在將自定義的網(wǎng)絡(luò)權(quán)重加載到網(wǎng)絡(luò)中時，報錯：
AttributeError: 'dict' object has no attribute 'seek'. You can only torch.load from a file that is seekable. Please pre-load the data into a buffer like io.BytesIO and try to load from it instead.
我們一步一步分析。
模型網(wǎng)絡(luò)權(quán)重保存額代碼是：torch.save(net.state_dict(),'net.pkl')
（1）查看獲取模型權(quán)重的源碼：
pytorch源碼：net.state_dict()
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
|
def state_dict(self, destination=None, prefix='', keep_vars=False): r"""Returns a dictionary containing a whole state of the module. Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Returns: dict: a dictionary containing a whole state of the module Example:: >>> module.state_dict().keys() ['bias', 'weight'] """ |
將網(wǎng)絡(luò)中所有的狀態(tài)保存到一個字典中了，我自己構(gòu)建的就是一個字典，沒問題！
（2）查看保存模型權(quán)重的源碼:
pytorch源碼：torch.save()
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
|
def save(obj, f, pickle_module=pickle, pickle_protocol=DEFAULT_PROTOCOL): """Saves an object to a disk file. See also: :ref:`recommend-saving-models` Args: obj: saved object f: a file-like object (has to implement write and flush) or a string containing a file name pickle_module: module used for pickling metadata and objects pickle_protocol: can be specified to override the default protocol .. warning:: If you are using Python 2, torch.save does NOT support StringIO.StringIO as a valid file-like object. This is because the write method should return the number of bytes written; StringIO.write() does not do this. Please use something like io.BytesIO instead. |
函數(shù)功能是將字典保存為磁盤文件（二進制數(shù)據(jù)），那么我們在torch.load（）時，就是在內(nèi)存中加載二進制數(shù)據(jù)，這就是報錯點。
解決方案：將字典保存為BytesIO文件之后，模型再net.load_state_dict()
|
1
2
3
|
#b為自定義的字典torch.save(b,'new.pkl')net.load_state_dict(torch.load(b)) |
解決方法很簡單，主要記錄解決思路。
以上這篇pytorch加載自定義網(wǎng)絡(luò)權(quán)重的實現(xiàn)就是小編分享給大家的全部內(nèi)容了，希望能給大家一個參考，也希望大家多多支持服務(wù)器之家。
原文鏈接：https://blog.csdn.net/qq_34789262/article/details/83376374
