SChem5Labels-google-t5-v1_1-large-inter_model-frequency-model_annots_str

This model is a fine-tuned version of google/t5-v1_1-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9844

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: 0.0001
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss
20.3817	1.0	25	23.8120
19.4374	2.0	50	21.8918
18.745	3.0	75	19.3959
16.896	4.0	100	15.4970
15.3045	5.0	125	11.5374
12.9955	6.0	150	9.6467
11.2112	7.0	175	8.9925
9.4851	8.0	200	8.7994
8.7487	9.0	225	8.5320
8.1197	10.0	250	8.3570
7.9164	11.0	275	8.2662
7.7789	12.0	300	8.1800
7.6671	13.0	325	8.0987
7.5107	14.0	350	7.9659
7.457	15.0	375	7.6850
7.1712	16.0	400	7.3914
6.9462	17.0	425	7.2019
6.8109	18.0	450	7.0657
6.7403	19.0	475	6.9778
6.5766	20.0	500	6.9288
6.505	21.0	525	6.8702
6.5148	22.0	550	6.8175
6.541	23.0	575	6.7619
4.3898	24.0	600	1.0917
1.0874	25.0	625	0.7681
0.8058	26.0	650	0.7295
0.7847	27.0	675	0.7244
0.7779	28.0	700	0.7195
0.7741	29.0	725	0.7205
0.7606	30.0	750	0.7222
0.7613	31.0	775	0.7189
0.7676	32.0	800	0.7119
0.7547	33.0	825	0.7138
0.7433	34.0	850	0.7148
0.7729	35.0	875	0.7202

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.6.1
• Tokenizers 0.14.1