V0.11

README.md

@@ -21,12 +21,8 @@ In order to valid the annotation, we search an agreement between raters to emoti
 #### sentiment analysis  				
 |              | precision | recall | f1-score |
 |--------------|-----------|--------|----------|
-| 0            | 0.95      | 0.95   | 0.95     |
-| 1            | 0.90      | 0.90   | 0.90     |
-| accuracy     |           |        | 0.93     |
-| macro avg    | 0.92      | 0.93    | 0.92     |
-| weighted avg | 0.93      | 0.93   | 0.93     |
-
-trained on [Amram, A., Ben-David, A., and Tsarfaty, R. (2018) dataset](https://github.com/omilab/Neural-Sentiment-Analyzer-for-Modern-Hebrew) and ours (will publish soon)
-evaluated by test data set by Amram et al.
-
+| 0            | 0.95      | 0.97   | 0.96     |
+| 1            | 0.91      | 0.84   | 0.87     |
+| accuracy     | 0.94      | 305    | 0.92     |
+| macro avg    | 0.93      | 0.91   | 0.92     |
+| weighted avg | 0.94      | 0.94   | 0.94     |

config.json

@@ -15,7 +15,7 @@
   "num_attention_heads": 12,
   "num_hidden_layers": 6,
   "pad_token_id": 0,
-  "total_flos": 9592087686854400,
+  "total_flos": 3118796722622472192,
   "type_vocab_size": 1,
   "vocab_size": 52000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfa1a0245b234a22e519db453643ac314d4d8d1989e44cbac793a7e07e3e707b
+oid sha256:f8c18b6c74323cc1d85483e93c48b66f6ad0a456bfc427e0ce74a1e667fe2825
 size 333858111

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c5f7d9521882eb40fcbec83b7f666ba49c66ce5e296948d034effa92d7710c3
+oid sha256:7f50e8cde1db05799ace5177491a0b663684c58e0226b05be15789d636ce053e
 size 1839