[🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' #1240

GitHub-ZHY · 2022-04-07T08:11:27Z

描述这个 bug
GCMC Hyper_tuning 代码不通，总是有不同的模型会在Hp时报这个Bug，rng.integers(low, high, size) no intergers。

如何复现
复现这个 bug 的步骤：

您引入的额外 yaml 文件

gcmc.yaml
embedding_size: 64
sparse_feature: True
class_num: 2
learner: adam
weight_decay: 0.995
train_batch_size: 4096
eval_batch_size: 40960000
metrics: ["Recall","NDCG","Hit","Precision","MRR"]
topk: [ 1,2,5,10,20,40,50,60,80,100 ]
valid_metric: Recall@20
hyper.gcmc
learning_rate choice [0.01]
dropout_prob choice [0.3,0.4,0.5,0.6,0.7,0.8]
accum choice ['stack','sum']
gcn_output_dim choice [256,500,1024]
num_basis_functions choice ['2']

您的代码
您的运行脚本
--model=GCMC --config_files=GCMC.yaml --dataset='ml-1m' --config_files=GCMC.yaml --params_file=hyper.gcmc --output_file=hyper_gcmc_ml1m.result

预期
通。

屏幕截图
Connected to pydev debugger (build 211.6693.115)

ERROR in rec_eval
EXCEPTION
<class 'AttributeError'>
'numpy.random.mtrand.RandomState' object has no attribute 'integers'
NODE
0 randint
1 Literal{1}
2 size =
3 len
4 array_union
5 array_union
6 array_union
7 Literal{new_ids}
8 rng =
9 Literal{rng-placeholder}

0%| | 0/36 [00:00<?, ?trial/s, best loss=?]
Traceback (most recent call last):
File "/home//anaconda3/envs//lib/python3.8/contextlib.py", line 131, in exit
self.gen.throw(type, value, traceback)
File "/home//anaconda3/envs//lib/python3.8/site-packages/hyperopt/progress.py", line 25, in tqdm_progress_callback
yield pbar
File "/home//anaconda3/envs//lib/python3.8/site-packages/hyperopt/fmin.py", line 278, in run
new_trials = algo(
File "/home///RecBole/recbole/trainer/hyper_tuning.py", line 110, in exhaustive_search
idxs, vals = pyll.rec_eval(domain.s_idxs_vals, memo={
File "/home//anaconda3/envs//lib/python3.8/site-packages/hyperopt/pyll/base.py", line 902, in rec_eval
rval = scope._impls[node.name](*args, **kwargs)
File "/home//anaconda3/envs//lib/python3.8/site-packages/hyperopt/pyll/stochastic.py", line 100, in randint
return rng.integers(low, high, size)
AttributeError: 'numpy.random.mtrand.RandomState' object has no attribute 'integers'

Process finished with exit code 1

链接
None

实验环境（请补全下列信息）：

操作系统: [Windows]
RecBole 版本 [0.1.0]
Python 版本 [3.8]
PyTorch 版本 [如 1.8.1]
cudatoolkit 版本 [11.4]

Ethan-TZ · 2022-04-07T08:25:43Z

@GitHub-ZHY 您好，感谢您的关注！
我们在#1192中修复了这个bug，请使用最新源码重新安装RecBole。

Ethan-TZ · 2022-04-07T11:16:22Z

你可以参考 #1181的解决方案

…

------------------ 原始邮件 ------------------ 发件人: "RUCAIBox/RecBole" ***@***.***>; 发送时间: 2022年4月7日(星期四) 下午4:33 ***@***.***>; ***@***.******@***.***>; 主题: Re: [RUCAIBox/RecBole] [🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' (Issue #1240) @chenyuwuxin 您好，我已经使用了最新的RecBole，#1192也是我提的Issue，好像问题并不出现在此，目前我仍在debug，烦请您再看下。 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

GitHub-ZHY · 2022-04-07T15:19:56Z

@chenyuwuxin 您好，上述问题或因服务器环境问题引起，然而我又发现一个新问题。即yaml内配置 learner 后，虽运行单组run_recbole调试时打印出相应的 learner 但是训练结果损失函数不下降，与此同时，指标结果时有下降时而保持不变，而且最终训练结果是保持不变的，无论其他参数如何修改。

