owinton commited on
Commit
b60419a
1 Parent(s): 9569e4e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -4
README.md CHANGED
@@ -19,11 +19,12 @@ The model is based on the [North-T5-NCC Large](https://huggingface.co/north/t5_l
19
  - **Model type:** Language model restricted to classification
20
  - **Language(s) (NLP):** Danish and Norwegian
21
  - **License:** [More Information Needed]
22
- - **Finetuned from model:** [More information needed]
23
 
24
 
25
  # Direct Use
26
- The model can be used directly to classify Danish and Norwegian social media posts (or similar pieces of text).
 
27
 
28
 
29
  # Bias, Risks, and Limitations
@@ -33,7 +34,8 @@ The model can be used directly to classify Danish and Norwegian social media pos
33
  # Training Data
34
  A collection of ~70k Norwegian and ~67k Danish social media posts have been manually annotated as 'attack' or 'not attack' by six individual coders. 5% of the posts have been annotated by more then one annotator, with the annotators in agreement for 83% of annotations.
35
 
36
- *Hvad er data-split metoden? Hvad er training-validation-test split?*
 
37
 
38
  # Evaluation
39
  <!-- This section describes the evaluation protocols and provides the results. -->
@@ -94,7 +96,7 @@ model = AutoModelForSeq2SeqLM.from_pretrained("ogtal/A-og-ttack2")
94
 
95
  # Give sample text. The example is from a social media comment.
96
  sample_text = "Velbekomme dit klamme usle løgnersvin!"
97
- input_ids = tokenizer("Velbekomme", return_tensors="pt").input_ids
98
 
99
  # Forward pass and print the output
100
  outputs = model.generate(input_ids)
 
19
  - **Model type:** Language model restricted to classification
20
  - **Language(s) (NLP):** Danish and Norwegian
21
  - **License:** [More Information Needed]
22
+ - **Finetuned from model:** [North-T5-NCC Large](https://huggingface.co/north/t5_large_NCC)
23
 
24
 
25
  # Direct Use
26
+ This model can be used for classifying Danish and Norwegian social media posts or similar text.
27
+
28
 
29
 
30
  # Bias, Risks, and Limitations
 
34
  # Training Data
35
  A collection of ~70k Norwegian and ~67k Danish social media posts have been manually annotated as 'attack' or 'not attack' by six individual coders. 5% of the posts have been annotated by more then one annotator, with the annotators in agreement for 83% of annotations.
36
 
37
+ [More information needed on the data split method and the training-validation-test split.]
38
+
39
 
40
  # Evaluation
41
  <!-- This section describes the evaluation protocols and provides the results. -->
 
96
 
97
  # Give sample text. The example is from a social media comment.
98
  sample_text = "Velbekomme dit klamme usle løgnersvin!"
99
+ input_ids = tokenizer(sample_text, return_tensors="pt").input_ids
100
 
101
  # Forward pass and print the output
102
  outputs = model.generate(input_ids)