From 0f6017bee3668222000ea788f552fac700362963 Mon Sep 17 00:00:00 2001 From: patrickvonplaten Date: Thu, 26 Dec 2019 00:35:11 +0100 Subject: [PATCH] improve comments for examples --- src/transformers/modeling_utils.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/transformers/modeling_utils.py b/src/transformers/modeling_utils.py index 5a36b436be..f2d3ca39f1 100644 --- a/src/transformers/modeling_utils.py +++ b/src/transformers/modeling_utils.py @@ -614,7 +614,7 @@ class PreTrainedModel(nn.Module): model = AutoModelWithLMHead.from_pretrained('openai-gpt') # Download model and configuration from S3 and cache. input_context = 'The dog' input_ids = torch.tensor(tokenizer.encode(input_context)).unsqueeze(0) # encode input context - outputs = model.generate(input_ids=input_ids, do_sample=True, num_beams=5, num_return_sequences=3) # generate 3 independent sequences using beam search decoding (5 beams) from initial context 'The dog' + outputs = model.generate(input_ids=input_ids, do_sample=True, num_beams=5, num_return_sequences=3, temperature=1.5) # generate 3 independent sequences using beam search decoding (5 beams) with sampling from initial context 'The dog' for i in range(3): # 3 output sequences were generated print('Generated {}: {}'.format(i, tokenizer.decode(outputs[0][i], skip_special_tokens=True))) @@ -622,7 +622,7 @@ class PreTrainedModel(nn.Module): model = AutoModelWithLMHead.from_pretrained('distilgpt2') # Download model and configuration from S3 and cache. input_context = 'The dog' input_ids = torch.tensor(tokenizer.encode(input_context)).unsqueeze(0) # encode input context - outputs = model.generate(input_ids=input_ids, max_length=40, do_sample=True, temperature=0.7, bos_token_id=tokenizer.bos_token_id, eos_token_ids=tokenizer.eos_token_id, num_beams=3) # generate sequences using beam search decoding (3 beams) + outputs = model.generate(input_ids=input_ids, max_length=40, temperature=0.7, bos_token_id=tokenizer.bos_token_id, eos_token_ids=tokenizer.eos_token_id, num_beams=3) # generate sequences using greedy beam search decoding (3 beams) print('Generated: {}'.format(tokenizer.decode(outputs[0], skip_special_tokens=True))) tokenizer = AutoTokenizer.from_pretrained('ctrl') # Initialize tokenizer