GitHub-ZHY · 2022-04-07T15:21:13Z

@chenyuwuxin 配置文件如下：# general
gpu_id: 0
use_gpu: True
seed: 2022
state: INFO
reproducibility: True
#data_path: 'dataset/ml-1m'
checkpoint_dir: 'saved'
show_progress: True
save_dataset: False
dataset_save_path: None
save_dataloaders: False
dataloaders_save_path: None
log_wandb: False
wandb_project: 'recbole'

training settings

epochs: 300
train_batch_size: 4096
learner: adam
learning_rate: 0.01
neg_sampling:
uniform: 1
eval_step: 1
stopping_step: 10
#clip_grad_norm: ~

clip_grad_norm: {'max_norm': 5, 'norm_type': 2}

weight_decay: 0.995
loss_decimal_place: 4
require_pow: False

evaluation settings

eval_args:
split: {'RS':[0.8,0.1,0.1]}
group_by: user
order: RO
mode: full
repeatable: False
metrics: ["Recall","NDCG","Hit","Precision","MRR"]
topk: [ 1,2,5,10,20,40,50,60,80,100 ]
valid_metric: Recall@20
valid_metric_bigger: True
eval_batch_size: 40960000
metric_decimal_place: 4

Other Hyper Parameters:

#accum: stack
embedding_size: 64
sparse_feature: True
class_num: 2
dropout_prob: 0.3
gcn_output_dim: 256
num_basis_functions: 2

GitHub-ZHY · 2022-04-07T15:22:13Z

Trainable parameters: 4914196
Train 0: 100%|████████████████████████| 197/197 [00:14<00:00, 13.62it/s, GPU RAM: 0.28 G/6.00 G]
07 Apr 23:13 INFO epoch 0 training [time: 14.47s, train loss: 136.5497]
Evaluate : 100%|████████████████████████████| 1/1 [00:00<00:00, 3.95it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:13 INFO epoch 0 evaluating [time: 0.29s, valid_score: 0.007500]
07 Apr 23:13 INFO valid result:
recall@10 : 0.0024 mrr@10 : 0.0075 ndcg@10 : 0.0032 hit@10 : 0.0311 precision@10 : 0.0032
07 Apr 23:13 INFO Saving current: saved\GCMC-Apr-07-2022_23-13-04.pth
Train 1: 100%|████████████████████████| 197/197 [00:14<00:00, 14.04it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:13 INFO epoch 1 training [time: 14.03s, train loss: 136.5497]
Evaluate : 100%|████████████████████████████| 1/1 [00:00<00:00, 4.08it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:13 INFO epoch 1 evaluating [time: 0.31s, valid_score: 0.004600]
07 Apr 23:13 INFO valid result:
recall@10 : 0.0022 mrr@10 : 0.0046 ndcg@10 : 0.0026 hit@10 : 0.0275 precision@10 : 0.003
Train 2: 100%|████████████████████████| 197/197 [00:13<00:00, 14.15it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:13 INFO epoch 2 training [time: 13.92s, train loss: 136.5497]
Evaluate : 100%|████████████████████████████| 1/1 [00:00<00:00, 4.05it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:13 INFO epoch 2 evaluating [time: 0.29s, valid_score: 0.068100]
07 Apr 23:13 INFO valid result:
recall@10 : 0.0185 mrr@10 : 0.0681 ndcg@10 : 0.0299 hit@10 : 0.2066 precision@10 : 0.0269
07 Apr 23:13 INFO Saving current: saved\GCMC-Apr-07-2022_23-13-04.pth
Train 3: 100%|████████████████████████| 197/197 [00:14<00:00, 13.93it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:14 INFO epoch 3 training [time: 14.15s, train loss: 136.5497]
Evaluate : 100%|████████████████████████████| 1/1 [00:00<00:00, 3.91it/s, GPU RAM: 0.78 G/6.00 G]
07 Apr 23:14 INFO epoch 3 evaluating [time: 0.31s, valid_score: 0.068100]
07 Apr 23:14 INFO valid result:
recall@10 : 0.0185 mrr@10 : 0.0681 ndcg@10 : 0.0299 hit@10 : 0.2066 precision@10 : 0.0269
07 Apr 23:14 INFO Saving current: saved\GCMC-Apr-07-2022_23-13-04.pth

