Tomás Fernandes commited on
Commit ·
1ee8539
1
Parent(s): dd37c58
Update model card metadata and W&B
Browse files
README.md
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
---
|
| 2 |
language: "en"
|
|
|
|
| 3 |
tags:
|
| 4 |
- financial-sentiment-analysis
|
| 5 |
- sentiment-analysis
|
|
@@ -11,12 +12,45 @@ widget:
|
|
| 11 |
- text: "ImaginaryJetCo bookings hit by Omicron variant as losses total £1bn."
|
| 12 |
- text: "Q1 ImaginaryGame's earnings beat expectations."
|
| 13 |
- text: "Should we buy IMAGINARYSTOCK today?"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
---
|
| 15 |
|
| 16 |
# FINPerceiver
|
| 17 |
FINPerceiver is a fine-tuned Perceiver IO language model for financial sentiment analysis.
|
| 18 |
More details on the training process of this model are available on the [GitHub repository](https://github.com/warwickai/fin-perceiver).
|
| 19 |
|
|
|
|
|
|
|
| 20 |
We achieved the following results with 10-fold cross validation.
|
| 21 |
|
| 22 |
```
|
|
|
|
| 1 |
---
|
| 2 |
language: "en"
|
| 3 |
+
license: apache-2.0
|
| 4 |
tags:
|
| 5 |
- financial-sentiment-analysis
|
| 6 |
- sentiment-analysis
|
|
|
|
| 12 |
- text: "ImaginaryJetCo bookings hit by Omicron variant as losses total £1bn."
|
| 13 |
- text: "Q1 ImaginaryGame's earnings beat expectations."
|
| 14 |
- text: "Should we buy IMAGINARYSTOCK today?"
|
| 15 |
+
metrics:
|
| 16 |
+
- recall
|
| 17 |
+
- f1
|
| 18 |
+
- accuracy
|
| 19 |
+
- precision
|
| 20 |
+
model-index:
|
| 21 |
+
- name: fin-perceiver
|
| 22 |
+
results:
|
| 23 |
+
- task:
|
| 24 |
+
name: Text Classification
|
| 25 |
+
type: text-classification
|
| 26 |
+
dataset:
|
| 27 |
+
name: financial_phrasebank
|
| 28 |
+
type: financial_phrasebank
|
| 29 |
+
args: sentences_50agree
|
| 30 |
+
metrics:
|
| 31 |
+
- name: Accuracy
|
| 32 |
+
type: accuracy
|
| 33 |
+
value: 0.8624
|
| 34 |
+
- name: F1
|
| 35 |
+
type: f1
|
| 36 |
+
value: 0.8416
|
| 37 |
+
args: macro
|
| 38 |
+
- name: Precision
|
| 39 |
+
type: precision
|
| 40 |
+
value: 0.8438
|
| 41 |
+
args: macro
|
| 42 |
+
- name: Recall
|
| 43 |
+
type: recall
|
| 44 |
+
value: 0.8415
|
| 45 |
+
args: macro
|
| 46 |
---
|
| 47 |
|
| 48 |
# FINPerceiver
|
| 49 |
FINPerceiver is a fine-tuned Perceiver IO language model for financial sentiment analysis.
|
| 50 |
More details on the training process of this model are available on the [GitHub repository](https://github.com/warwickai/fin-perceiver).
|
| 51 |
|
| 52 |
+
Weights & Biases was used to track experiments.
|
| 53 |
+
|
| 54 |
We achieved the following results with 10-fold cross validation.
|
| 55 |
|
| 56 |
```
|