algumusrende commited on
Commit
22042be
·
verified ·
1 Parent(s): e5e1651

Update README.md

Browse files

Created the model card and readme.md

Files changed (1) hide show
  1. README.md +56 -3
README.md CHANGED
@@ -1,3 +1,56 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: tr
3
+ license: mit
4
+ ---
5
+ # turkish-text-classification-model
6
+
7
+ https://huggingface.co/algumusrende/turkish-text-classification-model
8
+
9
+ This model is used for Sentiment Analysis, which is based on bert-base-turkish-sentiment-cased https://huggingface.co/savasy/bert-base-turkish-sentiment-cased
10
+
11
+ ## Dataset
12
+
13
+ The dataset is taken from https://www.kaggle.com/datasets/burhanbilenn/duygu-analizi-icin-urun-yorumlari?select=magaza_yorumlari_duygu_analizi.csv
14
+
15
+ Containing product reviews of electronics stores in Turkish Language, with 3 categories["Olumlu(Positive)", "Olumsuz(Negative)", "Tarafsız(Neutral)"]2 columns and 11429 rows (3 NaN rows), encoded in "utf-16"
16
+
17
+ *Dataset*
18
+
19
+ | *size* | *data* |
20
+ |--------|----|
21
+ | 5713 |train.csv|
22
+ | 2856 |val.csv|
23
+ | 2857 |test.tsv|
24
+ | *11426* |*total*|
25
+
26
+ ## Training and Results
27
+
28
+ |*index*|*eval\_loss*|*eval\_Accuracy*|*eval\_F1*|*eval\_Precision*|*eval\_Recall*|
29
+ |---|---|---|---|---|---|
30
+ |train|0\.41672539710998535|0\.8531419569403116|0\.8346503162224169|0\.842628684710363|0\.8315839726920476|
31
+ |val|0\.6787932515144348|0\.7545518207282913|0\.7277930570101517|0\.7311753495947505|0\.7293434379700242|
32
+ |test|0\.6885481476783752|0\.7434371718585929|0\.7170880702233838|0\.7189901255561661|0\.7180628887201386|
33
+
34
+ ## Code Usage
35
+
36
+ ```python
37
+ from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline
38
+
39
+ model = AutoModelForSequenceClassification.from_pretrained("algumusrende/turkish-text-classification-model")
40
+ tokenizer= AutoTokenizer.from_pretrained("algumusrende/turkish-text-classification-model")
41
+ pipe = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
42
+
43
+ pipe("Son zamanlarda ekonomideki istikrar, borsa endeksine de olumlu yansıdı.")
44
+ # [{'label': 'Olumlu', 'score': 0.6654265522956848}]
45
+
46
+ pipe("Geçirdiğim diş operasyonu için çekilen röntgen filmleri sağlık yardımı kapsamında ödenmedi.")
47
+ # [{'label': 'Olumsuz', 'score': 0.9064584970474243}]
48
+
49
+ pipe("Eskiden bayramlarda çikolata dağıtlırdı, artık bunu göremiyoruz.")
50
+ # [{'label': 'Olumsuz', 'score': 0.7049197554588318}]
51
+
52
+ pipe("Ürün genel itibari ile iyi sayılır, ancak bazı eksikleri de var.")
53
+ # [{'label': 'Tarafsız', 'score': 0.9369649887084961}]
54
+
55
+ ```
56
+ ---