Deprecate the tensorflow frontend

.coveragerc

@@ -17,6 +17,13 @@ exclude_lines =
     if __name__ == .__main__.:
     pass
 
-# Omit files that cannot be tested
 omit =
+    # Omit files that cannot be tested
     dace/jupyter.py
+
+    # Omit deprecated files
+    dace/frontend/tensorflow/__init__.py
+    dace/frontend/tensorflow/tensorflow.py
+    dace/frontend/tensorflow/winograd.py
+    dace/frontend/tensorflow/transformations/__init__.py
+    dace/frontend/tensorflow/transformations/redundant_array.py

README.md

@@ -9,8 +9,10 @@
 
 _Decoupling domain science from performance optimization._
 
-DaCe compiles code in various programming languages and paradigms (Python/Numpy, MATLAB, TensorFlow) and maps it efficiently to **CPUs, GPUs, and FPGAs** with high utilization, on par with the state-of-the-art. The key feature driving DaCe is its Stateful DataFlow multiGraph (SDFG) *data-centric intermediate representation*: A transformable, interactive representation of code based on data movement.
-With data-centric parallel programming, we enable **direct knowledge transfer** of performance optimization, regardless of the scientific application or the target processor.
+DaCe is a parallel programming framework that takes code in Python/NumPy and other programming languages, and maps it to high-performance **CPU, GPU, and FPGA** programs, which can be optimized to achieve state-of-the-art. Internally, DaCe uses the Stateful DataFlow multiGraph (SDFG) *data-centric intermediate representation*: A transformable, interactive representation of code based on data movement.
+Since the input code and the SDFG are separate, it is posible to optimize a program without
+changing its source, so that it stays readable. On the other hand, transformations are customizable and user-extensible, so they can be written once and reused in many applications.
+With data-centric parallel programming, we enable **direct knowledge transfer** of performance optimization, regardless of the application or the target processor.
 
 DaCe generates high-performance programs for:
  * Multi-core CPUs (tested on Intel and IBM POWER9)
@@ -48,7 +50,7 @@ Runtime dependencies:
 Running
 -------
 
-**Python scripts:** Run DaCe programs (in implicit, explicit, or TensorFlow syntax) using Python directly.
+**Python scripts:** Run DaCe programs (in implicit or explicit syntax) using Python directly.
 
 **[SDFV (standalone SDFG viewer)](https://spcl.github.io/dace/sdfv.html):** To view SDFGs separately, run the `sdfv` installed script with the `.sdfg` file as an argument. Alternatively, you can use the link or open `diode/sdfv.html` directly and choose a file in the browser.
 

codecov.yml

@@ -1,5 +1,6 @@
 ignore:
   - "dace/jupyter.py"  # Omit files that cannot be tested
+  - "dace/frontend/tensorflow/**/*"  # Omit deprecated files
 
 coverage:
   range: 40..90

dace/frontend/tensorflow/tensorflow.py

@@ -9,6 +9,7 @@
 import re
 import math
 from typing import Any, List
+import warnings
 
 import dace
 from dace.memlet import Memlet
@@ -130,6 +131,11 @@ def __init__(self, name: str = "tfsession", seed: int = None, config=None):
             :param name: (optional) The name of the resulting SDFG.
             :param seed: (optional) Fix random seed.
         """
+        warnings.warn(
+            'The TensorFlow DaCe frontend has been deprecated and will be '
+            'removed in a future version, please use daceml instead:\n'
+            'https://github.com/spcl/daceml', DeprecationWarning)
+
         self._internal_session = tf.Session(config=config)
 
         # Set for bookkeeping of already visited nodes
@@ -734,8 +740,9 @@ def tensorflow_callback(tf_op, *inputList, num_outputs=0):
                               *callback_input_types))
 
         # Register callback in SDFG
-        node_name, _ = self.graph.add_scalar(
-            node_name, dace_data_scalar.dtype, find_new_name=True)
+        node_name, _ = self.graph.add_scalar(node_name,
+                                             dace_data_scalar.dtype,
+                                             find_new_name=True)
         self.callbackTypeDict[node_name] = dace_data_scalar
         self.callbackFunctionDict[node_name] = tensorflow_callback