GitHub-ZHY · 2022-04-07T15:22:36Z

accum:sum, dropout_prob:0.8, gcn_output_dim:256, learning_rate:0.01, num_basis_functions:2
Valid result:
recall@1 : 0.0013 recall@2 : 0.0027 recall@5 : 0.0074 recall@10 : 0.0134 recall@20 : 0.0258 recall@40 : 0.0568 recall@50 : 0.072 recall@60 : 0.0858 recall@80 : 0.1162 recall@100 : 0.1511 ndcg@1 : 0.022 ndcg@2 : 0.023 ndcg@5 : 0.0242 ndcg@10 : 0.0245 ndcg@20 : 0.0276 ndcg@40 : 0.0382 ndcg@50 : 0.0437 ndcg@60 : 0.0486 ndcg@80 : 0.0589 ndcg@100 : 0.0703 hit@1 : 0.022 hit@2 : 0.0455 hit@5 : 0.1093 hit@10 : 0.1811 hit@20 : 0.2889 hit@40 : 0.4546 hit@50 : 0.5157 hit@60 : 0.5596 hit@80 : 0.6399 hit@100 : 0.706 precision@1 : 0.022 precision@2 : 0.0233 precision@5 : 0.024 precision@10 : 0.0225 precision@20 : 0.0219 precision@40 : 0.0228 precision@50 : 0.0231 precision@60 : 0.0229 precision@80 : 0.0228 precision@100 : 0.0232 mrr@1 : 0.022 mrr@2 : 0.0338 mrr@5 : 0.051 mrr@10 : 0.0603 mrr@20 : 0.0676 mrr@40 : 0.0733 mrr@50 : 0.0747 mrr@60 : 0.0754 mrr@80 : 0.0766 mrr@100 : 0.0773
Test result:
recall@1 : 0.0015 recall@2 : 0.003 recall@5 : 0.0076 recall@10 : 0.0133 recall@20 : 0.0252 recall@40 : 0.0588 recall@50 : 0.0762 recall@60 : 0.0914 recall@80 : 0.1217 recall@100 : 0.1572 ndcg@1 : 0.0265 ndcg@2 : 0.0266 ndcg@5 : 0.0259 ndcg@10 : 0.026 ndcg@20 : 0.0288 ndcg@40 : 0.0399 ndcg@50 : 0.0461 ndcg@60 : 0.0513 ndcg@80 : 0.0616 ndcg@100 : 0.0734 hit@1 : 0.0265 hit@2 : 0.0522 hit@5 : 0.1127 hit@10 : 0.1844 hit@20 : 0.2907 hit@40 : 0.4634 hit@50 : 0.5306 hit@60 : 0.5715 hit@80 : 0.6411 hit@100 : 0.7099 precision@1 : 0.0265 precision@2 : 0.0266 precision@5 : 0.025 precision@10 : 0.0236 precision@20 : 0.023 precision@40 : 0.0241 precision@50 : 0.0245 precision@60 : 0.0243 precision@80 : 0.024 precision@100 : 0.0245 mrr@1 : 0.0265 mrr@2 : 0.0393 mrr@5 : 0.0555 mrr@10 : 0.0649 mrr@20 : 0.072 mrr@40 : 0.0781 mrr@50 : 0.0796 mrr@60 : 0.0803 mrr@80 : 0.0813 mrr@100 : 0.0821

