You can deploy this model and start earning money today!

contemmcm/
b3032376a7ffa2e52661f77aee6cdb9d

Views

No views yet

b3032376a7ffa2e52661f77aee6cdb9d

This model is a fine-tuned version of google-bert/bert-large-cased-whole-word-masking-finetuned-squad on the dim/tldr_news dataset. It achieves the following results on the evaluation set:

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
num_epochs: 50

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rougel	Rougelsum
No log	0	0	1.9398	0	1.6219	0.0185	0.0073	0.0185	0.0185	0.0185
No log	1	178	1.4959	0.0078	3.1749	0.2820	0.0983	0.2812	0.2812	0.2820
No log	2	356	1.4121	0.0156	2.3455	0.2599	0.1028	0.2599	0.2599	0.2599
No log	3	534	1.1108	0.0312	3.0322	0.5241	0.3827	0.5249	0.5249	0.5234
No log	4	712	0.7557	0.0625	4.0711	0.7251	0.5748	0.7259	0.7259	0.7251
No log	5	890	0.7688	0.125	5.7070	0.7166	0.5465	0.7180	0.7173	0.7166
0.0567	6	1068	0.7546	0.25	8.0032	0.7195	0.5788	0.7202	0.7202	0.7202
0.5871	7	1246	0.6598	0.5	11.7694	0.7564	0.7337	0.7564	0.7571	0.7564
0.4853	8.0	1424	0.6008	1.0	20.7246	0.7827	0.7965	0.7834	0.7834	0.7827
0.335	9.0	1602	0.8023	1.0	20.5593	0.7543	0.7325	0.7550	0.7546	0.7550
0.2045	10.0	1780	1.0249	1.0	19.5754	0.7564	0.7868	0.7578	0.7564	0.7557
0.1174	11.0	1958	1.3285	1.0	19.7835	0.7472	0.7830	0.7479	0.7479	0.7472
0.1268	12.0	2136	1.2440	1.0	19.6191	0.7642	0.7929	0.7642	0.7642	0.7642