Model save
Browse files
README.md
ADDED
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
base_model: Qwen/Qwen1.5-4B
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: lmind_nq_train6000_eval6489_v1_doc_qa_v3_Qwen_Qwen1.5-4B_5e-4_lora2
|
10 |
+
results: []
|
11 |
+
library_name: peft
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# lmind_nq_train6000_eval6489_v1_doc_qa_v3_Qwen_Qwen1.5-4B_5e-4_lora2
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on an unknown dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 2.2417
|
22 |
+
- Accuracy: 0.5610
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 0.0005
|
42 |
+
- train_batch_size: 1
|
43 |
+
- eval_batch_size: 2
|
44 |
+
- seed: 42
|
45 |
+
- distributed_type: multi-GPU
|
46 |
+
- num_devices: 4
|
47 |
+
- gradient_accumulation_steps: 8
|
48 |
+
- total_train_batch_size: 32
|
49 |
+
- total_eval_batch_size: 8
|
50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
+
- lr_scheduler_type: constant
|
52 |
+
- lr_scheduler_warmup_ratio: 0.05
|
53 |
+
- num_epochs: 20.0
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
58 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
59 |
+
| 1.8583 | 1.0 | 529 | 1.6376 | 0.5726 |
|
60 |
+
| 1.6329 | 2.0 | 1058 | 1.6881 | 0.5713 |
|
61 |
+
| 1.3464 | 3.0 | 1587 | 1.8256 | 0.5663 |
|
62 |
+
| 1.1624 | 4.0 | 2116 | 1.9223 | 0.5652 |
|
63 |
+
| 0.964 | 5.0 | 2645 | 1.9720 | 0.5643 |
|
64 |
+
| 0.8117 | 6.0 | 3174 | 2.0016 | 0.5647 |
|
65 |
+
| 0.7242 | 7.0 | 3703 | 2.0785 | 0.5639 |
|
66 |
+
| 0.6381 | 8.0 | 4232 | 2.0954 | 0.5645 |
|
67 |
+
| 0.573 | 9.0 | 4761 | 2.1067 | 0.5623 |
|
68 |
+
| 0.5269 | 10.0 | 5290 | 2.1356 | 0.5646 |
|
69 |
+
| 0.5144 | 11.0 | 5819 | 2.1951 | 0.5616 |
|
70 |
+
| 0.4887 | 12.0 | 6348 | 2.1779 | 0.5631 |
|
71 |
+
| 0.4636 | 13.0 | 6877 | 2.1757 | 0.5611 |
|
72 |
+
| 0.467 | 14.0 | 7406 | 2.1781 | 0.5624 |
|
73 |
+
| 0.4613 | 15.0 | 7935 | 2.2312 | 0.5612 |
|
74 |
+
| 0.4405 | 16.0 | 8464 | 2.1800 | 0.5629 |
|
75 |
+
| 0.4308 | 17.0 | 8993 | 2.1960 | 0.5628 |
|
76 |
+
| 0.4401 | 18.0 | 9522 | 2.2355 | 0.5610 |
|
77 |
+
| 0.4334 | 19.0 | 10051 | 2.2380 | 0.5608 |
|
78 |
+
| 0.4218 | 20.0 | 10580 | 2.2417 | 0.5610 |
|
79 |
+
|
80 |
+
|
81 |
+
### Framework versions
|
82 |
+
|
83 |
+
- PEFT 0.5.0
|
84 |
+
- Transformers 4.40.2
|
85 |
+
- Pytorch 2.3.0
|
86 |
+
- Datasets 2.19.1
|
87 |
+
- Tokenizers 0.19.1
|