accum:sum, dropout_prob:0.5, gcn_output_dim:256, learning_rate:0.01, num_basis_functions:2
Valid result:
recall@1 : 0.0013 recall@2 : 0.0027 recall@5 : 0.0074 recall@10 : 0.0134 recall@20 : 0.0258 recall@40 : 0.0568 recall@50 : 0.072 recall@60 : 0.0858 recall@80 : 0.1162 recall@100 : 0.1511 ndcg@1 : 0.022 ndcg@2 : 0.023 ndcg@5 : 0.0242 ndcg@10 : 0.0245 ndcg@20 : 0.0276 ndcg@40 : 0.0382 ndcg@50 : 0.0437 ndcg@60 : 0.0486 ndcg@80 : 0.0589 ndcg@100 : 0.0703 hit@1 : 0.022 hit@2 : 0.0455 hit@5 : 0.1093 hit@10 : 0.1811 hit@20 : 0.2889 hit@40 : 0.4546 hit@50 : 0.5157 hit@60 : 0.5596 hit@80 : 0.6399 hit@100 : 0.706 precision@1 : 0.022 precision@2 : 0.0233 precision@5 : 0.024 precision@10 : 0.0225 precision@20 : 0.0219 precision@40 : 0.0228 precision@50 : 0.0231 precision@60 : 0.0229 precision@80 : 0.0228 precision@100 : 0.0232 mrr@1 : 0.022 mrr@2 : 0.0338 mrr@5 : 0.051 mrr@10 : 0.0603 mrr@20 : 0.0676 mrr@40 : 0.0733 mrr@50 : 0.0747 mrr@60 : 0.0754 mrr@80 : 0.0766 mrr@100 : 0.0773
Test result:
recall@1 : 0.0015 recall@2 : 0.003 recall@5 : 0.0076 recall@10 : 0.0133 recall@20 : 0.0252 recall@40 : 0.0588 recall@50 : 0.0762 recall@60 : 0.0914 recall@80 : 0.1217 recall@100 : 0.1572 ndcg@1 : 0.0265 ndcg@2 : 0.0266 ndcg@5 : 0.0259 ndcg@10 : 0.026 ndcg@20 : 0.0288 ndcg@40 : 0.0399 ndcg@50 : 0.0461 ndcg@60 : 0.0513 ndcg@80 : 0.0616 ndcg@100 : 0.0734 hit@1 : 0.0265 hit@2 : 0.0522 hit@5 : 0.1127 hit@10 : 0.1844 hit@20 : 0.2907 hit@40 : 0.4634 hit@50 : 0.5306 hit@60 : 0.5715 hit@80 : 0.6411 hit@100 : 0.7099 precision@1 : 0.0265 precision@2 : 0.0266 precision@5 : 0.025 precision@10 : 0.0236 precision@20 : 0.023 precision@40 : 0.0241 precision@50 : 0.0245 precision@60 : 0.0243 precision@80 : 0.024 precision@100 : 0.0245 mrr@1 : 0.0265 mrr@2 : 0.0393 mrr@5 : 0.0555 mrr@10 : 0.0649 mrr@20 : 0.072 mrr@40 : 0.0781 mrr@50 : 0.0796 mrr@60 : 0.0803 mrr@80 : 0.0813 mrr@100 : 0.0821

accum:sum, dropout_prob:0.3, gcn_output_dim:1024, learning_rate:0.01, num_basis_functions:2
Valid result:
recall@1 : 0.0013 recall@2 : 0.0027 recall@5 : 0.0074 recall@10 : 0.0134 recall@20 : 0.0258 recall@40 : 0.0568 recall@50 : 0.072 recall@60 : 0.0858 recall@80 : 0.1162 recall@100 : 0.1511 ndcg@1 : 0.022 ndcg@2 : 0.023 ndcg@5 : 0.0242 ndcg@10 : 0.0245 ndcg@20 : 0.0276 ndcg@40 : 0.0382 ndcg@50 : 0.0437 ndcg@60 : 0.0486 ndcg@80 : 0.0589 ndcg@100 : 0.0703 hit@1 : 0.022 hit@2 : 0.0455 hit@5 : 0.1093 hit@10 : 0.1811 hit@20 : 0.2889 hit@40 : 0.4546 hit@50 : 0.5157 hit@60 : 0.5596 hit@80 : 0.6399 hit@100 : 0.706 precision@1 : 0.022 precision@2 : 0.0233 precision@5 : 0.024 precision@10 : 0.0225 precision@20 : 0.0219 precision@40 : 0.0228 precision@50 : 0.0231 precision@60 : 0.0229 precision@80 : 0.0228 precision@100 : 0.0232 mrr@1 : 0.022 mrr@2 : 0.0338 mrr@5 : 0.051 mrr@10 : 0.0603 mrr@20 : 0.0676 mrr@40 : 0.0733 mrr@50 : 0.0747 mrr@60 : 0.0754 mrr@80 : 0.0766 mrr@100 : 0.0773
Test result:
recall@1 : 0.0015 recall@2 : 0.003 recall@5 : 0.0076 recall@10 : 0.0133 recall@20 : 0.0252 recall@40 : 0.0588 recall@50 : 0.0762 recall@60 : 0.0914 recall@80 : 0.1217 recall@100 : 0.1572 ndcg@1 : 0.0265 ndcg@2 : 0.0266 ndcg@5 : 0.0259 ndcg@10 : 0.026 ndcg@20 : 0.0288 ndcg@40 : 0.0399 ndcg@50 : 0.0461 ndcg@60 : 0.0513 ndcg@80 : 0.0616 ndcg@100 : 0.0734 hit@1 : 0.0265 hit@2 : 0.0522 hit@5 : 0.1127 hit@10 : 0.1844 hit@20 : 0.2907 hit@40 : 0.4634 hit@50 : 0.5306 hit@60 : 0.5715 hit@80 : 0.6411 hit@100 : 0.7099 precision@1 : 0.0265 precision@2 : 0.0266 precision@5 : 0.025 precision@10 : 0.0236 precision@20 : 0.023 precision@40 : 0.0241 precision@50 : 0.0245 precision@60 : 0.0243 precision@80 : 0.024 precision@100 : 0.0245 mrr@1 : 0.0265 mrr@2 : 0.0393 mrr@5 : 0.0555 mrr@10 : 0.0649 mrr@20 : 0.072 mrr@40 : 0.0781 mrr@50 : 0.0796 mrr@60 : 0.0803 mrr@80 : 0.0813 mrr@100 : 0.0821

