left_padding0model

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

• Accuracy: 0.9248
• Loss: 0.6918

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: 10

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
0.2227	1.0	1563	0.9217	0.2214
0.1325	2.0	3126	0.9312	0.2340
0.1433	3.0	4689	0.9273	0.2143
0.1093	4.0	6252	0.9267	0.3209
0.0601	5.0	7815	0.9276	0.3734
0.0497	6.0	9378	0.9174	0.4883
0.0632	7.0	10941	0.9169	0.4722
0.0301	8.0	12504	0.9048	0.5964
0.0292	9.0	14067	0.9261	0.4406
0.0119	10.0	15630	0.9264	0.5227
0.0218	11.0	17193	0.9294	0.5665
0.0161	12.0	18756	0.9276	0.5829
0.0068	13.0	20319	0.928	0.5820
0.0265	14.0	21882	0.9229	0.5842
0.0098	15.0	23445	0.9283	0.6034
0.0081	16.0	25008	0.9251	0.6624
0.0062	17.0	26571	0.9138	0.5561
0.0153	18.0	28134	0.9223	0.6722
0.0213	19.0	29697	0.9233	0.6735
0.0148	20.0	31260	0.9283	0.5918
0.0076	21.0	32823	0.9248	0.7200
0.0088	22.0	34386	0.9221	0.6554
0.0072	23.0	35949	0.9248	0.6918

Framework versions

• Transformers 4.35.0
• Pytorch 2.0.0+cu117
• Datasets 2.14.6
• Tokenizers 0.14.1