Update README.md

README.md

@@ -1,4 +1,4 @@
-This is the IndicBART model. For detailed documentation look here: https://indicnlp.ai4bharat.org/indic-bart/ and https://github.com/AI4Bharat/indic-bart/
+This is the IndicBART model fine-tuned on the PMI and PIB dataset. For detailed documentation look here: https://indicnlp.ai4bharat.org/indic-bart/ and https://github.com/AI4Bharat/indic-bart/
 
 Usage:
 
@@ -6,13 +6,13 @@ Usage:
 from transformers import MBartForConditionalGeneration, AutoModelForSeq2SeqLM
 from transformers import AlbertTokenizer, AutoTokenizer
 
-tokenizer = AutoTokenizer.from_pretrained("prajdabre/IndicBART", do_lower_case=False, use_fast=False, keep_accents=True)
+tokenizer = AutoTokenizer.from_pretrained("prajdabre/IndicBART-XXEN", do_lower_case=False, use_fast=False, keep_accents=True)
 
-# Or use tokenizer = AlbertTokenizer.from_pretrained("prajdabre/IndicBART", do_lower_case=False, use_fast=False, keep_accents=True)
+# Or use tokenizer = AlbertTokenizer.from_pretrained("prajdabre/IndicBART-XXEN", do_lower_case=False, use_fast=False, keep_accents=True)
 
-model = AutoModelForSeq2SeqLM.from_pretrained("prajdabre/IndicBART")
+model = AutoModelForSeq2SeqLM.from_pretrained("prajdabre/IndicBART-XXEN")
 
-# Or use model = MBartForConditionalGeneration.from_pretrained("prajdabre/IndicBART")
+# Or use model = MBartForConditionalGeneration.from_pretrained("prajdabre/IndicBART-XXEN")
 
 # Some initial mapping
 bos_id = tokenizer._convert_token_to_id_with_added_voc("<s>")
@@ -20,10 +20,10 @@ eos_id = tokenizer._convert_token_to_id_with_added_voc("</s>")
 pad_id = tokenizer._convert_token_to_id_with_added_voc("<pad>")
 # To get lang_id use any of ['<2as>', '<2bn>', '<2en>', '<2gu>', '<2hi>', '<2kn>', '<2ml>', '<2mr>', '<2or>', '<2pa>', '<2ta>', '<2te>']
 
-# First tokenize the input and outputs. The format below is how IndicBART was trained so the input should be "Sentence </s> <2xx>" where xx is the language code. Similarly, the output should be "<2yy> Sentence </s>". 
-inp = tokenizer("I am a boy </s> <2en>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids # tensor([[  466,  1981,    80, 25573, 64001, 64004]])
+# First tokenize the input and outputs. The format below is how IndicBART-XXEN was trained so the input should be "Sentence </s> <2xx>" where xx is the language code. Similarly, the output should be "<2yy> Sentence </s>". 
+inp = tokenizer("मैं  एक लड़का हूँ </s> <2hi>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids # tensor([[  466,  1981,    80, 25573, 64001, 64004]])
 
-out = tokenizer("<2hi> मैं  एक लड़का हूँ </s>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids # tensor([[64006,   942,    43, 32720,  8384, 64001]])
+out = tokenizer("<2en> I am a boy </s>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids # tensor([[64006,   942,    43, 32720,  8384, 64001]])
 
 model_outputs=model(input_ids=inp, decoder_input_ids=out[:,0:-1], labels=out[:,1:])
 
@@ -37,7 +37,7 @@ model_outputs.logits
 
 model.eval() # Set dropouts to zero
 
-model_output=model.generate(inp, use_cache=True, num_beams=4, max_length=20, min_length=1, early_stopping=True, pad_token_id=pad_id, bos_token_id=bos_id, eos_token_id=eos_id, decoder_start_token_id=tokenizer._convert_token_to_id_with_added_voc("<2en>"))
+model_output=model.generate(inp, use_cache=True, num_beams=4, max_length=20, min_length=1, early_stopping=True, pad_token_id=pad_id, bos_token_id=bos_id, eos_token_id=eos_id, decoder_start_token_id=tokenizer._convert_token_to_id_with_added_voc("<2hi>"))
 
 
 # Decode to get output strings
@@ -46,17 +46,6 @@ decoded_output=tokenizer.decode(model_output[0], skip_special_tokens=True, clean
 
 print(decoded_output) # I am a boy
 
-# What if we mask?
-
-inp = tokenizer("I am [MASK] </s> <2en>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids
-
-model_output=model.generate(inp, use_cache=True, num_beams=4, max_length=20, min_length=1, early_stopping=True, pad_token_id=pad_id, bos_token_id=bos_id, eos_token_id=eos_id, decoder_start_token_id=tokenizer._convert_token_to_id_with_added_voc("<2en>"))
-
-decoded_output=tokenizer.decode(model_output[0], skip_special_tokens=True, clean_up_tokenization_spaces=False)
-
-print(decoded_output) # I am happy
-```
-
 Notes:
 1. This is compatible with the latest version of transformers but was developed with version 4.3.2 so consider using 4.3.2 if possible.
 2. While I have only shown how to let logits and loss and how to generate outputs, you can do pretty much everything the MBartForConditionalGeneration class can do as in https://huggingface.co/docs/transformers/model_doc/mbart#transformers.MBartForConditionalGeneration