accum:stack, dropout_prob:0.4, gcn_output_dim:500, learning_rate:0.01, num_basis_functions:2
Valid result:
recall@1 : 0.0013 recall@2 : 0.0027 recall@5 : 0.0074 recall@10 : 0.0134 recall@20 : 0.0258 recall@40 : 0.0568 recall@50 : 0.072 recall@60 : 0.0858 recall@80 : 0.1162 recall@100 : 0.1511 ndcg@1 : 0.022 ndcg@2 : 0.023 ndcg@5 : 0.0242 ndcg@10 : 0.0245 ndcg@20 : 0.0276 ndcg@40 : 0.0382 ndcg@50 : 0.0437 ndcg@60 : 0.0486 ndcg@80 : 0.0589 ndcg@100 : 0.0703 hit@1 : 0.022 hit@2 : 0.0455 hit@5 : 0.1093 hit@10 : 0.1811 hit@20 : 0.2889 hit@40 : 0.4546 hit@50 : 0.5157 hit@60 : 0.5596 hit@80 : 0.6399 hit@100 : 0.706 precision@1 : 0.022 precision@2 : 0.0233 precision@5 : 0.024 precision@10 : 0.0225 precision@20 : 0.0219 precision@40 : 0.0228 precision@50 : 0.0231 precision@60 : 0.0229 precision@80 : 0.0228 precision@100 : 0.0232 mrr@1 : 0.022 mrr@2 : 0.0338 mrr@5 : 0.051 mrr@10 : 0.0603 mrr@20 : 0.0676 mrr@40 : 0.0733 mrr@50 : 0.0747 mrr@60 : 0.0754 mrr@80 : 0.0766 mrr@100 : 0.0773
Test result:
recall@1 : 0.0015 recall@2 : 0.003 recall@5 : 0.0076 recall@10 : 0.0133 recall@20 : 0.0252 recall@40 : 0.0588 recall@50 : 0.0762 recall@60 : 0.0914 recall@80 : 0.1217 recall@100 : 0.1572 ndcg@1 : 0.0265 ndcg@2 : 0.0266 ndcg@5 : 0.0259 ndcg@10 : 0.026 ndcg@20 : 0.0288 ndcg@40 : 0.0399 ndcg@50 : 0.0461 ndcg@60 : 0.0513 ndcg@80 : 0.0616 ndcg@100 : 0.0734 hit@1 : 0.0265 hit@2 : 0.0522 hit@5 : 0.1127 hit@10 : 0.1844 hit@20 : 0.2907 hit@40 : 0.4634 hit@50 : 0.5306 hit@60 : 0.5715 hit@80 : 0.6411 hit@100 : 0.7099 precision@1 : 0.0265 precision@2 : 0.0266 precision@5 : 0.025 precision@10 : 0.0236 precision@20 : 0.023 precision@40 : 0.0241 precision@50 : 0.0245 precision@60 : 0.0243 precision@80 : 0.024 precision@100 : 0.0245 mrr@1 : 0.0265 mrr@2 : 0.0393 mrr@5 : 0.0555 mrr@10 : 0.0649 mrr@20 : 0.072 mrr@40 : 0.0781 mrr@50 : 0.0796 mrr@60 : 0.0803 mrr@80 : 0.0813 mrr@100 : 0.0821

