Bencode92 commited on
Commit
8d09bbe
ยท
1 Parent(s): df1d07e

๐Ÿ”„ Incremental label | Acc: 0.714, F1: 0.519

Browse files
.gitignore ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # TradePulse ML - Artefacts ร  ignorer
2
+ logs/
3
+ *.pt
4
+ *.pth
5
+ *.bin
6
+ checkpoints/
7
+ runs/
8
+ wandb/
9
+ *.log
10
+ __pycache__/
11
+ *.pyc
12
+ .DS_Store
13
+ tmp_eval/
README.md CHANGED
@@ -1,71 +1,62 @@
1
  ---
2
- base_model: yiyanghkust/finbert-tone
 
3
  tags:
4
- - generated_from_trainer
5
- metrics:
6
- - accuracy
7
- - precision
8
- - recall
9
- - f1
10
- model-index:
11
- - name: tradepulse-finbert-sentiment
12
- results: []
13
  ---
14
 
15
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
- should probably proofread and complete it, then remove this comment. -->
17
 
18
- # tradepulse-finbert-sentiment
 
19
 
20
- This model is a fine-tuned version of [yiyanghkust/finbert-tone](https://huggingface.co/yiyanghkust/finbert-tone) on the None dataset.
21
- It achieves the following results on the evaluation set:
22
- - Loss: 2.7144
23
- - Accuracy: 0.3846
24
- - Precision: 0.4462
25
- - Recall: 0.3846
26
- - F1: 0.4126
27
- - F1 Macro: 0.3258
28
- - Precision Macro: 0.3500
29
- - Recall Macro: 0.3056
30
 
31
- ## Model description
32
 
33
- More information needed
 
34
 
35
- ## Intended uses & limitations
 
 
 
 
 
 
 
36
 
37
- More information needed
38
 
39
- ## Training and evaluation data
 
 
 
 
 
40
 
41
- More information needed
42
 
43
- ## Training procedure
 
44
 
45
- ### Training hyperparameters
 
46
 
47
- The following hyperparameters were used during training:
48
- - learning_rate: 2e-05
49
- - train_batch_size: 8
50
- - eval_batch_size: 8
51
- - seed: 42
52
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
- - lr_scheduler_type: linear
54
- - lr_scheduler_warmup_steps: 12
55
- - num_epochs: 3
56
 
57
- ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | F1 Macro | Precision Macro | Recall Macro |
60
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:--------:|:---------------:|:------------:|
61
- | No log | 1.0 | 7 | 5.3820 | 0.3077 | 0.3077 | 0.3077 | 0.3077 | 0.2222 | 0.2222 | 0.2222 |
62
- | 4.744 | 2.0 | 14 | 3.3730 | 0.3846 | 0.4462 | 0.3846 | 0.4126 | 0.3258 | 0.3500 | 0.3056 |
63
- | 1.0682 | 3.0 | 21 | 2.7144 | 0.3846 | 0.4462 | 0.3846 | 0.4126 | 0.3258 | 0.3500 | 0.3056 |
64
-
65
-
66
- ### Framework versions
67
-
68
- - Transformers 4.41.0
69
- - Pytorch 2.7.1+cpu
70
- - Datasets 2.19.1
71
- - Tokenizers 0.19.1
 
1
  ---
2
+ language: en
3
+ license: apache-2.0
4
  tags:
5
+ - finance
6
+ - sentiment-analysis
7
+ - finbert
8
+ - trading
9
+ pipeline_tag: text-classification
 
 
 
 
10
  ---
11
 
12
+ # Bencode92/tradepulse-finbert-sentiment
 
13
 
14
+ ## Description
15
+ Fine-tuned FinBERT model for financial sentiment analysis in TradePulse.
16
 
17
+ **Task**: Sentiment Classification
18
+ **Target Column**: `label`
19
+ **Labels**: ['negative', 'neutral', 'positive']
 
 
 
 
 
 
 
20
 
21
+ ## Performance
22
 
23
+ *Last training: 2025-07-09 15:39*
24
+ *Dataset: `news_20250709.csv` (56 samples)*
25
 
26
+ | Metric | Value |
27
+ |--------|-------|
28
+ | Loss | 1.6623 |
29
+ | Accuracy | 0.7857 |
30
+ | F1 Score | 0.7837 |
31
+ | F1 Macro | 0.7837 |
32
+ | Precision | 0.7934 |
33
+ | Recall | 0.7857 |
34
 
35
+ ## Training Details
36
 
37
+ - **Base Model**: Bencode92/tradepulse-finbert-sentiment
38
+ - **Training Mode**: Incremental
39
+ - **Epochs**: 2
40
+ - **Learning Rate**: 1e-05
41
+ - **Batch Size**: 4
42
+ - **Class Balancing**: None
43
 
