results_t5small

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

Loss: 0.0050

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: 0.005
train_batch_size: 48
eval_batch_size: 48
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.0235	0.12	500	0.0144
0.0151	0.25	1000	0.0119
0.0153	0.37	1500	0.0101
0.0173	0.5	2000	0.0109
0.0132	0.62	2500	0.0100
0.0123	0.75	3000	0.0092
0.0136	0.87	3500	0.0110
0.014	0.99	4000	0.0089
0.0123	1.12	4500	0.0094
0.0155	1.24	5000	0.0088
0.0119	1.37	5500	0.0095
0.0126	1.49	6000	0.0089
0.0111	1.61	6500	0.0083
0.0129	1.74	7000	0.0081
0.0115	1.86	7500	0.0087
0.0133	1.99	8000	0.0085
0.0167	2.11	8500	0.0105
0.0113	2.24	9000	0.0088
0.0106	2.36	9500	0.0077
0.0101	2.48	10000	0.0075
0.0086	2.61	10500	0.0076
0.0112	2.73	11000	0.0075
0.0097	2.86	11500	0.0076
0.0093	2.98	12000	0.0078
0.0086	3.1	12500	0.0076
0.0092	3.23	13000	0.0077
0.0089	3.35	13500	0.0074
0.0086	3.48	14000	0.0072
0.0084	3.6	14500	0.0072
0.0083	3.73	15000	0.0073
0.0076	3.85	15500	0.0072
0.0097	3.97	16000	0.0071
0.0074	4.1	16500	0.0071
0.0074	4.22	17000	0.0070
0.0089	4.35	17500	0.0069
0.0072	4.47	18000	0.0069
0.0072	4.6	18500	0.0068
0.0084	4.72	19000	0.0067
0.0076	4.84	19500	0.0072
0.0084	4.97	20000	0.0067
0.0062	5.09	20500	0.0065
0.007	5.22	21000	0.0064
0.0075	5.34	21500	0.0063
0.0067	5.46	22000	0.0064
0.0063	5.59	22500	0.0065
0.007	5.71	23000	0.0062
0.0064	5.84	23500	0.0062
0.0073	5.96	24000	0.0062
0.0064	6.09	24500	0.0063
0.0059	6.21	25000	0.0060
0.0062	6.33	25500	0.0060
0.0057	6.46	26000	0.0060
0.006	6.58	26500	0.0058
0.0053	6.71	27000	0.0058
0.0064	6.83	27500	0.0057
0.0061	6.95	28000	0.0057
0.0055	7.08	28500	0.0057
0.0065	7.2	29000	0.0057
0.0058	7.33	29500	0.0057
0.0052	7.45	30000	0.0055
0.0058	7.58	30500	0.0056
0.0055	7.7	31000	0.0055
0.0049	7.82	31500	0.0054
0.0053	7.95	32000	0.0053
0.005	8.07	32500	0.0054
0.0051	8.2	33000	0.0053
0.0047	8.32	33500	0.0053
0.0057	8.45	34000	0.0052
0.0049	8.57	34500	0.0053
0.0051	8.69	35000	0.0052
0.0047	8.82	35500	0.0052
0.0051	8.94	36000	0.0051
0.0044	9.07	36500	0.0052
0.0042	9.19	37000	0.0052
0.0045	9.31	37500	0.0051
0.0043	9.44	38000	0.0051
0.0041	9.56	38500	0.0051
0.0042	9.69	39000	0.0051
0.004	9.81	39500	0.0050
0.0038	9.94	40000	0.0050

Framework versions

Transformers 4.37.1
Pytorch 2.1.2
Datasets 2.16.1
Tokenizers 0.15.1

加载中…

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.005

train_batch_size: 48

eval_batch_size: 48

seed: 42

optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

lr_scheduler_type: linear

lr_scheduler_warmup_steps: 500

num_epochs: 10

mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.0235	0.12	500	0.0144
0.0151	0.25	1000	0.0119
0.0153	0.37	1500	0.0101
0.0173	0.5	2000	0.0109
0.0132	0.62	2500	0.0100
0.0123	0.75	3000	0.0092
0.0136	0.87	3500	0.0110
0.014	0.99	4000	0.0089
0.0123	1.12	4500	0.0094
0.0155	1.24	5000	0.0088
0.0119	1.37	5500	0.0095
0.0126	1.49	6000	0.0089
0.0111	1.61	6500	0.0083
0.0129	1.74	7000	0.0081
0.0115	1.86	7500	0.0087
0.0133	1.99	8000	0.0085
0.0167	2.11	8500	0.0105
0.0113	2.24	9000	0.0088
0.0106	2.36	9500	0.0077
0.0101	2.48	10000	0.0075
0.0086	2.61	10500	0.0076
0.0112	2.73	11000	0.0075
0.0097	2.86	11500	0.0076
0.0093	2.98	12000	0.0078
0.0086	3.1	12500	0.0076
0.0092	3.23	13000	0.0077
0.0089	3.35	13500	0.0074
0.0086	3.48	14000	0.0072
0.0084	3.6	14500	0.0072
0.0083	3.73	15000	0.0073
0.0076	3.85	15500	0.0072
0.0097	3.97	16000	0.0071
0.0074	4.1	16500	0.0071
0.0074	4.22	17000	0.0070
0.0089	4.35	17500	0.0069
0.0072	4.47	18000	0.0069
0.0072	4.6	18500	0.0068
0.0084	4.72	19000	0.0067
0.0076	4.84	19500	0.0072
0.0084	4.97	20000	0.0067
0.0062	5.09	20500	0.0065
0.007	5.22	21000	0.0064
0.0075	5.34	21500	0.0063
0.0067	5.46	22000	0.0064
0.0063	5.59	22500	0.0065
0.007	5.71	23000	0.0062
0.0064	5.84	23500	0.0062
0.0073	5.96	24000	0.0062
0.0064	6.09	24500	0.0063
0.0059	6.21	25000	0.0060
0.0062	6.33	25500	0.0060
0.0057	6.46	26000	0.0060
0.006	6.58	26500	0.0058
0.0053	6.71	27000	0.0058
0.0064	6.83	27500	0.0057
0.0061	6.95	28000	0.0057
0.0055	7.08	28500	0.0057
0.0065	7.2	29000	0.0057
0.0058	7.33	29500	0.0057
0.0052	7.45	30000	0.0055
0.0058	7.58	30500	0.0056
0.0055	7.7	31000	0.0055
0.0049	7.82	31500	0.0054
0.0053	7.95	32000	0.0053
0.005	8.07	32500	0.0054
0.0051	8.2	33000	0.0053
0.0047	8.32	33500	0.0053
0.0057	8.45	34000	0.0052
0.0049	8.57	34500	0.0053
0.0051	8.69	35000	0.0052
0.0047	8.82	35500	0.0052
0.0051	8.94	36000	0.0051
0.0044	9.07	36500	0.0052
0.0042	9.19	37000	0.0052
0.0045	9.31	37500	0.0051
0.0043	9.44	38000	0.0051
0.0041	9.56	38500	0.0051
0.0042	9.69	39000	0.0051
0.004	9.81	39500	0.0050
0.0038	9.94	40000	0.0050

Framework versions

Transformers 4.37.1

Pytorch 2.1.2

Datasets 2.16.1

Tokenizers 0.15.1

spidersouris/genre-t5-small

results_t5small

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

results_t5small

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions