README.md

metadata

language:
  - en
license: mit
tags:
  - generated_from_trainer
  - deberta-v3
datasets:
  - nyu-mll/glue
metrics:
  - accuracy
  - f1
model-index:
  - name: deberta-v3-small
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: GLUE MRPC
          type: glue
          args: mrpc
        metrics:
          - type: accuracy
            value: 0.8921568627450981
            name: Accuracy
          - type: f1
            value: 0.9233449477351917
            name: F1
      - task:
          type: natural-language-inference
          name: Natural Language Inference
        dataset:
          name: glue
          type: glue
          config: mrpc
          split: validation
        metrics:
          - type: accuracy
            value: 0.8921568627450981
            name: Accuracy
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZmQ3MjM2NTJiZjJmM2UxYTlmMDczOTQ2MjY4OTdhZTAyM2RiMTc2YjZiNWIwZDk1ZGUxMjgzMDBiOWVjZTQ4OCIsInZlcnNpb24iOjF9.yerN7Izy0yT3ykyO3t5Mr-TO3oxpTMfijCWJKnA_XO_rt81LP3-9qbqknXur6ahHqKN-1BLtr_fmAu0-IPQyDA
          - type: precision
            value: 0.8983050847457628
            name: Precision
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZWQxODVmYTM4OThlMjNhY2MzZTBhMWJmMmNjMDMyYjYyNzc4NWI3YzJjZDkzMTcyOWEwN2IxOWYyOGQ5NTY5MSIsInZlcnNpb24iOjF9.cfqvd8wnSqhHj5fKlIb6JN9He8ooAu94tFJytw2I93qqGSVvaTktM0Ib_DqPuHYneGY1DGbgb6Nsl90DiZSMCQ
          - type: recall
            value: 0.9498207885304659
            name: Recall
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNjg3Y2Y1NGY0NTRjMWFhYTAxMWYxMTcxNWM2ZDU5NGY1ZTk3OTJmZWQyYmIzMGJiZWQ0YWQ2MjNhOGU2MGU0ZCIsInZlcnNpb24iOjF9.jj7VNaWQU3u3tnngqCixlfkwF8h6ykzvHm4tgezJe1pacAU0Tsugn7IPvAJTrvNE0sU8_Q7dm-C_UKQGzmlIBw
          - type: auc
            value: 0.9516129032258065
            name: AUC
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMDhkOTQ0ZmVlYTYwNTdjY2IxYTM5ZThhYzgzZWMxMGQzMThmZDkwNTcyMWZiNzg4Y2I3NjZhMzVjYmNmN2FlZiIsInZlcnNpb24iOjF9.28hOJFgnyNHXMpaFbNTEcolUcuNVqrXNSuT6hTs2vrjlAIWVnzxUfaHjH2kVYh1-sOSNSE9maetd1CtQ7i78CQ
          - type: f1
            value: 0.9233449477351917
            name: F1
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZmY0ZWE5Y2Q5YmZlOWM4OTU0OGIwOWEwNDk3MTlkYTY5YzgwMjQwNDFjYWU4ZDdmZWY4Nzc0MzQzMTM2YTRhYyIsInZlcnNpb24iOjF9.NymiR2fVXaI6ytAGZFM8HuQLxTJlxuUsWziVNaauyuJ9xfOLOGVJ6VI_H7CoBwc-pZKbKiQOvtfpOGwt1J22CA
          - type: loss
            value: 0.2787226438522339
            name: loss
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZGNhMDgyMGI3ZWI4NDVkYzM0NjE1ZTk0YjczYzU4NmRhOGYxM2RlMjU3YThhY2QzNmU3NmJhM2IzMWI5MDMwNyIsInZlcnNpb24iOjF9.HFdpBkvu0671KUgkOtpSgeGBr3wU7g51zVt3-wEwVWhS4hMX4oPFAqF4JBxFx3mgbGjTDiRQ2xiA5lm0UnkdCg

DeBERTa v3 (small) fine-tuned on MRPC

This model is a fine-tuned version of microsoft/deberta-v3-small on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

• Loss: 0.2787
• Accuracy: 0.8922
• F1: 0.9233
• Combined Score: 0.9078

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
No log	1.0	230	0.2787	0.8922	0.9233	0.9078
No log	2.0	460	0.3651	0.875	0.9137	0.8944
No log	3.0	690	0.5238	0.8799	0.9179	0.8989
No log	4.0	920	0.4712	0.8946	0.9222	0.9084
0.2147	5.0	1150	0.5704	0.8946	0.9262	0.9104
0.2147	6.0	1380	0.5697	0.8995	0.9284	0.9140
0.2147	7.0	1610	0.6651	0.8922	0.9214	0.9068
0.2147	8.0	1840	0.6726	0.8946	0.9239	0.9093
0.0183	9.0	2070	0.7250	0.8848	0.9177	0.9012
0.0183	10.0	2300	0.7093	0.8922	0.9223	0.9072

Framework versions

• Transformers 4.13.0.dev0
• Pytorch 1.10.0+cu111
• Datasets 1.15.1
• Tokenizers 0.10.3