bart-large-samsum此模型使用微软的 Azure Machine Learning Service 进行训练。它基于 facebook/bart-large 检查点,在 samsum 语料库上进行了微调。
from transformers import pipeline
summarizer = pipeline("summarization", model="linydub/bart-large-samsum")
input_text = '''
Henry: Hey, is Nate coming over to watch the movie tonight?
Kevin: Yea, he said he'll be arriving a bit later at around 7 since he gets off of work at 6. Have you taken out the garbage yet?
Henry: Oh I forgot. I'll do that once I'm finished with my assignment for my math class.
Kevin: Yea, you should take it out as soon as possible. And also, Nate is bringing his girlfriend.
Henry: Nice, I'm really looking forward to seeing them again.
'''
summarizer(input_text)有关微调过程的更多信息(包括样本和基准):
[预览版] https://github.com/linydub/azureml-greenai-txtsum
这些结果来自 Azure Monitor 指标。所有实验均在 AzureML 低优先级计算集群上运行。
| 关键项 | 值 |
|---|---|
| 区域 | 美国西部 2 |
| AzureML 计算 SKU | STANDARD_ND40RS_V2 |
| 计算 SKU GPU 设备 | 8 x NVIDIA V100 32GB(NVLink) |
| 计算节点数 | 1 |
| 运行时长 | 6 分 48 秒 |
| 计算成本(专用/低优先级) | 2.50 美元 / 0.50 美元 |
| 平均 CPU 利用率 | 47.9% |
| 平均 GPU 利用率 | 69.8% |
| 平均 GPU 内存使用量 | 25.71 GB |
| 总 GPU 能源使用量 | 370.84 kJ |
*计算成本(美元)是根据运行时长、使用的计算节点数以及 SKU 的每小时价格估算得出的。最新的 SKU 定价可在 此处 找到。
这些结果是使用 CodeCarbon 获得的。碳排放量仅根据训练运行时间估算(不包括设置和评估运行时间)。
| 关键项 | 值 |
|---|---|
| timestamp | 2021-09-16T23:54:25 |
| duration | 263.2430217266083 |
| emissions | 0.029715544634717518 |
| energy_consumed | 0.09985062041235725 |
| country_name | USA |
| region | Washington |
| cloud_provider | azure |
| cloud_region | westus2 |
| ROUGE 指标 | 分数 |
|---|---|
| eval_rouge1 | 55.0234 |
| eval_rouge2 | 29.6005 |
| eval_rougeL | 44.914 |
| eval_rougeLsum | 50.464 |
| predict_rouge1 | 53.4345 |
| predict_rouge2 | 28.7445 |
| predict_rougeL | 44.1848 |
| predict_rougeLsum | 49.1874 |
| 指标 | 数值 |
|---|---|
| epoch | 3.0 |
| eval_gen_len | 30.6027 |
| eval_loss | 1.4327096939086914 |
| eval_runtime | 22.9127 |
| eval_samples | 818 |
| eval_samples_per_second | 35.701 |
| eval_steps_per_second | 0.306 |
| predict_gen_len | 30.4835 |
| predict_loss | 1.4501988887786865 |
| predict_runtime | 26.0269 |
| predict_samples | 819 |
| predict_samples_per_second | 31.467 |
| predict_steps_per_second | 0.269 |
| train_loss | 1.2014821151207233 |
| train_runtime | 263.3678 |
| train_samples | 14732 |
| train_samples_per_second | 167.811 |
| train_steps_per_second | 1.321 |
| total_steps | 348 |
| total_flops | 4.26008990669865e+16 |