depression_suggestion

This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.3740

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 64
• 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: 70

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	3	60.7965
No log	2.0	6	60.5778
No log	3.0	9	60.1954
No log	4.0	12	59.6487
No log	5.0	15	58.9372
No log	6.0	18	58.0582
No log	7.0	21	57.0106
No log	8.0	24	55.7910
No log	9.0	27	54.3934
No log	10.0	30	52.8099
No log	11.0	33	51.0219
No log	12.0	36	49.0127
No log	13.0	39	46.7522
No log	14.0	42	44.2033
No log	15.0	45	41.3146
No log	16.0	48	37.9982
No log	17.0	51	34.2236
No log	18.0	54	29.8068
No log	19.0	57	24.9750
No log	20.0	60	20.0707
No log	21.0	63	15.5166
No log	22.0	66	12.0328
No log	23.0	69	9.1012
No log	24.0	72	7.2116
No log	25.0	75	6.3149
No log	26.0	78	5.8127
No log	27.0	81	5.4548
No log	28.0	84	5.1684
No log	29.0	87	4.8927
No log	30.0	90	4.6128
No log	31.0	93	4.3782
No log	32.0	96	4.1996
No log	33.0	99	4.0981
No log	34.0	102	4.0022
No log	35.0	105	3.9224
No log	36.0	108	3.8381
No log	37.0	111	3.7660
No log	38.0	114	3.6887
No log	39.0	117	3.6483
No log	40.0	120	3.6020
No log	41.0	123	3.5590
No log	42.0	126	3.5199
No log	43.0	129	3.4646
No log	44.0	132	3.4098
No log	45.0	135	3.3684
No log	46.0	138	3.3290
No log	47.0	141	3.3113
No log	48.0	144	3.3033
No log	49.0	147	3.2928
No log	50.0	150	3.2776
No log	51.0	153	3.2587
No log	52.0	156	3.2487
No log	53.0	159	3.2390
No log	54.0	162	3.2318
No log	55.0	165	3.2311
No log	56.0	168	3.2377
No log	57.0	171	3.2554
No log	58.0	174	3.2720
No log	59.0	177	3.2781
No log	60.0	180	3.2882
No log	61.0	183	3.3089
No log	62.0	186	3.3352
No log	63.0	189	3.3519
No log	64.0	192	3.3233
No log	65.0	195	3.3028
No log	66.0	198	3.3153
No log	67.0	201	3.3422
No log	68.0	204	3.3753
No log	69.0	207	3.4003
No log	70.0	210	3.3740

Framework versions

• Transformers 4.19.2
• Pytorch 1.11.0+cu113
• Datasets 2.2.2
• Tokenizers 0.12.1