tf-2.6 to ci, reflect tf-2.6 in readme (#1680)

* tf-2.6 to ci, reflect tf-2.6 in readme

Signed-off-by: Guenther Schmuelling <[email protected]>

* remove tflite/tfjs 2.6 for now

Signed-off-by: Guenther Schmuelling <[email protected]>

* clear session since we run models in process back to back

Signed-off-by: Guenther Schmuelling <[email protected]>

Author: guschmue

README.md

@@ -1,12 +1,11 @@
 <!--- SPDX-License-Identifier: Apache-2.0 -->
 
-# tf2onnx - Convert TensorFlow, Keras and Tflite models to ONNX.
+# tf2onnx - Convert TensorFlow, Keras, Tensorflow.js and Tflite models to ONNX.
 
-tf2onnx converts TensorFlow (tf-1.x or tf-2.x), tf.keras and tflite models to ONNX via command 
+tf2onnx converts TensorFlow (tf-1.x or tf-2.x), keras, tensorflow.js and tflite models to ONNX via command 
 line or python api.
 
-__Note: after tf2onnx-1.8.3 we made a change that impacts the output names for the ONNX model.
-Instead of taking the output names from the tensorflow graph (ie. for keras models this is frequently Identity:0) we decided that it is better to use the structured output names of the model so the output names are now identical to the names in the keras or saved model.__
+__Note: tensorflow.js support was just added. While we tested it with many tfjs models from tfhub, it should be considered experimental.__
 
 TensorFlow has many more ops than ONNX and occasionally mapping a model to ONNX creates issues.
 
@@ -18,8 +17,8 @@ The common issues we run into we try to document here [Troubleshooting Guide](Tr
 
 | Build Type | OS | Python | Tensorflow | ONNX opset | Status |
 | ---        | ---    | ---    | ---        | ---        | ---    |
-| Unit Test - Basic | Linux, MacOS<sup>\*</sup>, Windows<sup>\*</sup> | 3.6-3.9 | 1.12-1.15, 2.1-2.5 | 8-14 | [![Build Status](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_apis/build/status/unit_test?branchName=master)](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_build/latest?definitionId=16&branchName=master) |
-| Unit Test - Full | Linux, MacOS, Windows | 3.6-3.9 | 1.12-1.15, 2.1-2.5 | 8-14 | [![Build Status](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_apis/build/status/unit_test-matrix?branchName=master)](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_build/latest?definitionId=18&branchName=master) | |
+| Unit Test - Basic | Linux, MacOS<sup>\*</sup>, Windows<sup>\*</sup> | 3.6-3.9 | 1.12-1.15, 2.1-2.6 | 8-14 | [![Build Status](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_apis/build/status/unit_test?branchName=master)](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_build/latest?definitionId=16&branchName=master) |
+| Unit Test - Full | Linux, MacOS, Windows | 3.6-3.9 | 1.12-1.15, 2.1-2.6 | 8-14 | [![Build Status](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_apis/build/status/unit_test-matrix?branchName=master)](https://dev.azure.com/tensorflow-onnx/tensorflow-onnx/_build/latest?definitionId=18&branchName=master) | |
 <br/>
 
 ## Supported Versions
@@ -127,7 +126,8 @@ We recently added support for tflite. You convert ```tflite``` models via comman
 python -m tf2onnx.convert
     --saved-model SOURCE_SAVED_MODEL_PATH |
     --checkpoint SOURCE_CHECKPOINT_METAFILE_PATH |
-    --tflite SOURCE_TFLITE_PATH |
+    --tflite TFLITE_MODEL_PATH |
+    --tfjs TFJS_MODEL_PATH | 
     --input | --graphdef SOURCE_GRAPHDEF_PB
     --output TARGET_ONNX_MODEL
     [--inputs GRAPH_INPUTS]
@@ -157,14 +157,18 @@ TensorFlow model as saved_model. We expect the path to the saved_model directory
 
 TensorFlow model as checkpoint. We expect the path to the .meta file.
 
-#### --tflite
-
-Convert a tflite model by providing a path to the .tflite file. Inputs/outputs do not need to be specified.
-
 #### --input or --graphdef
 
 TensorFlow model as graphdef file.
 
