commit files to HF hub

README.md

@@ -0,0 +1,11 @@
+ # `cardiffnlp/xlm-v-base-tweet-sentiment-ar`
+This model is a fine-tuned version of [facebook/xlm-v-base](https://huggingface.co/facebook/xlm-v-base) on the 
+[cardiffnlp/tweet_sentiment_multilingual](https://huggingface.co/datasets/cardiffnlp/tweet_sentiment_multilingual) (arabic).
+Following metrics are computed on the `test` split of 
+[cardiffnlp/tweet_sentiment_multilingual](https://huggingface.co/datasets/cardiffnlp/tweet_sentiment_multilingual)(arabic). 
+
+|    |   eval_f1_micro |   eval_recall_micro |   eval_precision_micro |   eval_f1_macro |   eval_recall_macro |   eval_precision_macro |   eval_accuracy |
+|---:|----------------:|--------------------:|-----------------------:|----------------:|--------------------:|-----------------------:|----------------:|
+|  0 |              60 |                  60 |                     60 |           59.83 |                  60 |                  59.94 |              60 |
+
+Check the result file [here](https://huggingface.co/cardiffnlp/xlm-v-base-tweet-sentiment-ar/raw/main/eval.json).

eval.json

@@ -0,0 +1 @@
+{"eval_loss": 0.9538047909736633, "eval_f1_micro": 0.6, "eval_recall_micro": 0.6, "eval_precision_micro": 0.6, "eval_f1_macro": 0.5982602849285922, "eval_recall_macro": 0.6, "eval_precision_macro": 0.5994300793632982, "eval_accuracy": 0.6, "eval_runtime": 2.2191, "eval_samples_per_second": 392.056, "eval_steps_per_second": 49.12}