Hugging Face's logo

Bencode92
/
tradepulse-finbert-sentiment

Text Classification
TensorBoard
Safetensors
English
bert
finance
sentiment-analysis
finbert
trading
Model card Files
xet
Metrics Community
Bencode92 commited on
Commit
295bf3e
·
1 Parent(s): 8c5fac6

🔄 Incremental label | Acc: 0.643, F1: 0.612

Browse files
Files changed (9) hide show
  1. README.md +8 -8
  2. {checkpoint-24 → checkpoint-48}/config.json +0 -0
  3. {checkpoint-24 → checkpoint-48}/model.safetensors +1 -1
  4. {checkpoint-24 → checkpoint-48}/special_tokens_map.json +0 -0
  5. {checkpoint-24 → checkpoint-48}/tokenizer.json +0 -0
  6. {checkpoint-24 → checkpoint-48}/tokenizer_config.json +0 -0
  7. {checkpoint-24 → checkpoint-48}/trainer_state.json +50 -21
  8. {checkpoint-24 → checkpoint-48}/vocab.txt +0 -0
  9. model.safetensors +1 -1
README.md CHANGED
@@ -21,19 +21,19 @@ Fine-tuned FinBERT model for financial sentiment analysis in TradePulse.
21
 
22
  ## Performance
23
 
24
- *Last training: 2025-07-24 21:18*
25
  *Dataset: `news_20250724.csv` (125 samples)*
26
 
27
  | Metric | Value |
28
  |--------|-------|
29
- | Loss | 1.0782 |
30
- | Accuracy | 0.7500 |
31
- | F1 Score | 0.7473 |
32
 
33
- | F1 Macro | 0.7473 |
34
 
35
- | Precision | 0.7618 |
36
- | Recall | 0.7500 |
37
 
38
  ## Training Details
39
 
@@ -65,4 +65,4 @@ predictions = outputs.logits.softmax(dim=-1)
65
  ## Model Card Authors
66
 
67
  - TradePulse ML Team
68
- - Auto-generated on 2025-07-24 21:18:33
 
21
 
22
  ## Performance
23
 
24
+ *Last training: 2025-07-24 22:19*
25
  *Dataset: `news_20250724.csv` (125 samples)*
26
 
27
  | Metric | Value |
28
  |--------|-------|
29
+ | Loss | 1.0535 |
30
+ | Accuracy | 0.6875 |
31
+ | F1 Score | 0.6935 |
32
 
