meghanaraok
commited on
Commit
•
638a7ef
1
Parent(s):
29c9017
Upload 2 files
Browse files- config.json +60 -0
- model.safetensors +3 -0
config.json
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model_name_or_path": "pretrained/ClinicalplusXLNet/",
|
3 |
+
"tokenizer_name": "xlnet-base-cased",
|
4 |
+
"output_dir": "../model/",
|
5 |
+
"overwrite_output_dir": true,
|
6 |
+
|
7 |
+
"task_name": "mimic3-50",
|
8 |
+
"max_seq_length": 512,
|
9 |
+
"train_file": "../data/mimic3/5/train_data_5_level_1_new.csv",
|
10 |
+
"validation_file": "../data/mimic3/5/dev_data_5_level_1_new.csv",
|
11 |
+
"test_file": "../data/mimic3/5/test_data_5_level_1_new.csv",
|
12 |
+
"label_dictionary_file": "../data/mimic3/5/labels_dictionary_5_level_1.csv",
|
13 |
+
"ignore_keys_for_eval": ["preds", "label_attention_weights", "chunk_attention_weights"],
|
14 |
+
"use_cached_datasets": true,
|
15 |
+
"data_segmented": true,
|
16 |
+
"lazy_loading": false,
|
17 |
+
|
18 |
+
"d_model": 768,
|
19 |
+
"dropout": 0.1,
|
20 |
+
"dropout_att": 0.1,
|
21 |
+
"num_chunks_per_document": 8,
|
22 |
+
"transformer_layer_update_strategy": "all",
|
23 |
+
"use_code_representation": false,
|
24 |
+
"multi_head_attention": false,
|
25 |
+
"chunk_attention": true,
|
26 |
+
"multi_head_chunk_attention": false,
|
27 |
+
"linear_init_mean": 0.0,
|
28 |
+
"linear_init_std": 0.03,
|
29 |
+
"document_pooling_strategy": "mean",
|
30 |
+
|
31 |
+
"do_train": true,
|
32 |
+
"do_eval": true,
|
33 |
+
"do_predict": true,
|
34 |
+
"evaluation_strategy": "steps",
|
35 |
+
"per_device_train_batch_size": 2,
|
36 |
+
"per_device_eval_batch_size": 2,
|
37 |
+
"gradient_accumulation_steps": 1,
|
38 |
+
"learning_rate": 5e-5,
|
39 |
+
"weight_decay": 0.1,
|
40 |
+
"num_train_epochs": 100,
|
41 |
+
"max_steps": 10500,
|
42 |
+
"warmup_steps": 500,
|
43 |
+
"log_level": "error",
|
44 |
+
"logging_strategy": "steps",
|
45 |
+
"logging_steps": 500,
|
46 |
+
"save_strategy": "steps",
|
47 |
+
"save_steps": 500,
|
48 |
+
"seed": 2022,
|
49 |
+
"dataloader_drop_last": false,
|
50 |
+
"disable_tqdm": false,
|
51 |
+
"label_names": ["targets"],
|
52 |
+
"load_best_model_at_end": true,
|
53 |
+
"metric_for_best_model": "micro_f1",
|
54 |
+
"greater_is_better": true,
|
55 |
+
"remove_unused_columns": false,
|
56 |
+
"report_to": "none",
|
57 |
+
"num_hidden_layers": 2,
|
58 |
+
"resume_from_checkpoint": "C:/Users/mkanneganti/Downloads/new/model/checkpoint-10500- Long HiLAT 05"
|
59 |
+
|
60 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62227d686b620b51b91edf0a6dd1a51f14df1fd35e7290024c2b5affd9c53296
|
3 |
+
size 471649956
|