accum:stack, dropout_prob:0.3, gcn_output_dim:500, learning_rate:0.01, num_basis_functions:2
Valid result:
recall@1 : 0.0013 recall@2 : 0.0027 recall@5 : 0.0074 recall@10 : 0.0134 recall@20 : 0.0258 recall@40 : 0.0568 recall@50 : 0.072 recall@60 : 0.0858 recall@80 : 0.1162 recall@100 : 0.1511 ndcg@1 : 0.022 ndcg@2 : 0.023 ndcg@5 : 0.0242 ndcg@10 : 0.0245 ndcg@20 : 0.0276 ndcg@40 : 0.0382 ndcg@50 : 0.0437 ndcg@60 : 0.0486 ndcg@80 : 0.0589 ndcg@100 : 0.0703 hit@1 : 0.022 hit@2 : 0.0455 hit@5 : 0.1093 hit@10 : 0.1811 hit@20 : 0.2889 hit@40 : 0.4546 hit@50 : 0.5157 hit@60 : 0.5596 hit@80 : 0.6399 hit@100 : 0.706 precision@1 : 0.022 precision@2 : 0.0233 precision@5 : 0.024 precision@10 : 0.0225 precision@20 : 0.0219 precision@40 : 0.0228 precision@50 : 0.0231 precision@60 : 0.0229 precision@80 : 0.0228 precision@100 : 0.0232 mrr@1 : 0.022 mrr@2 : 0.0338 mrr@5 : 0.051 mrr@10 : 0.0603 mrr@20 : 0.0676 mrr@40 : 0.0733 mrr@50 : 0.0747 mrr@60 : 0.0754 mrr@80 : 0.0766 mrr@100 : 0.0773
Test result:
recall@1 : 0.0015 recall@2 : 0.003 recall@5 : 0.0076 recall@10 : 0.0133 recall@20 : 0.0252 recall@40 : 0.0588 recall@50 : 0.0762 recall@60 : 0.0914 recall@80 : 0.1217 recall@100 : 0.1572 ndcg@1 : 0.0265 ndcg@2 : 0.0266 ndcg@5 : 0.0259 ndcg@10 : 0.026 ndcg@20 : 0.0288 ndcg@40 : 0.0399 ndcg@50 : 0.0461 ndcg@60 : 0.0513 ndcg@80 : 0.0616 ndcg@100 : 0.0734 hit@1 : 0.0265 hit@2 : 0.0522 hit@5 : 0.1127 hit@10 : 0.1844 hit@20 : 0.2907 hit@40 : 0.4634 hit@50 : 0.5306 hit@60 : 0.5715 hit@80 : 0.6411 hit@100 : 0.7099 precision@1 : 0.0265 precision@2 : 0.0266 precision@5 : 0.025 precision@10 : 0.0236 precision@20 : 0.023 precision@40 : 0.0241 precision@50 : 0.0245 precision@60 : 0.0243 precision@80 : 0.024 precision@100 : 0.0245 mrr@1 : 0.0265 mrr@2 : 0.0393 mrr@5 : 0.0555 mrr@10 : 0.0649 mrr@20 : 0.072 mrr@40 : 0.0781 mrr@50 : 0.0796 mrr@60 : 0.0803 mrr@80 : 0.0813 mrr@100 : 0.0821

Ethan-TZ · 2022-04-08T11:21:50Z

@GitHub-ZHY 可能是配置出现了问题，general 类模型在ml-1m数据集运行的示例配置文件为：

# dataset config
field_separator: "\t"
seq_separator: " "
USER_ID_FIELD: user_id
ITEM_ID_FIELD: item_id
RATING_FIELD: rating
NEG_PREFIX: neg_
LABEL_FIELD: label
load_col:
    inter: [user_id, item_id, rating]
val_interval:
    rating: "[3,inf)"    
unused_col: 
    inter: [rating]