33
+ | F1 Macro | 0.6935 |
34
 
35
+ | Precision | 0.7055 |
36
+ | Recall | 0.6875 |
37
 
38
  ## Training Details
39
 
 
65
  ## Model Card Authors
66
 
67
  - TradePulse ML Team
68
+ - Auto-generated on 2025-07-24 22:19:24
{checkpoint-24 → checkpoint-48}/config.json RENAMED
File without changes
{checkpoint-24 → checkpoint-48}/model.safetensors RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4cdbcb8ed6aa8aad2daf6ad1c9a844eac8def03b4d00f6f22449004ff204fdf7
3
  size 439039996
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61ee365bf408a3d23544228dcf9a097de22b17658c4294f8dfee7d83144a79b0
3
  size 439039996
{checkpoint-24 → checkpoint-48}/special_tokens_map.json RENAMED
File without changes
{checkpoint-24 → checkpoint-48}/tokenizer.json RENAMED
File without changes
{checkpoint-24 → checkpoint-48}/tokenizer_config.json RENAMED
File without changes
{checkpoint-24 → checkpoint-48}/trainer_state.json RENAMED
@@ -1,41 +1,70 @@
1
  {
2
- "best_metric": 0.7473118279569892,
3
- "best_model_checkpoint": "hf-sentiment-production/checkpoint-24",
4
- "epoch": 1.0,
5
  "eval_steps": 500,
6
- "global_step": 24,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.4166666666666667,
13
- "grad_norm": 3.0344228744506836,
14
  "learning_rate": 4.347826086956522e-06,
15
- "loss": 0.2894,
16
  "step": 10
17
  },
18
  {
19
  "epoch": 0.8333333333333334,
20
- "grad_norm": 1.2045588493347168,
21
  "learning_rate": 8.695652173913044e-06,
22
- "loss": 0.2505,
23
  "step": 20
24
  },
25
  {
26
  "epoch": 1.0,
27
- "eval_accuracy": 0.75,
28
- "eval_f1": 0.7473118279569892,
29
- "eval_f1_macro": 0.7240143369175627,
30
- "eval_loss": 1.0782057046890259,
31
- "eval_precision": 0.7618117559523809,
32
- "eval_precision_macro": 0.7205687830687831,
33
- "eval_recall": 0.75,
34
- "eval_recall_macro": 0.75,
35
- "eval_runtime": 17.8529,
36
- "eval_samples_per_second": 1.792,
37
- "eval_steps_per_second": 0.448,
38
  "step": 24
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39
  }
40
  ],
41
  "logging_steps": 10,
@@ -59,12 +88,12 @@
59
  "should_evaluate": false,
60
  "should_log": false,
61
  "should_save": true,
62
- "should_training_stop": false
63
  },
64
  "attributes": {}
65
  }
66
  },
67
- "total_flos": 24469547848704.0,
68
  "train_batch_size": 4,
69
  "trial_name": null,
70
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.6935308441558441,
3
+ "best_model_checkpoint": "hf-sentiment-production/checkpoint-48",
4
+ "epoch": 2.0,
5
  "eval_steps": 500,
6
+ "global_step": 48,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.4166666666666667,
13
+ "grad_norm": 0.17290420830249786,
14
  "learning_rate": 4.347826086956522e-06,
15
+ "loss": 0.0432,
16
  "step": 10
17
  },
18
  {
19
  "epoch": 0.8333333333333334,
20
+ "grad_norm": 0.24503789842128754,
21
  "learning_rate": 8.695652173913044e-06,
22
+ "loss": 0.0711,
23
  "step": 20
24
  },
25
  {
26
  "epoch": 1.0,
27
+ "eval_accuracy": 0.65625,
28
+ "eval_f1": 0.6526988636363636,
29
+ "eval_f1_macro": 0.5954545454545455,
30
+ "eval_loss": 1.0106450319290161,
31
+ "eval_precision": 0.6558823529411765,
32
+ "eval_precision_macro": 0.6019607843137256,
33
+ "eval_recall": 0.65625,
34
+ "eval_recall_macro": 0.5944444444444444,
35
+ "eval_runtime": 17.7449,
36
+ "eval_samples_per_second": 1.803,
37
+ "eval_steps_per_second": 0.451,
38
  "step": 24
39
+ },
40
+ {
41
+ "epoch": 1.25,
42
+ "grad_norm": 3.8800034523010254,
43
+ "learning_rate": 7.2000000000000005e-06,
44
+ "loss": 0.0533,
45
+ "step": 30
46
+ },
47
+ {
48
+ "epoch": 1.6666666666666665,
49
+ "grad_norm": 1.5228019952774048,
50
+ "learning_rate": 3.2000000000000003e-06,
51
+ "loss": 0.0525,
52
+ "step": 40
53
+ },
54
+ {
55
+ "epoch": 2.0,
56
+ "eval_accuracy": 0.6875,
57
+ "eval_f1": 0.6935308441558441,
58
+ "eval_f1_macro": 0.657056277056277,
59
+ "eval_loss": 1.0534968376159668,
60
+ "eval_precision": 0.7055288461538461,
61
+ "eval_precision_macro": 0.6538461538461539,
62
+ "eval_recall": 0.6875,
63
+ "eval_recall_macro": 0.6666666666666666,
64
+ "eval_runtime": 17.653,
65
+ "eval_samples_per_second": 1.813,
66
+ "eval_steps_per_second": 0.453,
67
+ "step": 48
68
  }
69
  ],
70
  "logging_steps": 10,
 
88
  "should_evaluate": false,
89
  "should_log": false,
90
  "should_save": true,
91
+ "should_training_stop": true
92
  },
93
  "attributes": {}
94
  }
95
  },
96
+ "total_flos": 48939095697408.0,
97
  "train_batch_size": 4,
98
  "trial_name": null,
99
  "trial_params": null
{checkpoint-24 → checkpoint-48}/vocab.txt RENAMED
File without changes
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4cdbcb8ed6aa8aad2daf6ad1c9a844eac8def03b4d00f6f22449004ff204fdf7
3
  size 439039996
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61ee365bf408a3d23544228dcf9a097de22b17658c4294f8dfee7d83144a79b0
3
  size 439039996