seed: 314
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 256
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 1500
  • num_epochs: 30.0

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.2702 0.56 2500 0.1290 0.7832 0.7783
0.1246 1.12 5000 0.1047 0.8169 0.8137
0.1066 1.69 7500 0.0945 0.8301 0.8276
0.0975 2.25 10000 0.0888 0.8386 0.8367
0.0917 2.81 12500 0.0849 0.8445 0.8428
0.0865 3.37 15000 0.0818 0.8496 0.8484
0.0835 3.94 17500 0.0796 0.8526 0.8509
0.08 4.5 20000 0.0777 0.8552 0.8542
0.0778 5.06 22500 0.0763 0.8580 0.8567
0.0753 5.62 25000 0.0744 0.8604 0.8592
0.0739 6.19 27500 0.0738 0.8614 0.8603
0.0716 6.75 30000 0.0729 0.8630 0.8620
0.0701 7.31 32500 0.0719 0.8645 0.8638
0.0689 7.87 35000 0.0708 0.8657 0.8647
0.067 8.43 37500 0.0705 0.8671 0.8660
0.0669 9.0 40000 0.0699 0.8681 0.8674
0.0647 9.56 42500 0.0697 0.8683 0.8673
0.0641 10.12 45000 0.0693 0.8691 0.8681
0.063 10.68 47500 0.0685 0.8702 0.8694
0.0618 11.25 50000 0.0681 0.8709 0.8701
0.0614 11.81 52500 0.0675 0.8720 0.8712
0.0601 12.37 55000 0.0678 0.8724 0.8713
0.0598 12.93 57500 0.0670 0.8732 0.8725
0.0584 13.5 60000 0.0670 0.8732 0.8723
0.0584 14.06 62500 0.0665 0.8740 0.8732
0.0572 14.62 65000 0.0665 0.8744 0.8734
0.0567 15.18 67500 0.0661 0.8753 0.8745
0.0561 15.74 70000 0.0660 0.8756 0.8750
0.0554 16.31 72500 0.0661 0.8759 0.8751
0.0552 16.87 75000 0.0656 0.8755 0.8749
0.0544 17.43 77500 0.0657 0.8762 0.8754
0.0544 17.99 80000 0.0654 0.8767 0.8760
0.0534 18.56 82500 0.0654 0.8767 0.8759
0.0534 19.12 85000 0.0653 0.8773 0.8767
0.0528 19.68 87500 0.0649 0.8775 0.8768
0.0525 20.24 90000 0.0651 0.8776 0.8769
0.0523 20.81 92500 0.0649 0.8775 0.8768
0.0517 21.37 95000 0.0648 0.8782 0.8775
0.0516 21.93 97500 0.0648 0.8783 0.8776
0.0511 22.49 100000 0.0648 0.8781 0.8774
0.0511 23.05 102500 0.0647 0.8783 0.8776
0.0508 23.62 105000 0.0647 0.8785 0.8778
0.0505 24.18 107500 0.0647 0.8785 0.8777
0.0505 24.74 110000 0.0646 0.8788 0.8781
0.0503 25.3 112500 0.0646 0.8786 0.8779
0.0502 25.87 115000 0.0646 0.8789 0.8782
0.0501 26.43 117500 0.0646 0.8788 0.8781
0.0501 26.99 120000 0.0645 0.8791 0.8784
0.05 27.55 122500 0.0646 0.8790 0.8783
0.0497 28.12 125000 0.0645 0.8792 0.8785
0.0499 28.68 127500 0.0645 0.8791 0.8784
0.0499 29.24 130000 0.0645 0.8792 0.8785
0.0497 29.8 132500 0.0645 0.8791 0.8784

Framework versions

  • Transformers 4.17.0
  • Pytorch 1.11.0+cu113
  • Datasets 2.3.2
  • Tokenizers 0.11.6

Model info

text-classification transformers
transformers pytorch tensorboard onnx roberta text-classification generated_from_trainer text-embeddings-inference endpoints_compatible region:us
↓ 4 ♥ 0

Created: 2022-08-11

Updated: 2022-08-12

View on Hugging Face

Files (20)

.gitattributes
.gitignore
README.md
all_results.json
categorization-finetuned-20220721-164940-distilled-20220811-013354-default.onnx ONNX
categorization-finetuned-20220721-164940-distilled-20220811-013354-int8.onnx ONNX
config.json
merges.txt
pytorch_model.bin
runs/Aug11_01-34-27_product-categorization/1660181758.762739/events.out.tfevents.1660181758.product-categorization.2923.1
runs/Aug11_01-34-27_product-categorization/events.out.tfevents.1660181758.product-categorization.2923.0
runs/Aug11_01-34-27_product-categorization/events.out.tfevents.1660201727.product-categorization.2923.2
special_tokens_map.json
test_results.json
tokenizer.json
tokenizer_config.json
train_results.json
trainer_state.json
training_args.bin
vocab.json

Related models

More from carted-nlp

More text-classification models

Browse all text-classification models

Popular models