# training and evaluation
epochs: 500
train_batch_size: 4096
valid_metric: MRR@10

# model
embedding_size: 64

wuxinran-Angel · 2022-04-10T03:06:58Z

@GitHub-ZHY 您好，感谢您的关注！我们在#1192中修复了这个bug，请使用最新源码重新安装RecBole。

您好我在使用最新源码后出现了和上述同样的问题；但使用之前的版本不会出现bug；

config

environment setting

show_progress: False

dataset settings

dataset: 'yoochoose-sample'
field_separator: "\t"
seq_separator: " "
USER_ID_FIELD: 'session_id'
load_col:
unused_col:
inter: [timestamp]

training settings

epochs: 300
train_batch_size: 4096
neg_sampling:

evaluation settings

eval_args:
split: {'RS': [0.8,0.1,0.1]}

split: {'LS': 'valid_and_test'}

order: 'TO'
mode: 'full'
group_by: 'none'
metrics: ['MRR', 'Recall']
topk: [20]
valid_metric: 'MRR@20'
stopping_step: 10

model config

model: 'GRU4Rec'
embedding_size: 64

Ethan-TZ · 2022-04-12T00:55:24Z

@wuxinran-Angel 您好，这个问题是由于hyperopt包的版本与numpy不匹配，而与RecBole的版本无关，如果出现冲突，请尝试降低hyperopt的版本(例如降低到0.2.5)。

wuxinran-Angel · 2022-04-12T01:03:49Z

谢谢您的回复我昨天尝试了这个问题还没有来得及在github上回复。我之前采用了hyperopt 0.2.7版本（用pip安装recbole 1.1可以正常运行）；现在下降到0.2.5版本（用git clone最新的包可以运行）。谢谢您的解答吴忻冉

…

------------------ Original ------------------ From: Tian Zhen ***@***.***> Date: Tue,Apr 12,2022 8:55 AM To: RUCAIBox/RecBole ***@***.***> Cc: Xinran Wu ***@***.***>, Mention ***@***.***> Subject: Re: [RUCAIBox/RecBole] [��BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' (Issue #1240)

GitHub-ZHY added the bug Something isn't working label Apr 7, 2022

Ethan-TZ self-assigned this Apr 7, 2022

Ethan-TZ closed this as completed Apr 12, 2022

Sherry-XLL added the environment label Feb 8, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' #1240

[🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' #1240

GitHub-ZHY commented Apr 7, 2022

Ethan-TZ commented Apr 7, 2022

Ethan-TZ commented Apr 7, 2022 via email

GitHub-ZHY commented Apr 7, 2022

GitHub-ZHY commented Apr 7, 2022

GitHub-ZHY commented Apr 7, 2022

GitHub-ZHY commented Apr 7, 2022

Ethan-TZ commented Apr 8, 2022

wuxinran-Angel commented Apr 10, 2022

Ethan-TZ commented Apr 12, 2022

wuxinran-Angel commented Apr 12, 2022 via email

[🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' #1240

[🐛BUG] GCMC Hyper_tuning 'numpy.random.mtrand.RandomState' object has no attribute 'integers' #1240

Comments

GitHub-ZHY commented Apr 7, 2022

屏幕截图 Connected to pydev debugger (build 211.6693.115)

ERROR in rec_eval EXCEPTION <class 'AttributeError'> 'numpy.random.mtrand.RandomState' object has no attribute 'integers' NODE 0 randint 1 Literal{1} 2 size = 3 len 4 array_union 5 array_union 6 array_union 7 Literal{new_ids} 8 rng = 9 Literal{rng-placeholder}

Ethan-TZ commented Apr 7, 2022

Ethan-TZ commented Apr 7, 2022 via email

GitHub-ZHY commented Apr 7, 2022

GitHub-ZHY commented Apr 7, 2022

training settings

clip_grad_norm: {'max_norm': 5, 'norm_type': 2}

evaluation settings

Other Hyper Parameters:

GitHub-ZHY commented Apr 7, 2022

GitHub-ZHY commented Apr 7, 2022

Ethan-TZ commented Apr 8, 2022

wuxinran-Angel commented Apr 10, 2022

config

environment setting

dataset settings

training settings

evaluation settings

split: {'LS': 'valid_and_test'}

model config

Ethan-TZ commented Apr 12, 2022

wuxinran-Angel commented Apr 12, 2022 via email

屏幕截图
Connected to pydev debugger (build 211.6693.115)

ERROR in rec_eval
EXCEPTION
<class 'AttributeError'>
'numpy.random.mtrand.RandomState' object has no attribute 'integers'
NODE
0 randint
1 Literal{1}
2 size =
3 len
4 array_union
5 array_union
6 array_union
7 Literal{new_ids}
8 rng =
9 Literal{rng-placeholder}