DrishtiSharma commited on
Commit
72016f6
1 Parent(s): 11c57c1

End of training

Browse files
README.md ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - patent-classification
8
+ metrics:
9
+ - accuracy
10
+ base_model: roberta-large
11
+ model-index:
12
+ - name: roberta-large-flash-attention-2-lora-patent-classification
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # roberta-large-flash-attention-2-lora-patent-classification
20
+
21
+ This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the patent-classification dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.8395
24
+ - Accuracy: 0.6304
25
+ - Precision Macro: 0.6136
26
+ - Recall Macro: 0.5995
27
+ - F1-score Macro: 0.5984
28
+
29
+ ## Model description
30
+
31
+ More information needed
32
+
33
+ ## Intended uses & limitations
34
+
35
+ More information needed
36
+
37
+ ## Training and evaluation data
38
+
39
+ More information needed
40
+
41
+ ## Training procedure
42
+
43
+ ### Training hyperparameters
44
+
45
+ The following hyperparameters were used during training:
46
+ - learning_rate: 2e-05
47
+ - train_batch_size: 6
48
+ - eval_batch_size: 4
49
+ - seed: 42
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: cosine
52
+ - num_epochs: 3
53
+ - mixed_precision_training: Native AMP
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision Macro | Recall Macro | F1-score Macro |
58
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------------:|:------------:|:--------------:|
59
+ | 0.7566 | 1.0 | 4167 | 0.9131 | 0.5692 | 0.6231 | 0.5423 | 0.5631 |
60
+ | 0.6974 | 2.0 | 8334 | 0.8428 | 0.6174 | 0.6169 | 0.5910 | 0.5942 |
61
+ | 0.7219 | 3.0 | 12501 | 0.8395 | 0.6304 | 0.6136 | 0.5995 | 0.5984 |
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - PEFT 0.7.2.dev0
67
+ - Transformers 4.37.0.dev0
68
+ - Pytorch 2.1.0+cu121
69
+ - Datasets 2.16.2.dev0
70
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "roberta-large",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.1,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 16,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "query",
23
+ "value"
24
+ ],
25
+ "task_type": "SEQ_CLS",
26
+ "use_rslora": false
27
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20b0336c319bc589e9462f42d5c9ef4da8aa99e5fc999024eca138990e03c993
3
+ size 10541212
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46338800b6b20cd70bd1a903b363f6e7286e6e5f6f9182eae7e1a7b80099155d
3
+ size 4792