dgl._ffi.base.DGLError: [17:19:38] /opt/dgl/src/runtime/c_runtime_api.cc:88: Check failed: allow_missing: Device API gpu is not enabled. Please install the cuda version of dgl.

[BEILOP]

dgl==0.9.0,torch==1.10.0,cuda=10.2 pytorch 10.2 INFO:root:Namespace(aggregator='mean', backprop_every=128, bs=1, cache_dim=32, cache_updater='gru', data='THA', dim=32, drop_out=0.2, embedding_module='graph_attention', gpu=0, hist_wind=7, leadtime=1, lr=0.001, max_pool=True, message_dim=32, message_function='identity', n_degree=10, n_epoch=20, n_head=2, n_layer=1, n_runs=1, patience=5, predwind=1, prefix='', uniform=False) pINFO:root:num of training instances: 316 INFO:root:start 0 epoch 0%| | 0/3 [00:00<?, ?it/s] Traceback (most recent call last): File "train.py", line 214, in y_hat = model.predict(sources_batch, destinations_batch, edge_idxs_batch, timestamps_batch, batch_idx, story_ids_batch) File "/home/jiaojiaxin/MTG/model/mtg.py", line 249, in predict pred = self.gnn_model(t_list, all_nodes_cache, all_rels_cache) File "/home/anaconda3/envs/mtg/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl return forwardcall(*input, **kwargs) File "/home/jiaojiaxin/MTG/gnn/models.py", line 60, in forward pred, idx, = self.__get_pred_embeds(t_list, ent_memory, rel_memory) File "/home/jiaojiaxin/MTG/gnn/models.py", line 68, in get_pred_embeds embed_seq_tensor, len_non_zero = self.aggregator(sorted_t, ent_memory, rel_memory, self.ent_embeds, File "/home/anaconda3/envs/mtg/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl return forward_call(*input, **kwargs) File "/home/jiaojiaxin/MTG/gnn/aggregators.py", line 62, in forward batched_g = batched_g.to(device) # torch.device('cuda:0') File "/home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/heterograph.py", line 5448, in to ret._graph = self._graph.copy_to(utils.to_dgl_context(device)) File "/home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/heterograph_index.py", line 236, in copy_to return _CAPI_DGLHeteroCopyTo(self, ctx.device_type, ctx.device_id) File "dgl/_ffi/_cython/./function.pxi", line 293, in dgl._ffi._cy3.core.FunctionBase.call__ File "dgl/_ffi/_cython/./function.pxi", line 225, in dgl._ffi._cy3.core.FuncCall File "dgl/_ffi/_cython/./function.pxi", line 215, in dgl._ffi._cy3.core.FuncCall3 dgl._ffi.base.DGLError: [17:19:38] /opt/dgl/src/runtime/c_runtime_api.cc:88: Check failed: allow_missing: Device API gpu is not enabled. Please install the cuda version of dgl. Stack trace: [bt] (0) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4f) [0x7fd1e344d29f] [bt] (1) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::runtime::DeviceAPIManager::GetAPI(std::string, bool)+0x37c) [0x7fd1e371f72c] [bt] (2) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::runtime::DeviceAPI::Get(DLContext, bool)+0x1f4) [0x7fd1e3719b14] [bt] (3) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::runtime::NDArray::CopyFromTo(DLTensor, DLTensor, void)+0x242) [0x7fd1e373ac72] [bt] (4) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::runtime::NDArray::CopyTo(DLContext const&, void const&) const+0xfb) [0x7fd1e377722b] [bt] (5) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::UnitGraph::CopyTo(std::shared_ptr, DLContext const&, void const&)+0x317) [0x7fd1e3887fa7] [bt] (6) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(dgl::HeteroGraph::CopyTo(std::shared_ptr, DLContext const&, void const&)+0x109) [0x7fd1e3789069] [bt] (7) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(+0x4eb739) [0x7fd1e3796739] [bt] (8) /home/anaconda3/envs/mtg/lib/python3.8/site-packages/dgl/libdgl.so(DGLFuncCall+0x48) [0x7fd1e37190e8]

[czkkkkkk]

Hi @BEILOP, could you try the latest DGL version?

[BEILOP]

because i must read some files encoded in the old DGL version,These files cannot be read with the latest  

孔明羽 @.***

 

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2023年8月2日(星期三) 中午11:44 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [dmlc/dgl] dgl._ffi.base.DGLError: [17:19:38] /opt/dgl/src/runtime/c_runtime_api.cc:88: Check failed: allow_missing: Device API gpu is not enabled. Please install the cuda version of dgl. (Issue #6078)

Hi @BEILOP, could you try the latest DGL version?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>

[czkkkkkk]

Did you install the CUDA version of DGL?

[github-actions[bot]]

This issue has been automatically marked as stale due to lack of activity. It will be closed if no further activity occurs. Thank you

[rudongyu]

Hi, I am closing this issue assuming you are happy about our response. Feel free to follow up and reopen the issue if you have more questions with regard to our response.