README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - rotten_tomatoes
metrics:
  - accuracy
base_model: distilbert-base-cased
model-index:
  - name: rtm_DistilBERT_5E
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: rotten_tomatoes
          type: rotten_tomatoes
          config: default
          split: train
          args: default
        metrics:
          - type: accuracy
            value: 0.82
            name: Accuracy

rtm_DistilBERT_5E

This model is a fine-tuned version of distilbert-base-cased on the rotten_tomatoes dataset. It achieves the following results on the evaluation set:

• Loss: 0.6835
• Accuracy: 0.82

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: 1e-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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6822	0.09	50	0.6391	0.76
0.5531	0.19	100	0.4684	0.7667
0.4546	0.28	150	0.4479	0.7733
0.4495	0.37	200	0.3953	0.8067
0.4239	0.47	250	0.4211	0.7933
0.3951	0.56	300	0.4126	0.7933
0.3861	0.66	350	0.3950	0.7933
0.4108	0.75	400	0.4091	0.82
0.3778	0.84	450	0.4107	0.7933
0.3627	0.94	500	0.4203	0.7933
0.3648	1.03	550	0.4190	0.8
0.2899	1.12	600	0.4436	0.8
0.2637	1.22	650	0.4504	0.82
0.2885	1.31	700	0.4406	0.82
0.3226	1.4	750	0.4398	0.8333
0.3147	1.5	800	0.4239	0.82
0.2937	1.59	850	0.4227	0.8133
0.3149	1.69	900	0.3791	0.82
0.3227	1.78	950	0.3888	0.8133
0.2727	1.87	1000	0.4215	0.82
0.2722	1.97	1050	0.4099	0.8333
0.1908	2.06	1100	0.4595	0.82
0.2276	2.15	1150	0.4572	0.84
0.2239	2.25	1200	0.4545	0.8333
0.1986	2.34	1250	0.4895	0.82
0.2388	2.43	1300	0.4352	0.86
0.1901	2.53	1350	0.4806	0.84
0.2227	2.62	1400	0.5473	0.8067
0.2221	2.72	1450	0.5010	0.84
0.1955	2.81	1500	0.5315	0.8267
0.2114	2.9	1550	0.5410	0.8133
0.1827	3.0	1600	0.5721	0.8133
0.1527	3.09	1650	0.5616	0.8133
0.1464	3.18	1700	0.5935	0.8067
0.135	3.28	1750	0.6145	0.82
0.1668	3.37	1800	0.6901	0.8067
0.1702	3.46	1850	0.6067	0.8133
0.1738	3.56	1900	0.5981	0.82
0.1506	3.65	1950	0.6073	0.8267
0.1584	3.75	2000	0.6549	0.8133
0.1698	3.84	2050	0.6660	0.8267
0.1626	3.93	2100	0.6645	0.8267
0.1483	4.03	2150	0.6497	0.82
0.1342	4.12	2200	0.6643	0.82
0.1064	4.21	2250	0.6775	0.82
0.1302	4.31	2300	0.6876	0.82
0.1847	4.4	2350	0.6821	0.8133
0.1055	4.49	2400	0.6928	0.8133
0.1372	4.59	2450	0.6877	0.8133
0.131	4.68	2500	0.6769	0.8267
0.1242	4.78	2550	0.6769	0.8267
0.1289	4.87	2600	0.6810	0.82
0.1488	4.96	2650	0.6835	0.82

Framework versions

• Transformers 4.24.0
• Pytorch 1.13.0
• Datasets 2.7.1
• Tokenizers 0.13.2