44
+ ## Usage
45
 
46
+ ```python
47
+ from transformers import AutoTokenizer, AutoModelForSequenceClassification
48
 
49
+ tokenizer = AutoTokenizer.from_pretrained("Bencode92/tradepulse-finbert-sentiment")
50
+ model = AutoModelForSequenceClassification.from_pretrained("Bencode92/tradepulse-finbert-sentiment")
51
 
52
+ # Example prediction
53
+ text = "Apple reported strong quarterly earnings beating expectations"
54
+ inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
55
+ outputs = model(**inputs)
56
+ predictions = outputs.logits.softmax(dim=-1)
57
+ ```
 
 
 
58
 
59
+ ## Model Card Authors
60
 
61
+ - TradePulse ML Team
62
+ - Auto-generated on 2025-07-09 15:39:06
 
 
 
 
 
 
 
 
 
 
 
checkpoint-11/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Bencode92/tradepulse-finbert-sentiment",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "neutral",
14
+ "2": "positive"
15
+ },
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "label2id": {
19
+ "negative": 0,
20
+ "neutral": 1,
21
+ "positive": 2
22
+ },
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "bert",
26
+ "num_attention_heads": 12,
27
+ "num_hidden_layers": 12,
28
+ "pad_token_id": 0,
29
+ "position_embedding_type": "absolute",
30
+ "problem_type": "single_label_classification",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.41.0",
33
+ "type_vocab_size": 2,
34
+ "use_cache": true,
35
+ "vocab_size": 30873
36
+ }
checkpoint-11/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beb7a90b969fd27377b1473dfa16619a7086d635988a964302c9887669d7fe5d
3
+ size 439039996
checkpoint-11/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-11/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-11/tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "2": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "3": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "4": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "5": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "never_split": null,
52
+ "pad_to_multiple_of": null,
53
+ "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
+ "sep_token": "[SEP]",
57
+ "stride": 0,
58
+ "strip_accents": null,
59
+ "tokenize_chinese_chars": true,
60
+ "tokenizer_class": "BertTokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
+ "unk_token": "[UNK]"
64
+ }
checkpoint-11/trainer_state.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7836734693877551,
3
+ "best_model_checkpoint": "hf-sentiment-production/checkpoint-11",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 11,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.9090909090909091,
13
+ "grad_norm": 20.06096839904785,
14
+ "learning_rate": 1e-05,
15
+ "loss": 0.5297,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accuracy": 0.7857142857142857,
21
+ "eval_f1": 0.7836734693877551,
22
+ "eval_f1_macro": 0.7746031746031745,
23
+ "eval_loss": 1.6622570753097534,
24
+ "eval_precision": 0.7933673469387755,
25
+ "eval_precision_macro": 0.7579365079365079,
26
+ "eval_recall": 0.7857142857142857,
27
+ "eval_recall_macro": 0.8055555555555555,
28
+ "eval_runtime": 7.8549,
29
+ "eval_samples_per_second": 1.782,
30
+ "eval_steps_per_second": 0.509,
31
+ "step": 11
32
+ }
33
+ ],
34
+ "logging_steps": 10,
35
+ "max_steps": 22,
36
+ "num_input_tokens_seen": 0,
37
+ "num_train_epochs": 2,
38
+ "save_steps": 500,
39
+ "stateful_callbacks": {
40
+ "EarlyStoppingCallback": {
41
+ "args": {
42
+ "early_stopping_patience": 1,
43
+ "early_stopping_threshold": 0.0
44
+ },
45
+ "attributes": {
46
+ "early_stopping_patience_counter": 0
47
+ }
48
+ },
49
+ "TrainerControl": {
50
+ "args": {
51
+ "should_epoch_stop": false,
52
+ "should_evaluate": false,
53
+ "should_log": false,
54
+ "should_save": true,
55
+ "should_training_stop": false
56
+ },
57
+ "attributes": {}
58
+ }
59
+ },
60
+ "total_flos": 11050763544576.0,
61
+ "train_batch_size": 4,
62
+ "trial_name": null,
63
+ "trial_params": null
64
+ }
checkpoint-11/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "models/sentiment-20250709_142546",
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
 
1
  {
2
+ "_name_or_path": "Bencode92/tradepulse-finbert-sentiment",
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:68968a5e5d224d145993c63372585240c43357c1f46c0d9a6d3d829c863a8b84
3
  size 439039996
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beb7a90b969fd27377b1473dfa16619a7086d635988a964302c9887669d7fe5d
3
  size 439039996
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dbb20ea164781712ff1ef105d582835c71998e3b3b930cbfc7ce4dddca8b9e20
3
- size 5585
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee63087a09f921beca0840c54a1806c0f34dd30d754242a62ed314cf5ffea0c3
3
+ size 5521