Update README.md

README.md

@@ -5,6 +5,42 @@ This is the converted model from Unbabel/wmt22-cometkiwi-da
 2) Renamed the keys to match the original Facebook/XLM-roberta-large
 3) kept the layer_wise_attention / estimator layers
 
+Because of a hack in HF's code I had to rename the "layerwise_attention.gamma" key to "layerwise_attention.gam"
+
+I changed the config.json key "layer_transformation" from sparsemax to softmax because there is a bug in COMET since the flag is not passed, the actual function used is the default which is softmax.
+
+Usage:
+
+```
+from transformers import XLMRobertaTokenizer, XLMRobertaTokenizerFast, AutoModel
+tokenizer = XLMRobertaTokenizerFast.from_pretrained("vince62s/wmt22-cometkiwi-da-roberta-large", trust_remote_code=True)
+model = AutoModel.from_pretrained("vince62s/wmt22-cometkiwi-da-roberta-large", trust_remote_code=True)
+
+text = "Hello world! </s> </s> Bonjour le monde"
+encoded_text = tokenizer(text, return_tensors='pt')
+print(encoded_text)
+output = model(**encoded_text)
+print(output[0])
+
+{'input_ids': tensor([[    0, 35378,  8999,    38,     2,     2, 84602,    95, 11146,     2]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])}
+tensor([[0.8640]], grad_fn=<AddmmBackward0>)
+
+```
+
+Let's double check with the original code from Unbabel Comet:
+
+```
+from comet import download_model, load_from_checkpoint
+model = load_from_checkpoint("/home/vincent/Downloads/cometkiwi22/checkpoints/model.ckpt") # this is the Unbabel checkpoint
+data = [{"mt": "Hello world!", "src": "Bonjour le monde"}]
+output = model.predict(data, gpus=0)
+print(output)
+
+Prediction([('scores', [0.863973081111908]),
+            ('system_score', 0.863973081111908)])
+```
+
+