Automatic update of fbcode/onnx to 7988d8360b11e6003560076e9b1d4aa426…

…db3244 (pytorch#25959)

Summary:
Pull Request resolved: pytorch#25959

Previous import was 28ca699b69b5a31892619defca2391044a9a6052

Included changes:
- **[7988d836](onnx/onnx@7988d836)**: Supporting negative axes for all existing onnx ops (pytorch#2281) <Negin Raoof>
- **[5ca0a09e](onnx/onnx@5ca0a09e)**: Update managingexperimentalops.md (pytorch#1981) <Joseph Spisak>
- **[bc0495c1](onnx/onnx@bc0495c1)**: Fix link to community docs in readme (pytorch#2261) <Prasanth Pulavarthi>
- **[2fdb3ef6](onnx/onnx@2fdb3ef6)**: move map and sequence types to onnx domain, (pytorch#2244) <Ke Zhang>
- **[568b65aa](onnx/onnx@568b65aa)**: Improve compatiblity with proto3 and enable reading attributes (pytorch#2288) <Dmitri Smirnov>
- **[1f350f2c](onnx/onnx@1f350f2c)**: Remove type info for loop variadic input in Loop op used to compose the Range op (pytorch#2287) <Hariharan Seshadri>
- **[eb139446](onnx/onnx@eb139446)**: Add Foundation WG to working-groups.md (pytorch#2276) <Ryan Loney>
- **[4eabc4b3](onnx/onnx@4eabc4b3)**: Fix testdata model for CumSum. Add exclusive attribute. (pytorch#2271) <jignparm>
- **[1a62afdb](onnx/onnx@1a62afdb)**: Support GatherND operator in ONNX (pytorch#2106) <Hariharan Seshadri>
- **[0e330e9d](onnx/onnx@0e330e9d)**: Support ScatterND operator in ONNX (pytorch#2220) <Bowen Bao>
- **[733f7a6a](onnx/onnx@733f7a6a)**: Add Det to ONNX (pytorch#2233) <Bowen Bao>
- **[52187738](onnx/onnx@52187738)**: Update the description of nearest_mode of resize op (pytorch#2257) <daquexian>
- **[64b4b686](onnx/onnx@64b4b686)**: Adding sparse tensor to ONNX (pytorch#2019) <G. Ramalingam>
- **[c8a8b7cc](onnx/onnx@c8a8b7cc)**: Support Range operator in ONNX (pytorch#2242) <Hariharan Seshadri>
- **[44b0d6d5](onnx/onnx@44b0d6d5)**: Update resize op (pytorch#2057) <daquexian>
- **[7d907964](onnx/onnx@7d907964)**: Add function to fuse dynamic quantization graph into 1 node (pytorch#2187) <Ashwini Khade>
- **[36f8e6d9](onnx/onnx@36f8e6d9)**: Update logo_request.md (pytorch#2231) <Prasanth Pulavarthi>
- **[4eb737c8](onnx/onnx@4eb737c8)**: Update Clip in opset 11 to support min/max as inputs instead of attributes (pytorch#2096) <Bowen Bao>
- **[a25e1388](onnx/onnx@a25e1388)**: Fix segfault in tile shape inference (pytorch#2221) <daquexian>
- **[2dc273c7](onnx/onnx@2dc273c7)**: update onehot shape inference to reflect the spec for depth input (pytorch#2224) <Ashwini Khade>
- **[665211c1](onnx/onnx@665211c1)**: Add GatherElements Op and Rename ScatterElements (pytorch#2143) <Lara Haidar>
- **[3ba2e31a](onnx/onnx@3ba2e31a)**: Unique (pytorch#2141) <liqunfu>
- **[5a5588ad](onnx/onnx@5a5588ad)**: Clarify dimension variable scoping (pytorch#2211) <G. Ramalingam>
- **[fabe39d5](onnx/onnx@fabe39d5)**: Liqun/topk sort (pytorch#2126) <liqunfu>
- **[453aa644](onnx/onnx@453aa644)**: Update document for NMS (pytorch#2193) <Hector Li>
- **[34e28ec2](onnx/onnx@34e28ec2)**: Handle negative 'axis' value in Split type and shape inferencing (pytorch#2177) <Scott McKay>
- **[28ec4583](onnx/onnx@28ec4583)**: depth to space shuffle order (pytorch#2163) <Negin Raoof>
- **[98f72629](onnx/onnx@98f72629)**: minor updates to fix links in readme (pytorch#2189) <Prasanth Pulavarthi>
- **[321d1467](onnx/onnx@321d1467)**: Add check to disallow squeezing input axes which are not 1 (pytorch#2204) <Ashwini Khade>
- **[573f0dc9](onnx/onnx@573f0dc9)**: fix a bug in fun shape inference (pytorch#2188) <Tang, Cheng>
- **[36dc7110](onnx/onnx@36dc7110)**: Clarify ambiguity in gather spec regarding indices expectation (pytorch#2202) <Ashwini Khade>
- **[a2449673](onnx/onnx@a2449673)**: Fix some minor issues in IR.md and Versioning.md (pytorch#2108) <edgchen1>
- **[349aff69](onnx/onnx@349aff69)**: Skip install typing package for python >=3.5 (pytorch#2199) <bddppq>

Test Plan: ci

Reviewed By: bddppq

Differential Revision: D17296390

fbshipit-source-id: 0b60ab87b1d23199c1653e53929a68703eb717fd

caffe2/python/onnx/tests/onnx_backend_test.py

@@ -30,6 +30,8 @@
                      '|test_reduce_log_sum.*'  # Does not support ReduceLogSum.
                      '|test_reduce_prod.*'  # Does not support ReduceProd.
                      '|test_reduce_sum_square.*'  # Does not support ReduceSumSquare
+                     '|test_det.*'  # Does not support Det
+                     '|test_range.*'  # Does not support Range
                      '|test_tile.*'  # Tile's Caffe2 implementation needs some tweak
                      '|test_lstm.*'  # Seems LSTM case has some problem
                      '|test_simple_rnn.*'  # Seems simple RNN case has some problem
@@ -76,6 +78,9 @@
                      '|test_gather_elements.*'  # opset 11 is not supported yet
                      '|test_scatter.*'  # opset 11 is not supported yet
                      '|test_unique.*'  # opset 11 is not supported yet
+                     '|test_gathernd.*'  # opset 11 is not supported yet
+                     '|test_.*negative_ax.*'  # negative axis is not supported yet
+                     '|test_.*negative_ind.*'  # negative axis is not supported yet
                      ')')
 
 # Quick patch to unbreak master CI, is working on the debugging.