akirazc / catn

The implementation of SIGIR 2020 paper "CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network“, Cheng Zhao, Chenliang Li, Rong Xiao, Hongbo Deng and Aixin Sun

CATN

Codes for SIGIR 2020 paper CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network.

Citation

Please cite our paper if you find this code useful for your research:

@inproceedings{sigir20:catn,
  author    = {Cheng Zhao and
               Chenliang Li and
               Rong Xiao and
               Hongbo Deng and
               Aixin Sun},
  title     = {{CATN:} Cross-Domain Recommendation for Cold-Start Users via Aspect
               Transfer Network},
  booktitle = {{SIGIR}},
  year      = {2020},
}

Requirement

• python 3.6
• tensorflow 1.10.0
• numpy
• pandas
• scipy
• gensim
• sklearn
• tqdm

Files in the folder

• dataset/
  • preprocessing.py: constructing cross-domain datasets;
• runner/
  • CATN_runner.py: the main runner (including the configurations);
• utils
  • CATN.py: CATN implementation.

Running the code

1. Download the original data from Amazon-5core, choose two relevant categories (e.g., Books, Movies and TV) and put them under the same directory in dataset/.

2. run python preprocessing.py.

3. run python CATN_runner.py.

catn's Issues

Code cannot run

hi，I've been studing your code recently，but there is a problem when I try to run this code.When I ran dataset/preprocessing.py, code stop at split_train_test（）; (this is the output of dataset/preprocessing.py ：
Load raw data from book2movie/reviews_Books_5.csv. Load raw data from book2movie/reviews_Movies_and_TV_5.csv. Source domain users 126666, items 63202, ratings 3494976. Target domain users 27822, items 12287, ratings 779376. Overlapping users 6074 (0.000%, 0.000%). Ratio 1.00, common users 3037, vali cold-start users 1214, test cold-start users 1823.)

so when I try to run runner/CATN_runner.py,there is a error just like this
Traceback (most recent call last): File "CATN_runner.py", line 106, in <module> with open(pkl_path % args.ratio, 'rb') as f: IOError: [Errno 2] No such file or directory: './dataset/book2movie/crossdata_i30_u10_0.20.pkl'

I wonder if u can help to figure this out ,thank u!!

preprocessing.py

As far as i'm concerned,it has a small bug.
In line 66,'index_t = df_t[["overall", "asin"]].groupby('asin').count() >= 30',I think you mean to write 'index_t = df_t[["overall", "asin"]].groupby('asin').count() >= thre_i'

Code cannot run

hi，I've been studing your code recently，but there is a problem when I try to run this code.when I try to run runner/CATN_runner.py,there is a error just like this

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "F:/Model-Code/CATN/runner/CATN_runner.py", line 162, in <module>
    catn_runner.step_train(sess)
  File "F:/Model-Code/CATN/runner/CATN_runner.py", line 90, in step_train
    self.catn.train_step(sess)
  File "F:\Model-Code\CATN\utils\CATN.py", line 372, in train_step
    self.ratings_ph: ratings_batch_s,
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\client\session.py", line 877, in run
    run_metadata_ptr)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\client\session.py", line 1100, in _run
    feed_dict_tensor, options, run_metadata)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\client\session.py", line 1272, in _do_run
    run_metadata)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\client\session.py", line 1291, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[57,12] = -1 is not in [0, 20001)
	 [[Node: embedding_lookup_2 = GatherV2[Taxis=DT_INT32, Tindices=DT_INT32, Tparams=DT_FLOAT, _class=["loc:@Adam/update_pre_word_embeddings/AssignSub"], _device="/job:localhost/replica:0/task:0/device:CPU:0"](pre_word_embeddings/read, embedding_lookup, embedding_lookup_5/axis)]]

Caused by op 'embedding_lookup_2', defined at:
  File "F:/Model-Code/CATN/runner/CATN_runner.py", line 162, in <module>
    catn_runner.step_train(sess)
  File "F:/Model-Code/CATN/runner/CATN_runner.py", line 89, in step_train
    self.catn = CATN(data, self.item_ave_rating_s, self.item_ave_rating_t, self.reviews, args, pkl_path)
  File "F:\Model-Code\CATN\utils\CATN.py", line 48, in __init__
    self.inference()
  File "F:\Model-Code\CATN\utils\CATN.py", line 103, in inference
    self.user_reviews_repr = tf.nn.embedding_lookup(self.word_embeddings, self.user_inputs)*self.user_inputs_mask
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\ops\embedding_ops.py", line 310, in embedding_lookup
    transform_fn=None)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\ops\embedding_ops.py", line 133, in _embedding_lookup_and_transform
    result = _clip(array_ops.gather(params[0], ids, name=name),
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\ops\array_ops.py", line 2659, in gather
    return gen_array_ops.gather_v2(params, indices, axis, name=name)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\ops\gen_array_ops.py", line 3761, in gather_v2
    "GatherV2", params=params, indices=indices, axis=axis, name=name)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 787, in _apply_op_helper
    op_def=op_def)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\util\deprecation.py", line 454, in new_func
    return func(*args, **kwargs)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\framework\ops.py", line 3155, in create_op
    op_def=op_def)
  File "E:\ruanjian\Miniconda3\envs\tensorflow110\lib\site-packages\tensorflow\python\framework\ops.py", line 1717, in __init__
    self._traceback = tf_stack.extract_stack()

InvalidArgumentError (see above for traceback): indices[57,12] = -1 is not in [0, 20001)
	 [[Node: embedding_lookup_2 = GatherV2[Taxis=DT_INT32, Tindices=DT_INT32, Tparams=DT_FLOAT, _class=["loc:@Adam/update_pre_word_embeddings/AssignSub"], _device="/job:localhost/replica:0/task:0/device:CPU:0"](pre_word_embeddings/read, embedding_lookup, embedding_lookup_5/axis)]]