+#### --tfjs
+
+Convert a tensorflow.js model by providing a path to the .tfjs file. Inputs/outputs do not need to be specified.
+
+#### --tflite
+
+Convert a tflite model by providing a path to the .tflite file. Inputs/outputs do not need to be specified.
+
 #### --output
 
 The target onnx file path.
@@ -260,7 +264,6 @@ optional arguments:
   --opset OPSET      target opset to use
   --perf csv-file    capture performance numbers for tensorflow and onnx runtime
   --debug            dump generated graph with shape info
-  --fold_const when set, TensorFlow fold_constants transformation will be applied before conversion. This will benefit features including Transpose optimization (e.g. Transpose operations introduced during tf-graph-to-onnx-graph conversion will be removed), and RNN unit conversion (for example LSTM).
 ```
 ```run_pretrained_models.py``` will run the TensorFlow model, captures the TensorFlow output and runs the same test against the specified ONNX backend after converting the model.
 

ci_build/azure_pipelines/onnxruntime_nightly_test.yml

@@ -76,6 +76,18 @@ stages:
           - template: 'unit_test.yml'
         report_coverage: 'True'
 
+    - template: 'templates/job_generator.yml'
+      parameters:
+        platforms: ['linux']
+        python_versions: ['3.9']
+        tf_versions: ['2.6.0']
+        onnx_opsets: ['']
+        onnx_backends: {onnxruntime: ['nightly']}
+        job:
+          steps:
+          - template: 'unit_test.yml'
+        report_coverage: 'True'
+
     - template: 'templates/job_generator.yml'
       parameters:
         platforms: ['windows']

ci_build/azure_pipelines/pretrained_model_test-matrix.yml

@@ -37,6 +37,15 @@ jobs:
       steps:
       - template: 'pretrained_model_test.yml'
 
+- template: 'templates/job_generator.yml'
+  parameters:
+    platforms: ['linux', 'windows']
+    python_versions: ['3.9']
+    tf_versions: ['2.6.0']
+    job:
+      steps:
+      - template: 'pretrained_model_test.yml'
+
 schedules:
 - cron: "0 10 * * *"
   displayName: pre-trained model test, full matrix

ci_build/azure_pipelines/pretrained_model_test.yml

@@ -15,7 +15,7 @@ jobs:
 - template: 'templates/job_generator.yml'
   parameters:
     python_versions: ['3.7']
-    tf_versions: ['1.15.5','2.4.1']
+    tf_versions: ['1.15.5','2.6.0']
     job:
       steps:
       - template: 'pretrained_model_test.yml'

ci_build/azure_pipelines/unit_test-matrix.yml

@@ -47,4 +47,15 @@ stages:
           - template: 'unit_test.yml'
         report_coverage: 'True'
 
+    - template: 'templates/job_generator.yml'
+      parameters:
+        platforms: ['linux', 'windows']
+        python_versions: ['3.8']
+        tf_versions: ['2.6.0']
+        onnx_opsets: ['']
+        job:
+          steps:
+          - template: 'unit_test.yml'
+        report_coverage: 'True'
+
   - template: 'templates/combine_test_coverage.yml'

ci_build/azure_pipelines/unit_test.yml

@@ -56,7 +56,7 @@ stages:
         # tf 1.12
         python_versions: [3.6']
         tf_versions: ['1.12.3']
-        onnx_opsets: ['']
+        onnx_opsets: ['9']
         job:
           steps:
           - template: 'unit_test.yml'
@@ -66,7 +66,7 @@ stages:
       parameters:
         platforms: ['windows']
         tf_versions: ['1.14.0']
-        onnx_opsets: ['']
+        onnx_opsets: ['14']
         job:
           steps:
           - template: 'unit_test.yml'
@@ -77,7 +77,7 @@ stages:
         python_versions: ['3.7']
         platforms: ['windows']
         tf_versions: ['2.4.1']
-        onnx_opsets: ['']
+        onnx_opsets: ['13']
         job:
           steps:
           - template: 'unit_test.yml'