[blocked by #3748] update logging docs

.github/ISSUE_TEMPLATE/bug_report.md

@@ -19,12 +19,8 @@ assignees: ''
 
 ### To Reproduce
 
-Steps to reproduce the behavior:
-
-1. Go to '...'
-2. Run '....'
-3. Scroll down to '....'
-4. See error
+Before reporting a bug, make sure that the bug can be reproduced with a minimal example. You can simply subclass our [miminal code example] (https://github.com/PyTorchLightning/pytorch-lightning/blob/master/pl_examples/bug_report_model.py) and add your relevant changes, to see if the issue persists. 
+If the test is failing, please add your test cases to the issue (as a draft PR, or simple paste the code to the isse description here).
 
 <!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
 

.github/workflows/ci_test-conda.yml

@@ -31,6 +31,7 @@ jobs:
         pip list
 
     - name: Cache datasets
+      # todo this probably does not work with docker images, rather cache dockers
       uses: actions/cache@v2
       with:
         path: Datasets # This path is specific to Ubuntu

.github/workflows/ci_test-full.yml

@@ -103,7 +103,8 @@ jobs:
         HOROVOD_BUILT=$(python -c "import horovod.torch; horovod.torch.nccl_built(); print('SUCCESS')" || true)
         if [[ $HOROVOD_BUILT != "SUCCESS" ]]; then
           pip uninstall -y horovod
-          pip install --no-cache-dir $(grep "horovod" requirements/extra.txt)
+          echo $(grep "horovod" requirements/extra.txt) > requirements/horovod.txt
+          pip install --no-cache-dir -r requirements/horovod.txt
         fi
         horovodrun --check-build
       shell: bash

.pyrightconfig.json

@@ -35,6 +35,7 @@
     "pytorch_lightning/trainer/connectors/checkpoint_connector.py",
     "pytorch_lightning/trainer/connectors/data_connector.py",
     "pytorch_lightning/trainer/connectors/logger_connector.py",
+    "pytorch_lightning/trainer/connectors/slurm_connector.py",
     "pytorch_lightning/distributed/dist.py",
     "pytorch_lightning/tuner",
     "pytorch_lightning/plugins"

CHANGELOG.md

@@ -31,6 +31,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Added `broadcast` to `TPUBackend` ([#3814](https://github.com/PyTorchLightning/pytorch-lightning/pull/3814))
 
+- Added `XLADeviceUtils` class to check XLA device type ([#3274](https://github.com/PyTorchLightning/pytorch-lightning/pull/3274))
+
 ### Changed
 
 - Changed `LearningRateLogger` to `LearningRateMonitor` ([#3251](https://github.com/PyTorchLightning/pytorch-lightning/pull/3251))
@@ -57,6 +59,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Deprecated
 
+- Rename Trainer arguments `row_log_interval` >> `log_every_n_steps` and `log_save_interval` >> `flush_logs_every_n_steps` ([#3748](https://github.com/PyTorchLightning/pytorch-lightning/pull/3748))
 
 ### Removed
 
@@ -101,10 +104,14 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Fixed counter-intuitive error being thrown in `Accuracy` metric for zero target tensor ([#3764](https://github.com/PyTorchLightning/pytorch-lightning/pull/3764))
 
+- Fixed Tuner dump: add `current_epoch` to dumped_params ([#3261](https://github.com/PyTorchLightning/pytorch-lightning/pull/3261))
+
 - Fixed aggregation of metrics ([#3517](https://github.com/PyTorchLightning/pytorch-lightning/pull/3517))
 
 - Fixed `current_epoch` and `global_step` properties mismatch between `Trainer` and `LightningModule` ([#3785](https://github.com/PyTorchLightning/pytorch-lightning/pull/3785))
 
+- Fixed learning rate scheduler for optimizers with internal state ([#3897](https://github.com/PyTorchLightning/pytorch-lightning/pull/3897))
+
 ## [0.9.0] - YYYY-MM-DD
 
 ### Added

dockers/base-cuda/Dockerfile

@@ -22,8 +22,11 @@
 ARG CUDNN_VERSION=7
 ARG CUDA_VERSION=10.1
 
-FROM nvidia/cuda:${CUDA_VERSION}-cudnn${CUDNN_VERSION}-devel
-# FROM nvidia/cuda:${CUDA_VERSION}-devel
+# FROM nvidia/cuda:${CUDA_VERSION}-devel-ubuntu20.04
+# FROM nvidia/cuda:${CUDA_VERSION}-cudnn${CUDNN_VERSION}-devel-ubuntu18.04
+FROM nvidia/cuda:${CUDA_VERSION}-cudnn${CUDNN_VERSION}-devel-ubuntu16.04
+# FROM nvidia/cuda:${CUDA_VERSION}-devel-ubuntu18.04
+# FROM nvidia/cuda:${CUDA_VERSION}-devel-ubuntu16.04
 
 ARG PYTHON_VERSION=3.7
 ARG PYTORCH_VERSION=1.6